Patent Application
20220401794
The present disclosure relates to exercise machines. More specifically, the present disclosure relates to using artificial intelligence to dynamically create an exercise program based on a user energy score.
Exercise and rehabilitation devices, such as a cycling machine and balance equipment, are used to facilitate exercise, strength training, osteogenesis, and/or rehabilitation of a user. A user may perform an exercise (e.g., cycling, balancing, bench press, pull down, arm curl, etc.) using the osteogenic isometric exercise, rehabilitation, and/or strength training equipment to improve osteogenesis, bone growth, bone density, muscular hypertrophy, flexibility, balance, coordination, reduce pain, decrease rehabilitation time, increase strength, or some combination thereof. The isometric exercise, rehabilitation, and/or strength training equipment may include moveable portions onto which the user adds a load or balances. For example, to perform a cycling exercise, the user may sit in a seat, place each of the user's feet on a respective pedal of an cycling machine, and push on the pedals with the user's feet while each of the pedals rotate in a circular motion. To perform a balancing exercise, the user may stand on a balance board and balance on top of the balance board as it shifts in one or more directions. The isometric exercise, rehabilitation, and/or strength training equipment may include non-movable portions onto which the user adds load. For example, to perform a leg-press-style exercise, the user may sit in a seat, place each of the user's feet on a respective foot plate, and push on the feet plates with the user's feet while the foot plates remain in the same position.
Representative implementations set forth herein disclose various techniques for an adjustment of exercise based on artificial intelligence, exercise plan, and user feedback. As used herein, the terms “exercise apparatus,” “exercise device,” “electromechanical device,” “exercise machine,” “rehabilitation device,” “cycling machine” “balance board,” and “isometric exercise and rehabilitation assembly” may be used interchangeably. The terms “exercise apparatus,” “exercise device,” “electromechanical device,” “exercise machine,” “rehabilitation device,” “cycling machine” “balance board,” and “isometric exercise and rehabilitation assembly” may also refer to an osteogenic, strength training, isometric exercise, and/or rehabilitation assembly.
The present disclosure provides a method for generating, by an artificial intelligence engine, an exercise program comprising a first user energy score. The method includes generating, by the artificial intelligence engine, an exercise program including an exercise plan, wherein the exercise plan includes a set of exercises. Each respective one of the set of exercises is associated with one or more user energy consumption metrics based at least on a metabolic equivalent of task (MET) value. Based on the one or more user energy consumption metrics, the first user energy score may be associated with the exercise program. The method also includes receiving data pertaining to a set of users, wherein the data includes physical activity goals the set of users desires to achieve. The method also includes determining second user energy scores for the physical activity goals and, based on the first and second user energy scores, assigning, by the artificial intelligence engine, at least a subset of the plurality of users to the exercise program.
The present disclosure also provides a system including a memory devices storing instructions and a processing device communicatively coupled to the memory device. The processing device executes the instructions to perform any of the methods and/or operations described herein.
The present disclosure further provides a tangible, non-transitory computer-readable medium storing instructions that, when executed, cause a processing device to perform any of the methods and/or operations described herein
Other technical features may be readily apparent to one skilled in the art from the following figures, descriptions, and claims.
For a more complete understanding of the present disclosure and its advantages, reference is now made to the following description, taken in conjunction with the accompanying drawings. It is emphasized that, according to common practice, the various features of the drawings are not to-scale. On the contrary, the dimensions of the various features are arbitrarily expanded or reduced for clarity.
Various terms are used to refer to particular system components. Different entities may refer to a component by different names—this document does not intend to distinguish between components that differ in name but not function. In the following discussion and in the claims, the terms “including” and “comprising” are used in an open-ended fashion, and thus should be interpreted to mean “including, but not limited to . . . .” Also, the term “couple” or “couples” is intended to mean either an indirect or direct connection. Thus, if a first device couples to a second device, that connection may be through a direct connection or through an indirect connection via other devices and connections.
Various terms are used to refer to particular system components. Different entities may refer to a component by different names—this document does not intend to distinguish between components that differ in name but not function. In the following discussion and in the claims, the terms “including” and “comprising” are used in an open-ended fashion, and thus should be interpreted to mean “including, but not limited to . . . .” Also, the term “couple” or “couples” is intended to mean either an indirect or direct connection. Thus, if a first device couples to a second device, that connection may be through a direct connection or through an indirect connection via other devices and connections.
The terminology used herein is for the purpose of describing particular example embodiments only, and is not intended to be limiting. As used herein, the singular forms “a,” “an,” and “the” may be intended to include the plural forms as well, unless the context clearly indicates otherwise. The method steps, processes, and operations described herein are not to be construed as necessarily requiring their performance in the particular order discussed or illustrated, unless specifically identified as an order of performance. It is also to be understood that additional or alternative steps may be employed.
The terms first, second, third, etc. may be used herein to describe various elements, components, regions, layers and/or sections; however, these elements, components, regions, layers and/or sections should not be limited by these terms. These terms may be only used to distinguish one element, component, region, layer or section from another region, layer or section. Terms such as “first,” “second,” and other numerical terms, when used herein, do not imply a sequence or order unless clearly indicated by the context. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the teachings of the example embodiments. The phrase “at least one of,” when used with a list of items, means that different combinations of one or more of the listed items may be used, and only one item in the list may be needed. For example, “at least one of: A, B, and C” includes any of the following combinations: A, B, C, A and B, A and C, B and C, and A and B and C. In another example, the phrase “one or more” when used with a list of items means there may be one item or any suitable number of items exceeding one.
Spatially relative terms, such as “inner,” “outer,” “beneath,” “below,” “lower,” “above,” “upper,” “top,” “bottom,” and the like, may be used herein. These spatially relative terms can be used for ease of description to describe one element's or feature's relationship to another element(s) or feature(s) as illustrated in the figures. The spatially relative terms may also be intended to encompass different orientations of the device in use, or operation, in addition to the orientation depicted in the figures. For example, if the device in the figures is turned over, elements described as “below” or “beneath” other elements or features would then be oriented “above” the other elements or features. Thus, the example term “below” can encompass both an orientation of above and below. The device may be otherwise oriented (rotated 90 degrees or at other orientations) and the spatially relative descriptions used herein interpreted accordingly.
Moreover, various functions described below can be implemented or supported by one or more computer programs, each of which is formed from computer readable program code and embodied in a computer readable medium. The terms “application” and “program” refer to one or more computer programs, software components, sets of instructions, procedures, functions, objects, classes, instances, related data, or a portion thereof adapted for implementation in a suitable computer readable program code. The phrase “computer readable program code” includes any type of computer code, including source code, object code, and executable code. The phrase “computer readable medium” includes any type of medium capable of being accessed by a computer, such as read only memory (ROM), random access memory (RAM), a hard disk drive, a compact disc (CD), a digital video disc (DVD), solid state drives (SSDs), flash memory, or any other type of memory. A “non-transitory” computer readable medium excludes wired, wireless, optical, or other communication links that transport transitory electrical or other signals. A non-transitory computer readable medium includes media where data can be permanently stored and media where data can be stored and later overwritten, such as a rewritable optical disc or an erasable memory device.
The term “bone geometry” may refer to bone diameter, bone density, bone shape, bone cross-section, bone length, bone weight, or any suitable bone dimension(s) and/or measurement(s).
The term “empirical data” may refer to data obtained and/or derived based on observation, experience, measurement, and/or research.
The term “strain,” when used in context with a bone of a user, may refer to an amount, proportion, or degree of deformation of the bone material.
The terms “exercise machine” and “isometric exercise and rehabilitation assembly” may be used interchangeably herein.
The terms “body part” and “body portion” may be used interchangeably herein.
The phrase “achieve a desired outcome” may refer to a completion of the desired outcome and/or also making progress toward the desired outcome. The term “attribute of a user” may include a quality or feature regarded as a characteristic or inherent part of someone.
The term “metabolic equivalent of task (MET) score” (also referred to as a MET indicator or a MET value) may refer to a ratio of working metabolic rate of a person relative to the person's resting metabolic rate.
The term “user energy consumption metric” may refer to a metric that is generated based on the MET score and/or a user fitness test, among other things.
The term “user energy score” may refer to an amount of energy it takes to attempt to achieve (including achieving) a physical activity goal of a user, and/or an amount of energy it takes to attempt to achieve (including achieving) completing an exercise program or any portion thereof.
Definitions for other certain words and phrases are provided throughout this patent document. Those of ordinary skill in the art should understand that in many if not most instances, such definitions apply to prior as well as future uses of such defined words and phrases.
As typically healthy people grow from infants to children to adults, they experience bone growth. Such, growth, however, typically stops at approximately age 30. After that point, without interventions as described herein, bone loss (called osteoporosis), can start to occur. This does not mean that the body stops creating new bone. Rather, it means that the rate at which it creates new bone tends to slow, while the rate at which bone loss occurs tends to increase.
In addition, as people age and/or become less active than they once were, they may experience muscle loss. For example, muscles that are not used often may reduce in muscle mass. As a result, the muscles become weaker. In some instances, people may be affected by a disease, such as muscular dystrophy, that causes the muscles to become progressively weaker and to have reduced muscle mass. To increase the muscle mass and/or reduce the rate of muscle loss, people may exercise a muscle to cause muscular hypertrophy, thereby strengthening the muscle as the muscle grows. Muscular hypertrophy may refer to an increase in a size of skeletal muscle through a growth in size of its component cells. There are two factors that contribute to muscular hypertrophy, (i) sarcoplasmic hypertrophy (increase in muscle glycogen storage), and (ii) myofibrillar hypertrophy (increase in myofibril size). The growth in the cells may be caused by an adaptive response that serves to increase an ability to generate force or resist fatigue.
The rate at which such bone or muscle loss occurs generally accelerates as people age. A net growth in bone can ultimately become a net loss in bone, longitudinally across time. By the time, in general, women are over 50 and men are over 70, net bone loss can reach a point where brittleness of the bones is so great that the risk of life-altering fractures can occur. Examples of such fractures include fractures of the hip and femur. Of course, fractures can also occur due to participation in athletics or due to accidents. In such cases, it is just as relevant to have a need for bone growth which heals or speeds the healing of the fracture.
To understand why such fractures occur, it is useful to recognize that bone is itself porous, with a somewhat-honeycomb like structure. This structure may be dense and therefore stronger or it may be variegated, spread out and/or sparse, such latter structure being incapable of continuously or continually supporting the weight (load) stresses experienced in everyday living. When such loads exceed the support capability of the structure at a stressor point or points, a fracture occurs. This is true whether the individual had a fragile bone structure or a strong one: it is a matter of physics, of the literal “breaking point.”
It is therefore preferable to have a means of mitigating or ameliorating bone loss and of healing fractures. Further, it is preferable to encourage new bone growth, thus increasing the density of the structure described hereinabove. The increased bone density may increase the load-bearing capacities of the bone, thus making first or subsequent fractures less likely to occur. Reduced fractures may improve a quality of life of the individual. The process of bone growth itself is referred to as osteogenesis, literally the creation of bone.
It is also preferable to have a means for mitigating or ameliorating muscle mass loss and weakening of the muscles. Further, it is preferable to encourage muscle growth by increasing the muscle mass through exercise. The increased muscle mass may enable a person to exert more force with the muscle and/or to resist fatigue in the muscle for a longer period of time.
In order to create new bone, at least three factors are necessary. First, the individual must have a sufficient intake of calcium, but second, in order to absorb that calcium, the individual must have a sufficient intake and absorption of Vitamin D, a matter problematic for those who have cystic fibrosis, who have undergone gastric bypass surgery or have other absorption disorders or conditions which limit absorption. Separately, supplemental estrogen for women and supplemental testosterone for men can further ameliorate bone loss. On the other hand, abuse of alcohol and smoking can harm one's bone structure. Medical conditions such as, without limitation, rheumatoid arthritis, renal disease, overactive parathyroid glands, diabetes or organ transplants can also exacerbate osteoporosis. Ethical pharmaceuticals such as, without limitation, hormone blockers, seizure medications and glucocorticoids are also capable of inducing such exacerbations. But even in the absence of medical conditions as described hereinabove, Vitamin D and calcium taken together do not create osteogenesis to a desirable degree or ameliorate bone loss to a desirable degree.
To achieve osteogenesis, therefore, one must add in the third factor: exercise. Specifically, one must subject one's bones to a force at least equal to certain multiple of body weight, such multiples varying depending on the individual and the specific bone in question. As used herein, “MOB” means Multiples of Body Weight. It has been determined through research that subjecting a given bone to a certain threshold MOB (this may also be known as a “weight-bearing exercise”), even for an extremely short period of time, one simply sufficient to exceed the threshold MOB, encourages and fosters osteogenesis in that bone.
Further, a person can achieve muscular hypertrophy by exercising the muscles for which increased muscle mass is desired. Strength training and/or resistance exercise may cause muscle tissue to increase. For example, pushing against or pulling on a stationary object with a certain amount of force may trigger the cells in the associated muscle to change and cause the muscle mass to increase.
The subject matter disclosed herein relates to a control system for an exercise machine, not only capable of enabling an individual, preferably an older, less mobile individual or preferably an individual recovering from a fracture, to engage easily in osteogenic exercises and/or muscle strengthening exercises, but capable of using predetermined thresholds or dynamically calculating them, such that the person using the machine can be immediately informed through real-time visual and/or other sensorial feedback, that the osteogenic threshold has been exceeded, thus triggering osteogenesis for the subject bone (or bones), and/or that the muscular strength threshold has been exceeded, thereby triggering muscular hypertrophy for the subject muscle (or muscles). The control system may be used to improve compliance with an exercise plan including one or more exercises.
The control system may receive one or more load measurements associated with forces exerted by both the left and right sides on left and right portions (e.g., handles, foot plate or platform) of the exercise machine to enhance osteogenesis, bone growth, bone density improvement, and/or muscle mass. The one or more load measurements may be a left load measurement of a load added to a left load cell on a left portion of the exercise machine and a right load measurement of a load added to a right load cell on a right portion of the exercise machine. A user interface may be provided by the control system that presents visual representations of the separately measured left load and right load where the respective left load and right load are added to the respective left load cell and right load cell at the subject portions of the exercise machine.
In some implementations, initially, the control system may receive load measurements via a data channel associated with each exercise of the machine. For example, there may be a data channel for a leg-press-style exercise, a pull-down-style exercise, a suitcase-lift-style exercise, an arm-curl-style exercise, and so forth. Each data channel may include one or more load cells (e.g., a left load cell and a right load cell) that measure added load or applied force and transmit the load measurement to the control system via its respective data channel. The control system may receive the load measurements from each of the data channels at a first rate (e.g., 1 Hertz). If the control system detects a load from a data channel (e.g., hands resting on the handles including the respective load cells, or feet resting on the feet plate including the respective load cells), the control system may set that data channel as active and start reading load measurements from that data channel at a second rate (e.g., 10 Hertz) that is higher than the first rate. Further, the control system may set the other exercises associated with the other data channels as inactive and stop reading load measurements from the other data channels until the active exercise is complete. The active exercise may be complete when the one or more load measurements received via the data channel exceed one or more target thresholds. In some implementations, the control system may determine an average load measurement by accumulating raw load measurements over a certain period of time (e.g., 5 seconds) and averaging the raw load measurements to smooth the data (e.g., eliminates jumps or spikes in data) in an average load measurement.
The control system may compare the one or more load measurements (e.g., raw load measurements, or averaged load measurements) to one or more target thresholds. In some implementations, a single load measurement may be compared to a single specific target threshold (e.g., a one-to-one relationship). In some implementations, a single load measurement may be compared to more than one specific target threshold (e.g., a one-to-many relationship). In some implementations, more than one load measurement may be compared to a single specific target threshold (e.g., a many-to-one relationship). In some implementations, more than one load measurement may be compared to more than one specific target threshold (e.g., a many-to-many relationship).
The target thresholds may be an osteogenesis target threshold, a muscular strength target threshold, and/or a rehabilitation threshold. The osteogenesis target threshold may be determined based on a disease protocol pertaining to the user, an age of the user, a gender of the user, a sex of the user, a height of the user, a weight of the user, a bone density of the user, etc. A disease protocol may refer to any illness, disease, fracture, or ailment experienced by the user and any treatment instructions provided by a caretaker for recovery and/or healing. The disease protocol may also include a condition of health where the goal is avoid a problem. The muscular strength target threshold may be determined based on a historical performance of the user using the exercise machine (e.g., number of pounds lifted for a particular exercise, amount of force applied associated with each body part, etc.) and/or other exercise machines, a fitness level (e.g., how active the user is) of the user, a diet of the user, a protocol for determining a muscular strength target, etc. The rehabilitation target threshold may be determined based on historical performance of the user using the exercise machine (e.g., amount of force applied associated with each body part, speed of cycling, level of stability, etc.) and/or other exercise machines, a fitness level (e.g., how active the user is, the flexibility of the user, etc.) of the user, a diet of the user, an exercise plan for determining a rehabilitation target, the condition of the user (e.g., type of surgery the user underwent, the type of injury the user sustained), physical characteristics of the user (e.g., an age of the user, a gender of the user, a sex of the user, a height of the user, a weight of the user, a bone density of the user), condition of the user's body part(s) (e.g., the pain level of a user), an exertion level of a user (e.g., how easy/hard the exercise session is for the user), any other suitable characteristic, or combination thereof.
The control system may determine whether the one or more load measurements exceed the one or more target thresholds. Responsive to determining that the one or more load measurements exceed the one or more target thresholds, the control system may cause a user interface to present an indication that the one or more target thresholds have been exceeded and an exercise is complete. Additionally, when the one or more target thresholds are exceeded, the control system may cause the user interface to present an indication that instructs the user to apply additional force (less than a safety limit) to attempt to set a personal maximum record of weight lifted, pressed, pulled, or otherwise exert force thereupon for that exercise.
Further, the user interface may present an indication when a load measurement is approaching a target threshold for the user. In another example, when the load measurement exceeds the target threshold, the user interface may present an indication that the target threshold has been exceeded, that the exercise is complete, and if there are any remaining incomplete exercises in the exercise plan, that there is another exercise to be completed by the user. If there are no remaining exercises in the exercise plan to complete, then the user interface may present an indication that all exercises in the exercise plan are complete and the user can rest. In addition, when the exercise plan is complete, the control system may generate a performance report that presents various information (e.g., charts and graphs of the right and left load measurements received during each of the exercises, left and right maximum loads for the user received during each of the exercises, historical right and left load measurements received in the past, comparison of the current right and left load measurements with the historical right and left load measurement, an amount of pounds lifted or pressed that is determined based on the load measurements for each of the exercises, percent gained in load measurements over time, etc.).
Further, the one or more load measurements may each be compared to a safety limit. For example, a left load measurement and a right load measurement may each be compared to the safety limit for the user. The safety limit may be determined for the user based on the user's disease protocol. There may be different safety limits for different portions of the user's body on the left and the right side, one extremity versus another extremity, a top portion of the user's body and a body portion of the user's body, etc., and for different exercises. For example, if someone underwent left knee surgery, the safety limit for a user for a left load measurement for a leg-press-style exercise may be different from the safety limit for a right load measurement for that exercise and user. If the safety limit is exceeded, an indication may be presented on the user interface to instruct to reduce the amount of force the user is applying and/or to instruct the user to stop applying force because the safety limit is exceeded.
For those with any or all of the osteoporosis-exacerbating medical conditions described herein, such a control system and exercise machine can slow the rate of net bone loss by enabling osteogenesis to occur without exertions which would not be possible for someone whose health is fragile, not robust. Another benefit of the present disclosure, therefore, is its ability to speed the healing of fractures in athletically robust individuals. Further, another benefit is the increase in muscle mass by using the exercise machine to trigger muscular hypertrophy. The control system may provide an automated interface that improves compliance with an exercise plan by using a real-time feedback loop to measure loads added during each of the exercises, compare the load measurements to target thresholds and/or safety limits that are uniquely determined for the user using the exercise machine, and provide various indications based on the comparison. For example, the indications pertain to when the user should add more load, when the target thresholds are exceeded, when the safety limit is exceeded, when the exercise is complete, when the user should begin another exercise, and so forth.
Bone Exercises and their Benefits
The following exercises achieve bone strengthening results by exposing relevant parts of a user to isometric forces which are selected multiples of body weight (MOB) of the user, a threshold level above which bone mineral density increases. A MOB may be any fraction or rational number excluding zero. The specific MOB-multiple threshold necessary to effect such increases will naturally vary from individual to individual and may be more or less for any given individual. “Bone-strengthening,” as used herein, specifically includes, without limitation, a process of osteogenesis, whether due to the creation of new bone as a result of an increase in the bone mineral density; or proximately to the introduction or causation of microfractures in the underlying bone. The exercises referred to are as follows.
Leg Press
A leg-press-style exercise to improve isometric muscular strength in the following key muscle groups: gluteals, hamstrings, quadriceps, spinal extensors and grip muscles as well as to increase resistance to skeletal fractures in leg bones such as the femur. In one example, the leg-press-style exercise can be performed approximately 4.2 MOB or more of the user.
Chest Press
A chest-press-style exercise to improve isometric muscular strength in the following key muscle groups: pectorals, deltoids, and tricep and grip muscles as well as in increasing resistance to skeletal fractures in the humerus, clavicle, radial, ulnar and rib pectoral regions. In one example, the chest-press-style exercise can be performed at approximately 2.5 MOB or more of the user.
Suitcase Lift
A suitcase-lift-style exercise to improve isometric muscular strength in the following key muscle groups: gluteals, hamstrings, quadriceps, spinal extensors, abdominals, and upper back and grip muscles as well as to increase resistance to skeletal fractures in the femur and spine. In one example, the suitcase-lift-style exercise can be performed at approximately 2.5 MOB or more of the user.
Arm Curl
An arm-curl-style exercise to improve isometric muscular strength in the following key muscle groups: biceps, brachialis, brachioradialis, grip muscles and trunk as well as in increasing resistance to skeletal fractures in the humerus, ribs and spine. In one example, the arm-curl-style exercise can be performed at approximately 1.5 MOB or more of the user.
Core Pull
A core-pull-style exercise to improve isometric muscular strength in the following key muscle groups: elbow flexors, grip muscles, latissimus dorsi, hip flexors and trunk as well as in increasing resistance to skeletal fractures in the ribs and spine. In one example, the core-pull-style exercise can be performed at approximately 1.5 MOB or more of the user.
Grip Strength
A grip-strengthening-style exercise which may preferably be situated around a station in an exercise machine, in order to improve strength in the muscles of the hand and forearm. Grip strength is medically salient because it has been positively correlated with better states of health.
In some implementations, a balance board may be communicatively coupled to the control system. For example, the balance board may include a network interface that communicates with the control system via any suitable interface protocol (e.g., Bluetooth, WiFi, cellular). The balance board may include pressure sensors and may obtain measurements of locations and amount of pressure applied to the balance board. The measurements may be transmitted to the control system. The control system may present a game or interactive exercise on a user interface. The game or interactive exercise may modify screens or adjust graphics that are displayed based on the measurements received from the balance board. The balance board may be used by a user to perform any suitable type of plank (e.g., knee plank, regular feet and elbow plank, table plank with elbows, or the like). Accordingly, the balance board may be configured to be used with arms on the balance board, knees on the balance board, and/or feet standing on the balance board. The games or interactive exercises may encourage the user during the game or interactive exercises to increase compliance and neuro-motor control after a surgery, for example.
The exercise machine, balance board, wristband, goniometer, and/or any suitable accessory may be used for various reasons in various markets. For example, users may use the exercise machine, balance board, wristband, goniometer, and/or any suitable accessory in the orthopedic market if the users suffer from chronic musculosketal pain (e.g., knees, hips, shoulders, and back). The exercise machine, balance board, wristband, goniometer, and/or any suitable accessory may be used to help with prehabilitation (prehab), as well as optimize post-surgical outcomes. Users may use the exercise machine, balance board, wristband, goniometer, and/or any suitable accessory in the back and neck pain market if the users suffer with chronic back and neck pain and they want to avoid surgery and experience long-term relief, as well as users that are in recovery following surgery. Users may use the exercise machine, balance board, wristband, goniometer, and/or any suitable accessory in the cardiovascular market if they desire to prevent or recover from life-threatening cardiovascular disease, especially heart attacks and stroke. Users may use the exercise machine, balance board, wristband, goniometer, and/or any suitable accessory in the neurological market if they desire to recover from stroke, or have conditions like Parkinson's Disease and/or Multiple Sclerosis, and the users desire to achieve better balance, strength, and muscle symmetry in order to slow progression of the medical condition.
A user may desire to perform a series of exercises to attempt to exert up to or approximately near to a target, total amount of energy. However, a user may not have proper knowledge, training, and/or education to determine which specific exercises to perform. A metabolic equivalent of task (MET) score (also referred to as a MET indicator or a MET value) represents a ratio of working metabolic rate of a person relative to the person's resting metabolic rate. Metabolic rate is a rate calculated by determining a measure of energy expended per unit of time. A MET score is one way to describe the intensity of an exercise or activity. However, MET scores alone may not provide a comprehensive approximation of the energy that a specific user exerts while performing different exercises. For example, MET scores are not intended to account for a specific user's fitness level. Further, MET scores are not intended to account for a specific user's pain levels and pain tolerance.
Accordingly, some implementations of the present disclosure provide a technical solution of comprising user energy consumption metrics for different exercises and generating an exercise plan that enables the user to attempt to achieve a user energy score. The system may use an artificial intelligence engine to generate machine learning models trained to generate user energy consumption metrics based on associations between user fitness test results and one or more exercises and on associations between MET scores and one or more exercises.
Further, the machine learning models may be trained based on associations between user pain levels and one or more exercises. In addition, the machine learning models may be trained to generate exercise programs including exercise plans, wherein the exercise plans include one or more exercises. Each of the one or more exercises may be associated with a user energy consumption metric. The user energy consumption metrics associated with the exercises may be summed, or in other embodiments, mathematically analyzed, processed or manipulated, to provide a user energy score associated with the exercise program. The machine learning models may be trained to determine user energy scores for physical activity goals. The machine learning models may be trained to match a first energy score that is associated with an exercise program with a second energy score that is associated with a physical activity goal selected by a user or selected for the user (e.g., by a coach based on user data related to a physical state, a mental or intellectual state, an emotional or psychological state, etc.).
In some implementations, the user energy score may be determined based on an amount of energy it takes to attempt to achieve (including achieving) a physical activity goal of the user. Bone growth, muscle growth, rehabilitation, prehabilitation, and the like may be preferred, desirable, or necessary to perform certain physical activities. For example, a person may require a certain amount of muscle mass to move an object having a particular weight. The physical activity may be desirable, but some people may lack the appropriate bone mass, muscle mass, or physical ability in general to perform the physical activity. In one example, a grandparent may desire to play with their grandchildren, and may want to select that physical activity as a goal.
Using various performance measurements from one or more sensors, attributes of users of the exercise machine, user-reported difficulty levels of exercises, user-reported pain levels, and the like, the user energy consumption metrics may be objectively monitored and/or measured. An onboarding protocol may be used to baseline a fitness level of the user, and the fitness level of the user may be used to generate use energy consumption metrics. A machine learning model may be trained to perform the onboarding protocol and to determine the fitness level of the user. The user energy consumption metrics (and the exercise plan) may be dynamically updated based on attributes of the user, selected physical activity levels, performance measurements, user-reported difficulty(-ies) of the exercises, user-reported pain levels, and the like. In some implementations, the exercise machines may be controlled using a signal that changes an attribute of an operating parameter of the exercise machine to comply with the exercise plan. Responsive to receiving the signal, the control system may change the attribute of the operating parameter.
In some embodiments, users may be grouped, based on energy scores, into cohorts and associated with a particular exercise program or class. For example, the users may select various physical activity goals wherein each physical activity goal is associated with a user energy score by one or more machine learning models. The machine learning models may be trained to associate the users with a particular exercise program by matching the user energy score associated with the exercise program and the user energy score associated with the selected physical activity goals. As used throughout this disclosure, the terms “match,” “match with,” or “matches with” may refer to an exact match, a correlative match, a substantial match, a statistically measured match, etc. Using the exercise machines, the users in the exercise program may perform an exercise plan associated with the exercise program. The user's progress toward achieving respective targets associated with user energy consumption metrics associated with exercises in the exercise plan may be monitored by one or more of the users and/or by one or more coaches in real-time and near real-time. In some embodiments, the one or more of the users may be in the same physical location and/or one or more of the users may be in a different physical location. In some embodiments, the one or more of the coaches may be in the same physical location and/or one or more of the coaches may be in a different physical location.
The following discussion is directed to various implementations of the present disclosure. Although these implementations are given as examples, the implementations disclosed should not be interpreted, or otherwise used, as limiting the scope of the present disclosure, including the claims. In addition, one of ordinary skill in the art will understand that the following description has broad application, and the discussion of any implementations is meant only to be exemplary of that implementations, and not intended to intimate that the scope of the present disclosure, including the claims, is limited to that implementations.
The network interface devices may enable communication via a wireless protocol for transmitting data over short distances, such as Bluetooth, ZigBee, near field communication (NFC), etc. In some implementations, the computing device 12 is communicatively coupled to the exercise machine 100 via Bluetooth. Additionally, the network interface devices may enable communicating data over long distances, and in one example, the computing device 12 may communicate with a network 20. Network 20 may be a public network (e.g., connected to the Internet via wired (Ethernet) or wireless (WiFi)), a private network (e.g., a local area network (LAN), wide area network (WAN), virtual private network (VPN)), or a combination thereof.
The computing device 12 may be any suitable computing device, such as a laptop, tablet, smartphone, or computer. The computing device 12 may include a display that is capable of presenting a user interface 18 of an application 17. The application 17 may be implemented in computer instructions stored on the one or more memory devices of the computing device 12 and executable by the one or more processing devices of the computing device 12. The application 17 may be a stand-alone application that is installed on the computing device 12 or may be an application (e.g., website) that executes via a web browser. The user interface 18 may present various screens to a user that enable the user to login, enter personal information (e.g., health information; a disease protocol prescribed by a physician, trainer, or caretaker; age; gender; activity level; bone density; weight; height; patient measurements; etc.), view an exercise plan, initiate an exercise in the exercise plan, view visual representations of left load measurements and right load measurements that are received from left load cells and right load cells during the exercise, view a weight in pounds that are pushed, lifted, or pulled during the exercise, view an indication when the user has almost reached a target threshold, view an indication when the user has exceeded the target thresholds, view an indication when the user has set a new personal maximum for a load measurement and/or pounds pushed, lifted, or pulled, view an indication when a load measurement exceeds a safety limit, view an indication to instruct the user to begin another exercise, view an indication that congratulates the user for completing all exercises in the exercise plan, and so forth, as described in more detail below. The computing device 12 may also include instructions stored on the one or more memory devices that, when executed by the one or more processing devices of the computing device 12, perform operations to control the exercise machine 100.
The computing device 15 may execute an application 21. The application 21 may be implemented in computer instructions stored on the one or more memory devices of the computing device 15 and executable by the one or more processing devices of the computing device 15. The application 21 may present a user interface 22 including various screens to a physician, trainer, or caregiver that enable the person to create an exercise plan for a user based on a treatment (e.g., surgery, medical procedure, etc.) the user underwent and/or injury (e.g., sprain, tear, fracture, etc.) the user suffered, view progress of the user throughout the exercise plan, and/or view measured properties (e.g., force exerted on portions of the exercise machine 100) of the user during exercises of the exercise plan. The exercise plan specific to a patient may be transmitted via the network 20 to the cloud-based computing system 16 for storage and/or to the computing device 12 so the patient may begin the exercise plan. The exercise plan may specifying one or more exercises that are available at the exercise machine 100.
The exercise machine 100 may be an osteogenic, muscular strengthening, isometric exercise and/or rehabilitation assembly. Solid state, static, or isometric exercise and rehabilitation equipment (e.g., exercise machine 100) can be used to facilitate osteogenic exercises that are isometric in nature and/or to facilitate muscular strengthening exercises. Such exercise and rehabilitation equipment can include equipment in which there are no moving parts while the user is exercising. While there may be some flexing under load, incidental movement resulting from the tolerances of interlocking parts, and parts that can move while performing adjustments on the exercise and rehabilitation equipment, these flexions and movements can comprise, without limitation, exercise and rehabilitation equipment from the field of isometric exercise and rehabilitation equipment.
The exercise machine 100 may include various load cells 110 disposed at various portions of the exercise machine 100. For example, one or more left load cells 110 may be located at one or more left feet plates or platforms, and one or more right load cells may be located at one or more right feet plates or platforms. Also, one or more left load cells may be located at one or more left handles, and one or more right load cells may be located at one or more right handles. Each exercise in the exercise system may be associated with both a left and a right portion (e.g., handle or foot plate) of the exercise machine 100. For example, a leg-press-style exercise is associated with a left foot plate and a right foot plate. The left load cell at the left foot plate and the right load cell at the right foot plate may independently measure a load added onto the left foot plate and the right foot plate, respectively, and transmit the left load measurement and the right load measurement to the computing device 12. The load added onto the load cells 110 may represent an amount of weight added onto the load cells. In some implementations, the load added onto the load cells 110 may represent an amount of force exerted by the user on the load cells. Accordingly, the left load measurement and the right load measurement may be used to present a left force (e.g., in Newtons) and a right force (e.g., in Newtons). The left force and right force may be totaled and converted into a total weight in pounds for the exercise. Each of the left force, the right force, and/or the total weight in pounds may be presented on the user interface 18.
In some implementations, the cloud-based computing system 16 may include one or more servers 28 that form a distributed, grid, and/or peer-to-peer (P2P) computing architecture. Each of the servers 28 may include one or more processing devices, memory devices, data storage, and/or network interface devices. The servers 28 may be in communication with one another via any suitable communication protocol. The servers 28 may store profiles for each of the users that use the exercise machine 100. The profiles may include information about the users such as one or more disease protocols, one or more exercise plans, a historical performance (e.g., loads applied to the left load cell and right load cell, total weight in pounds, etc.) for each type of exercise that can be performed using the exercise machine 100, health, age, race, credentials for logging into the application 17, and so forth.
In some implementations, the cloud-based computing system 16 may include a training engine 50 and/or an artificial intelligence engine 65. The cloud-based computing system 16 may include one or more servers 28 that execute the artificial intelligence engine 65 that uses one or more machine learning models 60 to perform at least one of the embodiments disclosed herein. In some implementations, the training engine 50 may be included as part of the artificial intelligence engine 65 and the artificial intelligence engine 65 may execute the training engine 50. In some implementations, the artificial intelligence engine 65 may use the training engine 50 to generate the one or more machine learning models 60.
The artificial intelligence engine 65, the training engine 50, and/or the one or more machine learning models 60 may be communicatively coupled to the servers 28 or may be included in one of the servers 28. In some implementations, the artificial intelligence engine 65, the training engine 50, and/or the machine learning models 60 may be included in the computing device 12.
The one or more of machine learning models 60 may refer to model artifacts created, using training data that includes training inputs and corresponding target outputs (correct answers for respective training inputs), by the artificial intelligence engine 65 and/or the training engine 50. The training engine 50 may find patterns in the training data that map the training input to the target output (the answer to be predicted), and provide the machine learning models 60 that capture these patterns. As described in more detail below, the set of machine learning models 60 may be composed of, e.g., a single level of linear or non-linear operations (e.g., a support vector machine (SVM)) or may be a deep network, i.e., a machine learning model composed of multiple levels of non-linear operations. Examples of deep networks are neural networks including convolutional neural networks, recurrent neural networks with one or more hidden layers, and fully connected neural networks.
In some implementations, the training data may include various inputs (e.g., a physical activity goal, range of motion of users, user-reported pain level of users, user-reported difficulty levels of exercises, exercise information, levels of attainment, attributes of users (e.g., age, weight, height, gender, procedures performed, condition of user, goals for outcomes of exercising, etc.), performance measurements, and the like) and mapped outputs. The mapped outputs may include an exercise plan composed on various exercise sessions each including various exercises, schedule of the exercise sessions, etc. In some implementations, the training data may include other inputs (e.g., state of the exercise session, exercise, exercise machine 100; progress of the user; events; attributes of the user; etc.) and other mapped outputs. The other mapped outputs may include comorbidity information pertaining to the user. The other mapped outputs may further include multimedia (e.g., video/audio) clips or segments for a virtual coach to speak, graphic images, video, and the like to be presented on the user interface 18 of the computing device 12 before, during, or after the user performs the exercises. The virtual coach may be implemented in computer instructions as part of application 17 executing on the computing device 12. The virtual coach may be driven and controlled by artificial intelligence (e.g., via one or more machine learning models 60). For example, the machine learning model 60 may be trained to implement the virtual coach. Further, the training data may include inputs pertaining to user feedback and/or progress of the user and outputs pertaining to a persona for the virtual coach to implement. The training data may include inputs of the progress of the user (e.g., completion of an exercise) and output various incentives, rewards, and/or certificates. The training data may include inputs of the progress of the user and/or the exercise plan and may output notifications pertaining to the progress and/or the exercise plan. The training data may include inputs of user-reported pain levels, user-reported difficulty of exercises, difficulty levels of the exercises, etc. and may include mapped outputs of modifying the exercise plan (e.g., removing an exercise, switching an exercise to another exercise, adding an exercise, modifying an exercise session, adding an exercise session, removing an exercise session, etc.). The machine learning model 60 may be trained using any and/or all of the training data.
In some embodiments, the machine learning model 60 may be trained with training data including inputs of exercise data associated with user energy consumption metrics and outputs of exercise plans to include in an exercise program (also referred to as a class herein) and an energy score associated with the exercise program. In some embodiments, the machine learning model 60 may be trained with training data including inputs of physical activity goals associated with energy scores and outputs of exercise programs having matching energy scores. In some embodiments, the machine learning model 60 may be trained with training data including inputs of user energy consumption metrics and outputs of signals to transmit to change an operating parameter of exercise machines 100.
In some implementations, the training engine 50 may train the machine learning models 60 to output an exercise plan, wherein such plan may include a schedule of exercise sessions and selected exercises for each of the exercise sessions. Based on the inputs described herein, the trained machine learning model 60 may select the exercises by filtering a set of exercises included in a tagged data structure (e.g., data source). The machine learning model 60 may be trained to control the virtual coach executing on the computing device 12. The machine learning model 60 may also be trained to provide incentives, rewards, and/or certificates to the user. The machine learning model 60 may also be trained to modify the exercise plan and/or directly or indirectly control the exercise machine 100 based on the progress of the user and/or feedback of the user (e.g., indications of a difficulty level of an exercise). For example, if the user indicates an exercise is too easy, the machine learning model 60 may choose a new intensity for the exercise and the cloud-based computing system 16 may distally control the exercise machine 100 by increasing the intensity. Any suitable number of machine learning models 60 may be used. For example, separate machine learning models 60 may be used for each respective function described above, and the machine learning models 60 may be linked such that the output from one machine learning model 60 may be input into another machine learning model 60.
The cloud-based computing system may include a data source 67 that stores the training data for the training engine 50 and/or the artificial intelligence engine 65 to use to train the one or more machine learning models 60. The data source may include exercises, physical activity goals, levels of attainment, body portions targeted by exercises, weights, and/or parameters used to configure a prioritization of certain levels of attainment throughout an exercise schedule, comorbidity information, health-related information, audio segments, video segments, motivational quotations, and so forth. The data source 67 may include various tags and/or keys (e.g., primary, foreign, etc.) to associate items of the data with each other in the data source 67. The data source 67 may be a relational database, a pivot table, or any suitable type of data structure configured to store data used for any of the operations described herein.
During exercise, a user can grip and apply force to one of the pairs of load handles 104, 106, 108. The term “apply force” can include a single force, more than one force, a range of forces, etc. and may be used interchangeably with “addition of load.” Each load handle in the pairs of load handles 104, 106, 108 can include at least one load cell 110 for separately and independently measuring a force applied to, or a load added onto, respective load handles. Further, each foot plate 118 (e.g., a left foot plate and a right foot plate) can include at least one load cell 110 for separately and independently measuring a force applied to, or a load added onto, respective foot plates.
The placement of a load cell 110 in each pair of load handles 104, 106, 108 and/or feet plates 118 can provide the ability to read variations in force applied between the left and right sides of the user. This allows a user or trainer to understand relative strength. This is also useful in understanding strength when recovering from an injury.
In some implementations, the assembly 101 further can include the computing device 12. One or more of the load cells 110 can be individually in electrical communication with the computing device 12 either via a wired or wireless connection. In some implementations, the user interface 18 presented via a display of the computing device 12 may indicate how to perform an exercise, how much force is being applied, a target force to be applied, historical information for the user about how much force they applied at prior sessions, comparisons to averages, etc., as well as additional information, recommendations, notifications, and/or indications described herein.
In some implementations, the assembly 101 further includes a seat 112 supported by the frame 102 in which a user sits while applying force to the load handles and/or feet plates. In some implementations, the seat 112 can include a support such as a backboard 114. In some implementations, the position of the seat 112 is adjustable in a horizontal and/or vertical dimension. In some implementations, the angle of the seat 112 is adjustable. In some implementations, the angle of the backboard 114 is adjustable. Examples of how adjustments to the seat 112 and backboard 114 can be implemented include, but are not limited to, using telescoping tubes and pins, hydraulic pistons, electric motors, etc. In some implementations, the seat 112 can further include a fastening system 116 (
In one example, the seat 112 can include a base 113 that is slidably mounted to a horizontal rail 111 of the frame 102. The seat 112 can be selectively repositionable and secured as indicated by the double-headed arrow. In another example, the seat 112 can include one or more supports 117 (e.g., two shown) that are slidably mounted to a substantially vertical rail 115 of the frame 102. The seat 112 can be selectively repositionable and secured as indicated by the double-headed arrow.
In some implementations, a pair of feet plate 118 can be located angled toward and in front of the seat 112. The user can apply force to the feet plate 118 (
In some implementations, adjustments can be made to the position of the pair of feet plate 118. For example, these adjustments can include the height of the pair of feet plate 118, the distance between the pair of feet plate 118 and the seat 112, the distance between each handle of the pair of feet plate 118, the angle of the pair of feet plate 118 relative to the user, etc. In some implementations, to account for natural differences in limb length or injuries, each foot plate of the pair of feet plate 118 can be adjusted separately.
In some implementations, a first pair of load handles 104 can be located above and in front of the seat 112. The user can apply force to the load handles 104 (
In some implementations, adjustments can be made to the position of the first pair of load handles 104. For example, these adjustments can include the height of the first pair of load handles 104, the distance between the first pair of load handles 104 and the seat 112, the distance between each handle of the first pair of load handles 104, the angle of the first load handles 104 relative to the user, etc. In some implementations, to account for natural differences in limb length or injuries, each handle of the first pair of load handles 104 can be adjusted separately.
In one example, the first pair of load handles 104 can include a sub-frame 103 that is slidably mounted to a vertical rail 105 of the frame 102. The first pair of load handles 104 can be selectively repositionable and secured as indicated by the double-headed arrow.
In some implementations, a second pair of load handles 106 can be spaced apart from and in the front of the seat 112. While seated (
In some implementations, adjustments can be made to the position of the second pair of load handles 106. These adjustments can include the height of the second pair of load handles 106, the distance between the second pair of load handles 106 and the seat 112, the distance between each handle of the second pair of load handles 106, the angle of the second load handles 106 relative to the user, etc. In some implementations, to account for natural differences in limb length or injuries, each handle of the second pair of load handles 106 can be adjusted separately.
In one example, the second pair of load handles 106 can include the sub-frame 103 that is slidably mounted to the vertical rail 105 of the frame 102. The sub-frame 103 can be the same sub-frame 103 provided for the first pair of load handles 104, or a different, independent sub-frame. The second pair of load handles 106 can be selectively repositionable and secured as indicated by the double-headed arrow.
In some implementations (
In some implementations, adjustments can be made to the position of the third pair of load handles 108. These adjustments can include the height of the third pair of load handles 108, the distance between the third pair of load handles 108 and the seat 112, the distance between each handle of the third pair of load handles 108, the angle of the third load handles 108 relative to the user, etc. In some implementations, to account for natural differences in limb length or injuries, each handle of the third pair of load handles 108 can be adjusted separately.
In one example, each load handle 108 of the third pair of load handles 108 can include a sub-frame 109 that is slidably mounted in or to a vertical tube 107 of the frame 102. Each load handle 108 of the third pair of load handles 108 can be selectively repositionable and secured as indicated by the double-headed arrows.
In other implementations (not shown), the third pair of load handles 108 can be reconfigured to be coaxial and located horizontally in front of the user along an axis that is perpendicular to the vertical plane. The user can apply force to the third pair of load handles 108 in a deadlift-style exercise. Like the suitcase-lift-style exercise, the deadlift-style exercise can provide or enable osteogenesis, bone growth or bone density improvement for a portion of the skeletal system of the user. Further, the deadlift-style exercise can provide or enable muscular hypertrophy for one or more muscles of the user. In the deadlift-style exercise, the user can stand on the floor or a horizontal portion of the frame 102, bend their knees, hold the third pair of load handles 108 in front of them, and extend their legs to apply an upward force to the third pair of load handles 108. In some implementations, the third pair of load handles 108 can be adjusted (e.g., rotated) from the described coaxial position used for the deadlift-style exercise, to the parallel position (
In general, the user interface 18 may present real-time visual feedback of the current load measurements or the current forces corresponding to the load measurements, a weight in pounds associated with the load measurements, incentive messages that encourage the user to exceed target thresholds (e.g., to trigger osteogenesis and/or muscular hypertrophy) and/or set personal records for maximum loads, historical performance of the user performing the exercise, and/or scripted prompts that display images of one or more body portions indicating proper technique for performing the exercise. The control system may provide various visual, audio, and/or haptic feedback to encourage the user to exceed their target thresholds.
Initially, when the user has not added load onto any portion of the exercise machine 100 including one or more load cells 110, the computing device 12 may be operating in an idle mode. During the idle mode, the computing device 12 may be receiving load measurements at a first frequency from each data channel associated with an exercise. For example, there may be four data channels, one for each of a chest-press-style exercise, a leg-press-style exercise, a suitcase-lift-style exercise, and a pulldown-style exercise. Although four data channels are described for explanatory purposes, it should be understood that there may be any suitable number of data channels, where “any” refers to one or more. Each data channel may provide load measurements to the computing device 12 from a respective left load cell and a respective right load cell that are located at the portion of the exercise machine 100 where the user pushes or pulls for the respective exercises. The user interface 18 may present the load measurement from each left and right load cells (e.g., 8 load measurements for the 4 data channels associated with the 4 exercises). Further, any target thresholds and/or safety limits for the user performing the exercises may be presented on the user interface 18 during the idle mode. For example, a left target threshold, a right target load threshold, a safety limit, and/or a total weight target threshold for each of the exercises may be presented on the user interface 18 during the idle mode.
If the computing device 12 detects a minimum threshold amount of load (e.g., at least 10 pound-force (lbf)) added onto any of the load cells, the computing device switches from an idle mode to an exercise mode. The data channel including the load cell that sent the detected load measurement may be set to active by the computing device 12. Further, the computing device 12 may set the other data channels to inactive and may stop receiving load measurements from the load cells corresponding to the inactive data channels. The computing device 12 may begin reading data from the load cells at the active data channel at a second frequency higher (e.g., high frequency data collection) than the first frequency when the computing device 12 was operating in the idle mode. Further, the user interface 18 may switch to presenting information pertaining to the exercise associated with the active data channel and stop presenting information pertaining to the exercises associated with the inactive data channels.
For example, the user may grip the second pair of handles 106 and apply force. The computing device 12 may detect the load from the load cells 110 located at the second pair of handles 106 and may set the data channel associated with the chest-press-style exercise to active to begin high frequency data collection from the load cells 110 via the active data channel.
As depicted, the user interface 18 presents a left load measurement 1000 as a left force and a right load measurement 1002 as a right force in real-time or near real-time as the user is pressing on the second pair of handles 106. The values of the forces for the left load measurement 1000 and the right load measurement 1002 are presented. There are separate visual representations for the left load measurement 1000 and the right load measurement 1002. In some implementations, these load measurements 1000 and 1002 may be represented in a bar char, line chart, graph, or any suitable visual representation. In some implementations, a left target threshold and a right target threshold for the user may be presented on the user interface 18. In some implementations, there may be more than one left target threshold and more than one right target threshold. For example, the left target thresholds may relate to an osteogenesis target threshold determined using a user's disease protocol and/or a muscular strength target threshold determined using a historical performance of the user for a particular exercise. The right target thresholds may relate to an osteogenesis target threshold determined using a user's disease protocol and/or a muscular strength target threshold determined using a historical performance of the user for a particular exercise. For example, if the user fractured their left arm and is rehabilitating the left arm, but the user's right arm is healthy, the left osteogenesis target threshold may be different from the right osteogenesis target threshold.
If the left load measurement 1000 exceeds any of the left target thresholds, an indication (e.g., starburst) may be presented on the user interface 18 indicating that the particular left target threshold has been exceeded and/or osteogenesis and/or muscular hypertrophy has been triggered in one or more portions of the body. If the right load measurement 1002 exceeds any of the right target thresholds, an indication (e.g., starburst) may be presented on the user interface 18 indicating that the particular right target threshold has been exceeded and/or osteogenesis and/or muscular hypertrophy has been triggered in another portion of the body. Further, if either or both of the left and right target thresholds are exceeded, the indication may indicate that the exercise is complete and a congratulatory message may be presented on the user interface 18. In some implementations, another message may be presented on the user interface 18 that encourages the user to continue adding load to set a new personal maximum left load measurement and/or right load measurement for the exercise.
In some implementations, there may be a single target threshold to which both the left load measurement and the right load measurement are compared. If either of the left or right load measurement exceed the single target threshold, the above-described indication may be presented on the user interface 18.
In some implementations, there may be a single safety limit to which the left and right load measurements are compared. The single safety limit may be determined based on the user's disease protocol (e.g., what type of disease the user has, a severity of the disease, an age of the user, the height of the user, the weight of the user, what type of injury the user sustained, what type of surgery the user underwent, the portion of the body affected by the disease, the exercise plan to rehabilitate the user's body, instructions from a caregiver, etc.). If either or both of the left and right load measurements exceed the single safety limit, an indication may be presented on the user interface 18. The indication may warn the user that the safety limit has been exceeded and recommend to reduce the amount of load added to the load cells 110 associated with the exercise being performed by the user.
In some implementations, more than one safety limit may be used. For example, if the user is rehabilitating a left leg, but a right leg is healthy, there may be a left safety limit that is determined for the left leg based on the user's disease protocol and there may be a right safety limit for the left leg determined based on the user's disease protocol. The left load measurement may be compared to the left safety limit, and the right load measurement may be compared to the right safety limit. If either or both the left load measurement and/or the right load measurement exceed the left safety limit and/or the right safety limit, respectively, an indication may be presented on the user interface 18. The indication may warn the user that the respective safety limit has been exceeded and recommend to reduce the amount of load added to the load cells 110 associated with the exercise being performed by the user.
Further, a total weight 1004 in pounds that is determined based on the left and right load measurements is presented on the user interface 18. The total weight 1004 may dynamically change as the user adds load onto the load cells 110. A target weight 1006 for the exercise for the current day is also presented. This target weight 1006 may be determined based on the user's historical performance for the exercise. If the total weight 1004 exceeds the target weight 1006, an indication (e.g., starburst) may be presented on the user interface 18 indicating that osteogenesis and/or muscular hypertrophy has been triggered. Further, the indication may indicate that the exercise is complete and a congratulatory message may be presented on the user interface 18. In some implementations, another message may be presented on the user interface 18 that encourages the user to continue adding load to set a new personal maximum record for the exercise.
Additionally, the user interface 18 may present a left grip strength 1008 and a right grip strength 1010. In some implementations, the left grip strength and the right grip strength may be determined based on the left load measurement and the right load measurement, respectively. Numerical values representing the left grip strength 1008 and the right grip strength 1010 are displayed. Any suitable visual representation may be used to present the grip strengths (e.g., bar chart, line chart, etc.). The grip strengths may only be presented when the user is performing an exercise using handles.
The user interface 18 may also present a prompt 1012 that indicates the body position the user should be in to perform the exercise, as well as indicate which body portions will be targeted by performing the exercise. The user interface 18 may present other current and historical information related to the user performing the particular exercise. For example, the user interface 18 may present a visual representation 1014 of the user's maximum weight lifted, pressed, pulled, or otherwise exerted force for the day or a current exercise session. The user interface 18 may present a visual representation 1016 of the user's previous maximum weight lifted, pressed, pulled, or otherwise exerted force. The user interface 18 may present a visual representation 1018 of the user's maximum weight lifted, pressed, pulled, or otherwise exerted force the first time the user performed the exercise. The user interface 18 may present one or more visual representations 1020 for a weekly goal including how many sessions should be performed in the week and progress of the sessions as they are being performed. The user interface 18 may present a monthly goal including how many sessions should be performed in the month and progress of the sessions as they are being performed. Additional information and/or indications (e.g., incentivizing messages, recommendations, warnings, congratulatory messages, etc.) may be presented on the user interface 18, as discussed further below.
After a person has an injury (e.g., sprain or fractured bone), a surgery (e.g., knee replacement), or a disease (e.g., muscular dystrophy), the person's body is typically in a weakened state (e.g., physically disabled). Thus, clinicians, such as doctors and physical therapists, can prescribe exercise plans for rehabilitating their patients. The exercises in these exercise plans help restore function, improve mobility, relieve pain, improve strength, improve flexibility, and, among other benefits, prevent or limit permanent physical disability in the patients. Patients who follow their exercise plans typically show signs of physical improvement and reduced pain at a faster the rate (i.e., a faster rate of recovery or rehabilitation).
In addition, after an injury or surgery, patients typically become less active than they once were, and they may experience muscle loss. As explained above, muscles that are not used often may reduce in muscle mass and become weaker. To increase the muscle mass and/or reduce the rate of muscle loss, people may conduct exercises according to an exercise plan.
Balancing and/or resistance exercise may cause muscle tissue to increase. For example, balancing on a balance board or pushing and pulling on a stationary object (e.g., pedals of an exercise cycle) with a certain amount of force may trigger the cells in the associated muscle to change and cause the muscle mass to increase.
The subject matter disclosed herein relates to a control system for an exercise machine, not only capable of enabling an individual, preferably an individual recovering from a fracture, an injury, or a surgery, to engage easily exercises according to an exercise plan, but capable of using predetermined thresholds or dynamically calculating them, such that the person using the exercise machine can be immediately informed through real-time visual and/or other sensorial feedback, that goals of the exercise plan has been met or exceeded, thus triggering osteogenesis for the subject bone (or bones), and/or that the muscular strength threshold has been exceeded, thereby triggering muscular hypertrophy for the subject muscle (or muscles). The control system may be used to improve compliance with an exercise plan, whereby the exercise plan includes one or more exercises.
The control system may receive one or more measurements, such as load measurements, associated with forces exerted by both the left and right sides on left and right portions (e.g., pedals, base, or platform) of the exercise machine to enhance osteogenesis, bone growth, bone density improvement, stability, flexibility, range of motion, and/or muscle mass. The one or more measurements (e.g., a load measurement) may be a left measurement of a load or an increased resistance added to a left load cell on a left portion of the exercise machine (e.g., a left pedal or a left portion of the platform) and a right measurement of a load or an increased resistance added to a right load cell on a right portion of the exercise machine (e.g., a right pedal or a right portion of the platform). A user interface may be provided by the control system that presents visual representations of the separately measured left and right loads or resistances where the respective left and right load or resistances are added to the respective left and right load cells or sensors at the subject portions of the exercise machine. For example, the user interface may provide a video game that has an avatar representing the user (e.g., the patient in rehabilitation). The avatar may move in the video game and those moves may correlate with the moves of the patient. As the one or more measurements increase, the movement of the avatar may increase (e.g., if the video game is a car racing video game, as the patient increases the force exerted on the pedals, the speed of the avatar, in its car, will increase). Similarly, the control system may receive one or more measurements associated with speed, repetitions, balance, any other suitable measurement, or combination thereof. Such measurements can be used to move the avatar. The measurements can be received from sensors coupled to the exercise machine. For example, sensors can be coupled to the pedals of the exercise machine or to a base of the exercise machine.
In some implementations, initially, the control system may determine measurements in accordance with an exercise plan associated with each exercise of the video game. For example, there may be a first level of the video game that applies a first resistance to the pedals of the exercise machine (e.g., the cycle machine) and a second level of the video game that applies a second resistance to the pedals. Further, the control system may receive measurements associated with each exercise as a patient is using the exercise machine. The control system may generate a target threshold in accordance with an exercise plan associated with each exercise of the video game. For example, there may be a first threshold associated with the first level and a second threshold associated with the second level. The exercise may be complete when the one or more measurements are received and the one or more measurements exceed one or more target thresholds. For example, if the patient is playing the first level of the video game and one or more measurements exceed a first target threshold, the first level may end and the control system will select the level two for the patient to play. In some implementations, the control system may determine an average measurement by accumulating raw measurements over a certain period of time (e.g., 5 seconds) and averaging the raw measurements to smooth the data (e.g., eliminates jumps or spikes in data) in an average measurement.
The control system may compare the one or more measurements (e.g., raw measurements, or averaged measurements) to one or more target thresholds. In some implementations, a single measurement may be compared to a single specific target threshold (e.g., a one-to-one relationship). In some implementations, a single measurement may be compared to more than one specific target threshold (e.g., a one-to-many relationship). In some implementations, more than one measurement may be compared to a single specific target threshold (e.g., a many-to-one relationship). In some implementations, more than one measurement may be compared to more than one specific target threshold (e.g., a many-to-many relationship).
The target thresholds may be an osteogenesis target threshold, a muscular strength target threshold, a balance threshold, a speed threshold, a range of motion threshold, a repetition threshold, any other suitable threshold, or combination thereof. In addition to the threshold explanations described above, the balance target threshold, the speed threshold, and/or the range of motion threshold may be determined based on a rehabilitation protocol pertaining to the user, an age of the user, a gender of the user, a sex of the user, a height of the user, a weight of the user, a bone density of the user, an injury of the user, a type of surgery of the user, a type of bone fracture of the user, etc. A rehabilitation protocol may refer to any illness, disease, fracture, surgery, or ailment experienced by the user and any treatment instructions provided by a caretaker for recovery and/or healing. The rehabilitation protocol may also include a condition of health where the goal is avoid a problem. Any of the target thresholds may be determined based on a historical performance of the user using the exercise machine (e.g., amount of pounds lifted for a particular exercise, amount of force applied associated with each body part, the range of motion for pedaling, the level of exertion, the level of pain, etc.) and/or other exercise machines, a fitness level (e.g., how active the user is) of the user, a diet of the user, a protocol for determining a muscular strength target, a range of motion target, etc.
The control system may determine whether the one or more measurements exceed the one or more target thresholds. Responsive to determining that the one or more measurements exceed the one or more target thresholds, the control system may cause a user interface to present an indication that the one or more target thresholds have been met or exceeded and an exercise is complete. For example, the user has completed a level of the video game. Additionally, when the one or more target thresholds are met or exceeded, the control system may cause the user interface to present an indication that instructs the user to apply additional force (less than a safety limit) to attempt to set a personal maximum record or achievement (e.g., of a rate of speed, of a level of stability, a number of repetitions, of an amount of weight lifted, pressed, pulled, or otherwise exerted force) for that exercise. The control system may also determine that one or more target thresholds (e.g., a level of pain or an exersion level) are met or exceeded and end the exercise game being played. The control system may present the same game at an easier exercise game level or present a different game for the user to engage in different exercises to reduce the level of pain. In this way, the user can continue exercising rather than stopping the rehabilitation session due to pain. The video game may have one or more games, each of which have one or more exercises that target one or more muscles groups at one or more different levels of intensity.
Further, the user interface may present an indication when a measurement is approaching a target threshold for the user. In another example, when the measurement meets or exceeds the target threshold, the user interface may present an indication that the target threshold has been met or exceeded, respectively, and that the exercise is complete. The control system may provide visual and/or audio encouragement and/or coaching to the user during a video game. For example, as the user is nearing the target threshold, the control system may provide an audio of a human voice encouraging the user to maintain or increase speed on the cycling machine to earn an achievement or reach the end of the exercise game level. The control system may indicate if there are any remaining incomplete exercise game levels the video game as part of the exercise plan, that there is another game or another level (e.g., with a difference exercise and/or goal) to be completed by the user. If there are no remaining games or levels (i.e., exercises in the exercise plan) to complete, then the user interface may present an indication that all exercises in the exercise plan are complete and the user can rest. In addition, when the exercise plan is complete, the control system may generate a performance report that presents various information (e.g., charts and graphs of the right and left measurements received during each of the exercises, left and right maximum loads for the user received during each of the exercises, historical right and left measurements received in the past, comparison of the current right and left measurements with the historical right and left measurement, an amount of pounds lifted or pressed that is determined based on the measurements for each of the exercises, percent gained in measurements over time, achievements earned, goals reached, exercise game levels completed, rankings as compared to a video game history of playing, etc.).
Further, the one or more measurements may each be compared to a safety limit. For example, a left measurement and a right measurement may each be compared to the safety limit for the user. The safety limit may be determined for the user based on the user's disease protocol. There may be different safety limits for different portions of the user's body on the left and the right side, one extremity versus another extremity, a top portion of the user's body and a body portion of the user's body, etc., and for different exercises. For example, if someone underwent left knee surgery, the safety limit for a user for a left measurement for a cycling using a left leg may be different from the safety limit for a right measurement for that exercise and user. If the safety limit is exceeded, an indication may be presented on the user interface to instruct to reduce the amount of force or speed that the user is applying and/or to instruct the user to stop applying force because the safety limit has been exceeded.
Another benefit of the present disclosure is its ability to speed the healing of fractures in athletically robust individuals. Further, another benefit is the increase in muscle mass by using the exercise machine to trigger muscular hypertrophy. The control system may provide an automated interface that improves compliance with an exercise plan by using a real-time feedback loop to measure loads added during each of the exercises, (e.g. resistance applied to the pedals) compare the measurements to target thresholds and/or safety limits that are uniquely determined for the user using the exercise machine, and provide various indications based on the comparison. For example, the indications pertain to when the user should add more load, when the target thresholds are met or exceeded, when the safety limit is met or exceeded, when the exercise is complete, when the user should begin another game, when the user should begin another level of the exercise game, and so forth.
Rehabilitation Exercises and their Benefits
The following exercises achieve rehabilitation results by exposing relevant parts of a user to exercises that build strength, increase flexibility, increase range of motion, increase balance, increase coordination, decrease pain, decrease the amount of time required for recovery, or any combination thereof. In addition to the exercises machines or devices described above in this disclosure, exercise machines or devices used to facilitate the rehabilitation exercises referred to are as follows.
Cycling Machine
A cycling machine refers to a stationary bicycle used as exercise equipment and/or rehabilitation equipment. The cycling machine includes pedals configured to rotate. The cycling machine may include attached handlebars or may be used in combination with detached handlebars. The cycling machine may include an attached seat or may be used in combination with a detached seat. The cycling machine can be used to for exercise targeted to improve the following key muscle groups: gluteals, hamstrings, quadriceps, thighs, adductors, abs, and grip muscles as well as to increase flexibility, range of motion, and strength.
Balance Equipment
Balance equipment refers to an exercise machine or device, such as a balance board or a rocker device, for a user to stand on and maintain balance and control as the balance board moves in various directions. The balance board can be used to for exercise targeted to can improve mobility, flexibility, proprioception, and strength in the following key muscle groups: peroneals, gluteals, hamstrings, quadriceps, thighs, adductors, abs, and grip muscles as well as to increase flexibility, range of motion, and core strength.
Exercise machines can include moving parts to provide dynamic exercises to facilitate rehabilitation. A dynamic exercise can be, but is not limited to an exercise where a user participates in an activity where the user moves and some resistance or load may be provided against the movement of the user. The
Implementations of a first housing 1914, generally indicated, can be coupled to the base 1902. The first housing 1914 can be disposed adjacent to the rear side 1906. A handlebar including one or more handles 1916 can be coupled to the first housing 1914. The handles 1916 can include grip pads to prevent slipping during use of the exercise machine 1900.
The exercise machine 1900 comprises a multidimensional exercise control system. The control system comprises a user interface 1918. The user interface can be coupled to the first housing 1914. The user interface 1918 may be or function as the user interface 18 in
Implementations of a second housing 1920, generally indicated, can be coupled to the base 1902. The second housing 1920 can be disposed between the front and rear sides 1904, 1906. The second housing 1920 can be disposed adjacent to and/or coupled to the first housing 1914. In the present implementation of the second housing 1920, and as illustrated in the drawings, the second housing 1920 is cylindrical shaped. However, the base 1902 could be of any shape.
A wheel 1926 can be operatively coupled to the exercise machine 1900. In certain implementations, the exercise machine 1900 can have the wheel 1926 coupled to the base 1902. The wheel 1926 can be a single wheel 1926, and the wheel 1926 may be a flywheel. In certain implementations, the exercise machine 1900 can have a pair of wheels, and the wheels may be flywheels. The wheel 1926 can be disposed in the second housing 1920, and the wheel 1926 can be independently rotatable about an axis. The wheel 1926 can be disposed in in a cavity of the second housing 1920. The wheel 1926 can be partially disposed in an opening of the second housing 1920. One of skill in the art will appreciate that the wheel 1926 may be coupled to the base 1902 by various means known in the art. As one example, a support beam can extend from the base 1902 to a first axial, where an axial extends along the axis. In this implementation, the wheel 1926 can be coupled to and independently rotatable about the axial.
In some implementations, pair of pedals (e.g., a right pedal 1922 and a left pedal 1924) can be coupled to and extend from the wheel 1926. The pedals 1922, 1924 can be configured to be engaged by the user, and the pedals 1922, 1924 can facilitate rotation of the respective wheel 1926. The pedals 1922, 1924 can be movably coupled to the wheel 1926. More specifically, the pedals 1922, 1924 can be adjusted radially by the user to various positions to accommodate the needs of the user. During use of the exercise machine 1900, the user can sit in a seat 1930 and engage the pedals 1922, 1924. The seat 1930 may be detached from the exercise machine 1900. In some implementations, the seat 1930 may be attached to the exercise machine 1900. It should be readily appreciated that the user may adjust the seat 1930 and/or the pedals 1922, 1924 to a desired position to accommodate the needs of the user for exercise or rehabilitation. When the user engages the pedals 1922, 1924, the user may apply a force to respective pedals 1922, 1924 to engage and cause rotation of a respective wheel 1926. By engaging respective pedals 1922, 1924 and applying a force to the same, the user, to support osteogenesis and/or increase a range of motion of a user's legs, engages various muscles to push the respective pedals 1922, 1924. The pedals 1922, 1924 may have straps or engagements for a user to engage with and pull the pedals 1922, 1924. Pulling the pedals 1922, 1924 may aid in the strength and rehabilitation of additional muscles. A sensor 1934 can be coupled to the right pedal 1922. An additional sensor 1936 can be coupled to the left pedal 1924. As described above, the sensors 1934, 1936 can be configured to collect sensor data correlating to the respective pedals 1922, 1924. The sensors 1934, 1936 can be a Bluetooth sensor, a load sensor, accelerometers, gyroscopes, magnetometers, any other suitable sensor, or combination thereof.
To further support osteogenesis during use of the exercise machine 1900 by a user, the exercise machine 1900 can include a first resistance mechanism (not shown). The resistance mechanism can be coupled to the base 1902, and the resistance mechanism can be disposed in the second housing 1920 adjacent to the wheel 1926. When the pedal 1922, 1924 are engaged by the user, the resistance mechanism can be configured to resist rotation of the wheel 1926. The resistance mechanisms may resist rotation of the wheel 1926 by any means known in the art.
It is to be appreciated that the exercise machine 1900 could comprise a motor coupled to each of the wheel 1926 and each motor is configured to affect or regulate the independent rotation of a respective wheel 1926. Moreover, the motor 1928 affects or regulates the independent rotation of the wheel 1926 by engaging the wheel 1926 and selectively causing or resisting rotation of the wheel 1926. The motor 1928 can engage the wheel 1926 by any means known in the art. In one example, the motor 1928 could engage gears to cause rotation of the wheel 1926. It is to be appreciated that the motor 1928 can operate congruently with or independently of the resistance mechanisms to affect or regulate the rotation of the wheel 1926. In certain implementations, the motor 1928 can cause rotation of the wheel 1926, and the motor 1928 can resist rotation of the wheel 1926. In other implementations with the motor 1928 and the resistance mechanism, the motor 1928 can rotate the wheel 1926 and the resistance mechanism can resist or stop rotation of the wheel 1926 when the motor 1928 stops rotating the wheel 1926. For regulating or affecting the rotation of the wheel 1926, the present disclosure allows for many variations and combinations of the motor 1928 and the resistance mechanism.
During use of the exercise machine 1900 by a user, when the user applies a force to the pedals 1922, 1924, the control system can maintain a constant rotational velocity between each of the wheel 1926. Alternatively, the wheel 1926 can be mechanically interconnected. For example, the wheel 1926 could be mechanically interconnected by a chain, belt, gear system, or any other means to maintain a constant rotational velocity between the wheel 1926.
In a further implementations of the exercise machine 1900, a control system can be coupled to an actuator, and the control system can be configured to control the actuator. Moreover, the control system can be configured to independently vary the resistance to each of the wheel 1926 to maintain a select rotational velocity thereof, and to independently stop rotation of the wheel 1926. More specifically, the control system can control the actuator to activate the resistance mechanism to independently vary the resistance of the wheel 1926. In certain implementations, the control system can be coupled to the motor 1928, and the control system can be configured to control the motor 1928. Additionally, the control system can be configured to independently maintain select rotational velocities of the wheel 1926, and to independently stop rotation of the wheel 1926. More specifically, the control system can control the motor 1928 to independently maintain select rotational velocities of the wheel 1926 by rotating, resisting, or stopping rotation of the wheel 1926. It is to be appreciated that the control system may control the actuator and/or the motor 1928 simultaneously or independently to maintain the select rotational velocities of the wheel 1926. For communicating the rotational velocities or accelerations of the wheel 1926 to the control system, the control system may also include sensors located on the user or coupled to the wheel 1926. With the rotational velocities or accelerations received from the sensors, the control system can determine, with a processor of the control system, a select rotational velocity of the wheel 1926. The control system can then control the motor 1928 and/or the actuator to maintain the select rotational velocities of the wheel 1926.
In some implementations of the exercise machine 1900, a switch, not illustrated, can be disposed on the first housing 1914 for activating the control system. In another implementations, a button, not illustrated, may be disposed on the first housing 1914 for activating the control system. In yet another implementation, a display 1932 of a user interface 1918, such as a computer screen, iPad, or like device, can be coupled to the exercise machine 1900 to activate the control system. The switch, display 1932, and/or button may be coupled to the exercise machine 1900 by alternative or other means. For example, the switch, display 1932, and/or button could be coupled to the handle 1916. It is further to be appreciated that alternative means could be used to activate the control system and the use of the switch, display 1932, or the button, is not meant to be limiting.
In another implementations, one or more biometric sensors, not shown, may be coupled to the exercise machine 1900 for activating the control system. The biometric sensor could be for, inter alia, detection, recognition, validation and/or analysis of data relating to: facial attributes; a fingerprint, hand, eye (iris), or voice signature; DNA; and/or handwriting. In yet another implementation, the biometric sensor can comprise position sensors located on the user. In addition, it is contemplated that advancements of such biometric sensors may result in alternative sensors that could be incorporated in the exercise machine 1900, i.e., biometric type sensors not currently on the market may be utilized. Further, the one or more biometric sensors may comprise a biometric system, which may be standalone or integrated.
In one implementation, adjustment of exercise based on artificial intelligence, exercise plan, and user feedback is disclosed. An exercise plan may include one or more exercise sessions. For example, an exercise plan may include a schedule of a certain number of exercises sessions for a certain time period (e.g., 3 exercise sessions each week for 4 weeks) that, if performed by the user, should result in a desired outcome (e.g., rehabilitation of a body part, strengthen a muscle, etc.). The exercise session may include one or more exercises for various sections (e.g., warm up, strength, flexibility, cycling, cool down, etc.) The exercise plan may be generated using artificial intelligence via one or more trained machine learning models as described herein. The exercise plan may include a plan of one or more exercise sessions including exercises for a patient for rehabilitating a body part. The exercise plan may include exercises for one or more muscle groups. The exercise plan may be generated by artificial intelligence and/or prescribed by a doctor, a physical therapist, or any other qualified clinician.
For example, a machine learning model may be trained to select one or more exercises for an exercise session based on various inputs. The inputs may include the pain level of the user, the range of motion of the user, and/or attributes of the user. These inputs may be used to determine an exercise level of the user. The machine learning model may receive the exercise level as input and select corresponding exercises from a data structure by matching the exercise level of the user to exercises having a tagged corresponding user exercise level. Various other techniques may be used to select the exercises for the exercise session.
The machine learning models may be trained to control a virtual coach executing on a computing device associated with the exercise machine 100. The virtual coach may speak via a speaker of the computing device, may be a virtual avatar displayed on the user interface 18 of the computing device 12, may cause one or more messages, emails, text, notifications, prompts, etc. to be presented on the user interface 18. The virtual coach may perform actions based on various information, such as progress of the user performing an exercise, the exercise plan details, user feedback, and the like. For example, the virtual coach may provide encouragement to the user based on the progress of the user during an exercise. The virtual coach may provide incentives, rewards, and/or certificates to the user as the user completes exercises. The virtual coach may have a particular persona that is selected for a particular user. For example, some users may respond better and perform exercises completely in response to a nice and encouraging persona for the virtual coach, while other users may respond better to a more demanding and strict (e.g., drill sergeant) persona.
By tailoring the exercise plan for the specific user and dynamically adjusting it using artificial intelligence, compliance with the exercise plan may be enhanced. Further, the user may achieve their desired goal faster by using the generated exercise plan because it is based on their progress and feedback (e.g., pain level, exercise difficulty level). By achieving the desired outcome faster, computing resources (e.g., processing, memory, network, etc.) may be reduced because the exercise machine 100, the computing device 12, and/or the cloud-based computing system 16 may not have to continuously update the exercise plan.
Further, the virtual coach may provide a companion type of feel for the user, which may further cause the user to comply with the exercise plan more efficiently and completely, thereby achieving their desired outcome faster. The virtual coach may improve the user experience of using the computing device 12 and/or the exercise machine 100 because the persona may be selected specifically for the particular user. In some instances, the user may form a bond with the persona of the virtual coach if the persona matches a friend in real life, family member, a significant other, or the like, and the bond may cause the user to feel a desire to want to listen to the virtual coach and/or complete the exercise plan such that they don't let the virtual coach down. Such a situation may also save computing resources because the exercise plan may not have to be adjusted and lengthened by adding additional exercise sessions.
As a result, various technical benefits may be achieved by the disclosed implementations, as described above. Further, the user experience of using the exercise machine 100, the computing device 12, or both may be improved based on the disclosed techniques due to exercising with the virtual coach, the incentives, the rewards, the certificates, and the like.
The processing device may be configured to execute the instructions to receive user input data. As illustrated in
At block 2702, the processing device may receive a set of inputs. The set of inputs may include an indication of a level of pain of the user, a range of motion of a body part of the user, a set of attributes of the user, or some combination thereof. The indication of the level of pain of the user may be entered by the user using any suitable peripheral device (e.g., microphone, keyboard, touchscreen, mouse, etc.) at a particular user interface 18 displayed on a computing device 12. The range of motion of the body part of the user may be determined by the processing device by the user performing a baseline exercise for a certain amount of time. The baseline exercise may include setting a pedal at an initial position and having the user cycle at that position for a period of time. If the user does not experience any pain after the period of time, the pedal may be moved to a second position and the user may cycle for the period of time again. If the user experiences pain, then the range of motion of the user may be determined based on the previous position of the pedal when the user was able to cycle without pain. As may be appreciated, if the user does not experience pain, the position of the pedal may continue to change until the user experiences pain and the ROM of the user may be determined based on the prior position where the user did not experience pain. The attributes of the user may include an age of the user, a height of the user, a weight of the user, a gender of the user, a condition that caused the pain in the body part, one or more procedures perform on the user, a goal of the user, whether the user is in a pre-procedure stage or a post-procedure stage, or some combination thereof. The attributes may be included in a user profile for the user that is stored at the cloud-based computing system 16, the computing device 12, the computing device 15, or both.
At block 2704, the processing device may determine, based on the set of inputs, an exercise level of the user. The exercise levels may range from 1-5, where 1 is the lowest exercise level and 5 is the most advanced exercise level. Any suitable range of exercise levels may be used. The following chart illustrates an example of how the exercise level may be determined:
At block 2706, the processing device may generate, using the machine learning model 60, an exercise session for the user by selecting, based on the exercise level of the user, one or more exercises to be performed by the user using the exercise machine 100. In some implementations, a data structure may include entries for a set of exercises (e.g., tens, hundreds, thousands, etc.) that are each tagged with an exercise level. For example, the processing device may tag each exercise of the set of exercises with a respective user exercise level. The machine learning model 60 may access the data structure to select the exercises for the exercise session by filtering the set of exercises, as further discussed with reference to
At block 2708, the processing device may cause initiation of the exercise session on the exercise machine 100 and a virtual coach executed by the computing device 12 associated with the exercise machine 100 to provide instructions pertaining to the exercise session. The virtual coach may be driven by artificial intelligence via one or more trained machine learning models 60. For example, the trained machine learning models may receive various inputs, such as the exercise session for the user, the exercise being performed, instructions pertaining to the exercise being performed, completion of the exercise being performed, progress of the exercise being performed, and may be trained to provide certain outputs based on the inputs. The virtual coach may output audible noise (e.g., speech) that pertain to the various inputs. For example, the virtual coach may say, via a speaker of the computing device 12, encouraging words while a user is performing an exercise, congratulatory words when the user completes an exercise, instructions when the user is about to start another exercise, and the like. The virtual coach may have a persona (e.g., a cheerleader type of persona, a drill sergeant type of persona) that is selected based on progress of the user, feedback of the user, or both, as described further below with reference to
In some implementations, the processing device may receive, from the user while the user is performing an exercise of the one or more exercises in the exercise session, feedback pertaining to the exercise. The feedback may include an indication that the exercise is too easy or too hard. For example, the user may use a display screen or microphone of the computing device 12 to enter or say the exercise is “too easy” or “too hard.” Responsive to receiving the feedback, the processing device may cause an intensity of the exercise to increase or decrease. For example, if the user says “too easy” the intensity of the exercise may be increased. If the user says “too hard,” the intensity of the exercise may be decreased. Other dimensions, parameters, attributes, etc. of the exercise or exercise session may be changed based on whether the feedback is too easy or too hard. For example, the other dimensions, parameters, attributes, etc. may include a number of sets, a number of repetitions, a hold time, a rest time, and the like. When one of the dimensions, parameters, attributes, etc. changes, the virtual coach may provide an indication of the change. For example, the virtual coach may say, via a speaker of the computing device 12, “The intensity for this exercise has increased.”
In some implementations, the processing device may track how many times the user has provided the feedback for a particular exercise in an exercise session or across every exercise session in an exercise plan for the user. Responsive to determining the feedback has been received more than a threshold number of times (e.g., 3, 4, 5, etc.), the processing device may control, in real-time or near real-time, the exercise machine 100 to initiate a more advanced exercise than the exercise currently being performed, a less advanced exercise than the exercise currently being performed, or the like. Further, the processing device may remove the exercise for which the feedback was received more than the threshold number of times from subsequent exercise sessions and replace it with another exercise. The processing device may cause the virtual coach to provide an indication via the computing device 12 (e.g., voice emitted through the speaker, graphic on the user interface 18, text on the user interface 18, or the like) of the change to the exercise.
In some implementations, the processing device may monitor the progress of the user while the user uses the exercise machine to perform the one or more exercises. The progress may include an amount of time the user performs the one or more exercises, the range of motion of the user while the user performs the one or more exercises, the level of pain of the user while the user performs the one or more exercises, whether the user completes the one or more exercises, an indication of the user of a level of difficulty of the one or more exercises, or some combination. The progress may be determined based on measurement data received from any sensor associated with the exercise machine 100, any user feedback received by the computing device 12, and the like. The user may use any suitable peripheral to input the level of difficulty (e.g., too hard or too easy) while the user performs the exercises. The processing device may adjust, by executing the machine learning model 60, a subsequent exercise session based on the progress of the user. The adjusting may be based on advancing the exercise level of the user to a next exercise level, achieving a desired goal as defined by the user, a medical professional, or both, or some combination thereof.
In some implementations, the processing device may monitor progress of the user while the user uses the exercise machine 100 to perform the one or more exercises. The processing device may cause, based on the progress of the user, an incentive, reward, or both to be elicited by the computing device 12 associated with the exercise machine 100. The incentive, reward, or both may include an animation, video, audio, haptic feedback, image, push notification, email, text, or some combination thereof. The processing device may cause the virtual coach to perform an encouraging action (e.g., shoot virtual fireworks on the user interface 18, cause an avatar displayed on the user interface 18 to dance or give a virtual high five, emit an audible noise from the speaker congratulating the user). Providing incentives, rewards, or both may encourage the user to continue to perform exercises and comply with the exercise session, which in turn, may decrease the amount of time it takes for the user to achieve their goal. Reducing the amount of time it takes for the user to achieve their goal may include technical benefits because if the user achieves their goal faster, the computing device 12, exercise machine 100, and/or the cloud-based computing system 16 may save computing resources (e.g., processing, memory, network) by not having to execute as long to guide the use through the exercise plan. That is, if the user does not comply with the exercise plan efficiently or as directed, then the exercise plan may be adjusted to add additional exercise sessions, thereby causing the computing device 12, exercise machine 100, and/or the cloud-based computing system 16 execute longer and waste computing resources until the user achieves their goal.
In some implementations, the processing device may determine when a number of incentives, rewards, or both elicited by the computing device 12 satisfy a threshold value (e.g., 3, 4, 5). Responsive to determining that threshold value is satisfied, the processing device may cause a certificate to be transmitted to the computing device 12 and associated with an account of the user using the exercise machine 100. For example, the certificate may be stored in a digital wallet of the user's account in the application 17 executing on the computing device 12. In some implementations, the certificate may have a particular value that may be exchanged for certain items (e.g., gift certificate, clothing, coupons, discounts, etc.).
In some implementations, the processing device may determine, by executing the machine learning model 60, a set of audio segments for the virtual coach to say while the user performs the one or more exercises. The audio segments may be based on a state of the exercise (e.g., beginning, middle, end), progress of the user performing the exercise, or any suitable information. For example, at the initiation of the exercise, the audio segment may provide instructions to the user on the details of the exercise (e.g., 2 reps, 30 seconds, etc.). Based on the progress of the user, the audio segment may say “pedal faster” if the user is not pedaling fast enough, “good job” if the user is satisfying the criteria for the exercise, “almost finished” if the user is almost finished with the exercise, or the like. The audio segments may be dynamically determined, in real-time or near real-time, by the machine learning model 60 based on the inputs described above. It should be noted that real-time or near real-time may refer to a relatively short amount of time (e.g., less than 5 seconds) after an action occurring.
In some implementations, the processing device may determine, by executing the machine learning model 60, a schedule of a set of exercise sessions to be performed by the user to achieve a desired goal specified by the user, a medical professional (e.g., physical therapist), or both. The machine learning model 60 may be trained to determine the schedule based on various inputs, such as the desired goal (e.g., full recovery, near full recovery at a fastest pace possible, strength improvement, flexibility improvement, etc.), a procedure performed on the user, attributes of the user (e.g., age, weight, height, etc.), a daily schedule of the user (e.g., job schedule, parenting schedule, school schedule, etc.), and the like. The schedule may be optimized for the user and may comply with the various inputs described above.
In some implementations, the virtual coach may be controlled, in real-time or near real-time, by the machine learning model 60. For example, the virtual coach may provide indications (e.g., emit audible noises, present various screens or notifications or indications or avatars or graphics, etc.) via the computing device 12 as parameters of the exercise, exercise session, exercise machine 100, etc. change, or as attributes or progress of the user changes.
The exercise Sitting Knee Extension has been tagged as a suitable exercise for levels 1, 2, and 3, and is an option for the section comprising Warm Up. It should be note that each exercise session may include various sections: warm up, cardio, strength, cycle, cool down, flexibility, etc. Each section of an exercise session may be assigned one or more exercises that are appropriate for that section, based on the entry in the data structure 2750, and the exercise level of the user that matches the level in the data structure 2750.
At block 2802, the processing device may filter a set of exercises to obtain the one or more exercises for a particular exercise session in an exercise plan. Block 2802 may include blocks 2804, 2806, 2808, 2810, and/or 2812.
At block 2804, the processing device may identify, based on the tagging of the exercises in the data structure, a subset of exercises having the respective user exercise level that matches the exercise level of the user. At block 2806, the processing device may identify a first subset of exercises having a respective section of a set of sections, wherein the set of sections include warm-up, cycling, strength, flexibility, or some combination thereof. At block 2808, the processing device may identify a second subset of exercises that result in a desired outcome specified by a medical professional, wherein the desired outcome pertains to increasing a range of motion, mobility, strength, flexibility, or some combination thereof. At block 2810, the processing device may identify, using a historical performance of the user, a third subset of exercises that have been performed by the user less than a threshold number of times. At block 2812, the processing device may identify, based on feedback from the user, a fourth subset of exercises that have been performed by the user and indicated as being too easy or too hard for the user.
In some implementations, the processing device may select at least one of the subset of exercises, the first subset of exercises, the second subset of exercises, the third subset of exercises, or the fourth subset of exercises as the one or more exercises for the exercise session. That is, any combination of the subset, the first subset, the second subset, the third subset, and the further subset of exercises may be selected as the one or more exercises for the exercise session.
At block 2902, the processing device may receive, from the user while the user is performing an exercise of the one or more exercises, feedback pertaining to the exercise, wherein the feedback includes an indication of a level of difficulty of the exercise. For example, the feedback may be entered by the user using any suitable peripheral (e.g., microphone, touchscreen, mouse, keyboard, etc.) of the computing device 12. The feedback may include the user saying the exercise is too easy or too hard.
At block 2904, the processing device may determine whether the feedback has been received more than a threshold number of times for the exercise. At 2906, responsive to determining the feedback has been received more than the threshold number of times for the exercise, the processing device may adjust, in real-time or near real-time, the exercise session. In some implementations, adjusting the exercise session may include changing to another exercise, controlling the exercise machine to stop the exercise, removing the exercise from the exercise session, changing an intensity of the exercise, or some combination thereof. At block 2908, the processing device may cause the virtual coach to provide an indication of the adjustment. The indication may be provided via the user interface 18, a speaker of the computing device 12, or the like.
At block 3002, the processing device may select, for the virtual coach, a persona from a plurality of personas. The virtual coach may be implemented in computer instructions stored in a memory device and executable by a processing device. The virtual coach may include a particular voice (e.g., male, female) and have a particular persona. The persona may be randomly selected at first and the user's response to the persona may be tracked over time. The response may include whether the user performs the exercises completely or incompletely as the virtual coach guides the user through the exercises. The personas may range from a cheerleader type that provides a lot of encouragement to a drill sergeant type that is more aggressive, harsher, stricter, and/or demands compliance with the exercise or demands the user tries harder.
At block 3004, the processing device may cause the virtual coach to provide instructions as the user performs the one or more exercises. The instructions may be provided visually on the user interface 18, audibly via a speaker of the computing device 12, or both.
At block 3006, the processing device may monitor a parameter associated with the user while the user performs the one or more exercises. The parameter may include a vital sign (e.g., heartrate, blood pressure), sensor measurement data (e.g., ROM, pressure exerted on pedals, etc.), attributes of the user (e.g., respiratory rate, temperature, perspiration, etc.). The monitoring may be based on any suitable sensor measurement data associated with the user, the exercise machine 100, or both. In some implementations, the parameter pertains to a progress of the user, an indication of whether the user likes the persona of the virtual coach, or both. For example, the user may provide feedback that they like the persona of the virtual coach via the user interface 18 or by speaking to the computing device 12 via a microphone.
At block 3008, the processing device may select, based on the parameter, a subsequent persona for the virtual coach. For example, if the user indicated the user does not like the persona, the processing device may select a different persona for a subsequent exercise and/or exercise session.
At block 3010, the processing device may switch, in real-time or near real-time, based on the parameter, to a different persona for the virtual coach while the user performs the one or more exercises. Dynamically switching may be based on whether the user is performing the exercise well or not. For example, if the user is pedaling at substantially slower rate than desired for the exercise, the processing device may determine the user is not responding well to the persona and may switch to a different persona immediately during the exercise. The progress of the user may be tracked to see if the switch of personas impacts the progress of the user. Further, if the user indicates the user does not like the persona, the processing may switch to a different persona immediately while the user performs the exercise.
Further, the user interface 1918 may present one or more visual representations 3106 of target load thresholds tailored for the user. For example, the one or more target thresholds may include a left target threshold, a right target threshold, or some combination thereof. Presenting the visual representations 3106 of the target thresholds concurrently with the real-time display of the measurements in the visual representations 3102 and/or 3104 may enable the user to determine how close they are to exceeding the target thresholds and/or when they exceed the target thresholds.
The computer system 3200 includes a processing device 3202, a main memory 3204 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM)), a static memory 3206 (e.g., flash memory, static random access memory (SRAM)), and a data storage device 3208, which communicate with each other via a bus 3210.
Processing device 3202 represents one or more general-purpose processing devices such as a microprocessor, central processing unit, or the like. More particularly, the processing device 3202 may be a complex instruction set computing (CISC) microprocessor, reduced instruction set computing (RISC) microprocessor, very long instruction word (VLIW) microprocessor, or a processor implementing other instruction sets or processors implementing a combination of instruction sets. The processing device 3202 may also be one or more special-purpose processing devices such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), network processor, or the like. The processing device 3202 is configured to execute instructions for performing any of the operations and steps discussed herein.
The computer system 3200 may further include a network interface device 3212. The computer system 3200 also may include a video display 3214 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), one or more input devices 3216 (e.g., a keyboard and/or a mouse), and one or more speakers 3218 (e.g., a speaker). In one illustrative example, the video display 3214 and the input device(s) 3216 may be combined into a single component or device (e.g., an LCD touch screen).
The data storage device 3208 may include a computer-readable storage medium 3220 on which the instructions 3222 (e.g., implementing the application 17 or 21 executed by any device and/or component depicted in the FIGURES and described herein) embodying any one or more of the methodologies or functions described herein are stored. The instructions 3222 may also reside, completely or at least partially, within the main memory 3204 and/or within the processing device 3202 during execution thereof by the computer system 3200. As such, the main memory 3204 and the processing device 3202 also constitute computer-readable media. The instructions 3222 may further be transmitted or received over a network via the network interface device 3212.
While the computer-readable storage medium 3220 is shown in the illustrative examples to be a single medium, the term “computer-readable storage medium” should be taken to include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) that store the one or more sets of instructions. The term “computer-readable storage medium” shall also be taken to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by the machine and that cause the machine to perform any one or more of the methodologies of the present disclosure. The term “computer-readable storage medium” shall accordingly be taken to include, but not be limited to, solid-state memories, optical media, and magnetic media.
In some implementations, the cloud-based computing system 16 and/or the computing device 12 may connect to and/or use an application programming interface (API) exposed by a third-party entity, such as an electronic medical records (EMR) system, and/or a social network system. The API may be used by the application 17 to extract information pertaining to EMRs of the user, if proper authorization is given and authorization of a user account is completed. The EMR information may automatically populate in the appropriate fields in the user profile. For example, the medical procedures identified in the EMR information may be populated. In some implementations, the format of the data obtained by the API may be in a different format than the format the application 17 uses. In such an instance, the application 17 may transform the data's format into an acceptable format (e.g., extensible markup language (XML)) for the application 17. In some implementations, the application 17 may use the API to access the user's information on a social media or social network system (e.g., Facebook®, Twitter®, Instagram®, etc.) to obtain information publicly available on the social network system.
The received input of the physical activity level and the pain level may be used by the one or more machine learning models 60 to generate an improved exercise plan. For example, the machine learning model may determine, using a data source including various associations, including, for example, the levels of attainment associated with achieving the physical activity level, where the levels of attainment may include range of motion, strength, endurance, balance, intelligence, neurological responsiveness, emotional well-being, and mobility. Further, the machine learning model 60 may determine which body portions to target for the various levels of attainment, and which exercises to select to include in the exercise plan that target the appropriate body portions. In some implementations, the pain level reported by the user may be used to select exercises, difficulty levels of the exercises, and the like.
In some implementations, upon the user's selecting the physical activity goal and the pain level, an onboarding protocol that uses a baseline fitness test may be initiated. For example,
Selection of the graphical element 3504 may cause the machine learning model 60 to select a next exercise that is more difficult. The onboarding protocol may include exercises having tiered difficulty levels and may select for subsequent exercises for the user to perform, wherein the subsequent exercises advance in difficulty until the user has either completed all of the exercises or reached a point where the user can no longer perform the exercise because it is too difficult or painful. The machine learning model 60 may determine a fitness level for the user based on a completion state (e.g., a degree of completion, a percentage of completion, a value of completion, etc.) of a last exercise performed by the user. The machine learning model 60 may select a difficulty level for each exercise in the improved exercise plan by associating the difficulty level for each exercise with the fitness level of the user.
In some implementations, a multimedia segment (e.g., recording or feed) may be presented in a digital media player 3506. The multimedia segment may include video and/or audio of a coaching character providing instructions and guidance on how to perform the first exercise. Various options may be provided by the digital media player that enable the user to play, pause, or stop the multimedia segment. There may be options to enable the user to fast forward or rewind the multimedia segment, as well.
As further depicted, each exercise includes an energy consumption metric (“50”). The energy consumption metric may vary for each exercise and it may provide a target metric for the user to achieve during each exercise. The energy consumption metric may be based on a combination of various types of information and metrics associated therewith, such as a metabolic indicator associated with performing the exercise, fitness results of the user, and/or a user-reported pain level of the user, among other information. The energy consumption metric may be determined for the user while they perform the exercise, and when the target energy consumption metric has been exceeded, the user may be done with the exercise. The application 17 may track the user's progress over time if and when the user exceeds or meets the target energy consumption metric. Each determined energy consumption metric for each exercise may be summed, mathematically analyzed, processed or manipulated to determine an energy score associated with an amount of energy it will take to achieve the physical activity goal. If the summed, mathematically analyzed, processed or manipulated energy consumption metrics equal or exceed the energy score, then the user may have enough energy to achieve the physical activity goal. As may be appreciated, the user may exceed or match the energy score faster or slower than predicted based on a number of factors, such as their performance, their drive, their health (e.g., physical and mental), their compliance with the exercise plan, and the like.
At block 4202, the processing device may generate, by the artificial intelligence engine 65, an exercise program including an exercise plan, wherein the exercise plan includes a set of exercises. Using training data, one or more machine learning models 60 may be trained to generate an exercise program including the exercise plan. For example, the trained machine learning model 60 may generate the exercise program by selecting a set of exercises to include in the exercise plan. The set of exercises may be selected to, inter alia, encourage compliance, improve user enjoyment of the exercise program, and/or ameliorate boredom. In some embodiments, the set of exercises may be selected based on different types of exercises (e.g., stretching, cycling, weightlifting, balancing, etc.) that focus on different parts of a user's body or improvements thereto, different movements, and the like. Each respective one of the set of exercises may be associated with one or more user energy consumption metrics, wherein the energy consumption metrics are based at least on a metabolic equivalent of task (MET) value. Each determined energy consumption metric for each exercise may be summed, or otherwise mathematically analyzed, processed or manipulated, to determine an energy score associated with an amount of energy it will take to achieve completing the exercise program.
User energy scores may be determined based on an amount of energy it takes to attempt to achieve (including achieving) a physical activity goal of the user. A user energy score may be associated with exercise programs (also referred to as classes herein), and the user energy score associated with the exercise program may indicate an amount of energy performing the exercise program consumes. Bone growth, muscle growth, rehabilitation, prehabilitation, and the like may be preferred, desirable, or necessary to perform certain physical activities. For example, a person may require a certain amount of muscle mass to move an object having a particular weight. The physical activity may be desirable, but some people may lack the appropriate bone mass, muscle mass, or physical ability in general to perform the physical activity. In one example, a grandparent may desire to play with their grandchildren, and may want to select that physical activity as a goal.
The processing device may generate the user energy consumption metrics based at least on MET values and/or user fitness test results. Each of the user energy consumption metrics is generated for a respective one of the set of exercises based on the MET value for the respective one of the set of exercises. Some MET values may be associated with a single exercise included in the set of exercises. For example, a first MET value may be associated with a first exercise and a second MET value may be associated with a second exercise. Some MET values may be associated with multiple exercises included in the plurality of exercises. For example, a third MET value may be associated with both a third exercise and a fourth exercise. In some implementations, the user energy consumption metrics may be generated based on one or more user-reported pain levels, an indication of a pain level the user is in, heartrate, step count, blood pressure, perspiration, blood oxygen levels, body temperature, or some combination thereof. In some implementations, the artificial intelligence engine 65 may generate one or more machine learning models 60 trained to generate the user energy consumption metrics.
Each determined energy consumption metric for each exercise may be summed, or otherwise mathematically analyzed, processed or manipulated, to determine an energy score associated with an amount of energy it will take to achieve the physical activity goal. If the summed energy consumption metrics equal or exceed the energy score, then the user may have enough energy to achieve the physical activity goal. As may be appreciated, based on a number of factors, such as their performance, their drive, their health (e.g., physical and mental), their compliance with the exercise plan, and the like, the user may exceed or match the energy score faster or slower than predicted.
At block 4204, the processing device may receive data pertaining to a set of users. The set of users may be grouped based on the users selecting a similar physical activity goal (e.g., Bayesian approach), based on a vote, based one or more measurements, based on a coach selecting the users, etc. The data may include physical activity goals the set of users desires to achieve. The physical activity goals may be selected from one or more user interfaces of computing devices associated with the set of users. The data pertaining to the set of users may include at least user fitness test results. The user fitness test results may indicate, for example, strength, mobility, endurance, pliability, a range of motion, flexibility, balance, or a combination thereof. In some embodiments, the data pertaining to the set of users may include at least one or more user-reported pain levels. The user fitness test results and/or the user-reported pain levels may be determined, for example, during an onboarding protocol, such as the one described above in relation to
Using various performance measurements from one or more sensors, attributes of users of the exercise apparatuses 100, user-reported difficulty levels of exercises, user-reported pain levels, and the like, the user energy consumption metrics may be objectively monitored and/or measured. While the set of users perform the exercise plan of the exercise program, the user energy consumption metrics may be monitored and/or measured. The user energy consumption metrics may be presented on computing devices associated with the users.
At 4206, the processing device may determine one or more second user energy scores for the physical activity goals. The one or more second user energy scores may be determined by a machine learning model 60 of the artificial intelligence engine 65. The second user energy scores may be the same, within a threshold deviation of each other, or outside the threshold deviation.
At 4208, based on the first and second user energy scores, the processing device may assign, by the artificial intelligence engine 65, at least a subset of the set of users to the exercise program. For example, the set of users may be grouped in a cohort based on how closely or to what extent the first energy score of the exercise program matches the second energy score of the desired physical activity goal.
If the second user energy scores are the same or within a designated range of each other, an exercise program having a matching or similar user energy score may be selected and assigned to the set of users. If the second user energy scores are within the threshold deviation, an exercise program having a matching or similar user energy score may be selected and assigned to the set of users. If the second user energy scores are outside the threshold deviation, different exercise programs may be selected and assigned to the user or users.
The processing device may transmit the exercise plan to one or more computing devices. The one or more computing devices may be associated with a coach and/or one or more users. The physical activity goal may include any activity relating to physical motion of a portion of the user's body. For example, the physical activity goal may include activities aimed at ameliorating knee pain, traversing stairs, gardening, performing yardwork, playing, walking, running, meditating, learning faster, improving concentration, improving focus, shortening response time to stimuli, improving relationships, improving libido, changing a state of mind, improving cardiovascular performance, improving heart rate, reducing blood pressure, sitting without pain, standing without pain, feeling energized, performing more advanced exercises, performing a greater quantity exercises, carrying groceries, performing house chores, losing weight, or some combination thereof. The physical activity goal may require one or more physical levels of attainment to achieve. As used herein, levels of attainment may refer to range of motion, strength, endurance, balance, intelligence, neurological responsiveness, emotional well-being, mobility, pliability or some combination thereof. In some implementations, the user may use a user interface, including one or more graphical elements, to select the physical activity goal via the computing device 12.
In some embodiments, the processing device may generate one or more machine learning models 60 trained to generate the exercise program and to assign the subset of the set of users to the exercise program. For example, the machine learning models 60 may be trained to match a user energy score associated with an exercise program and a user energy score associated with a physical activity goal selected by a user. The machine learning models 60 may select the subset of users who selected physical activity goals having a similar energy score, group them in a cohort, and assign them to the same exercise program. The user energy score associated with the physical activity goals and the user energy score associated with the same exercise program may match. In some embodiments, the machine learning models 60 may dynamically generate the exercise programs as a result of the user selecting a physical activity goal associated with a user energy consumption metric. The machine learning model 60 may select exercises to include in an exercise plan, such that a user energy consumption metric associated with the exercise plan matches the user energy consumption metric associated with the physical activity goal.
In some embodiments, the processing device may transmit a signal to a set of exercise apparatuses 100. When the signal is transmitted, the set of users may be performing at least one of the subset of the set of exercises included in the exercise plan on the exercise apparatuses 100. In response to the exercise apparatuses 100 receiving the signal, at least one portion of the exercise apparatuses 100 may be adjusted. The adjustment may be based on at least one operating parameter specified in the exercise plan.
At block 4302, the processing device may receive data pertaining to a user who is a member of the set of users. The data may include user fitness test results, user-reported pain levels, or both. The user fitness test results may indicate, for example, strength, mobility, endurance, pliability, a range of motion, flexibility, balance, or a combination thereof. In some embodiments, the data pertaining to the set of users may include at least one or more user-reported pain levels.
At block 4304, the processing device may generate, based on the data and the MET value, a second user energy consumption metric for at least one of the set of exercises. The generation of the second user energy consumption metric may also be based on heartrate, step count, blood pressure, perspiration, blood oxygen levels, progress of the exercise plan, weight height, range of motion measurement relating to any body part, body temperature, or some combination thereof.
At block 4306, based on the second user energy consumption metric for the at least one of the set of exercises, the processing device may generate an updated exercise plan for the user. The updated exercise plan for the user may include at least one modified exercise. The modification may include a type of the exercise (e.g., cycling, rowing, running, etc.) or a characteristic of the current exercise (e.g., a duration, a force, a speed, an interval, a frequency, a periodicity, an amount of weight, etc.). The modification to the exercise plan may include a modification to an exercise performed on or off the exercise machine 100. The processing device may transmit the updated exercise plan to the computing device.
The exercise program 4402 may be generated by the machine learning model 60. In some embodiments, the exercise program 4402 may be generated by the machine learning model 60 based on a target user energy score to be achieved by performing the exercise program 4402 and/or a user energy score associated with a physical activity goal 4404 selected by the user. As depicted, the user energy score (S1) of the physical activity goal (“Gardening”) and the user energy score (S1) of the exercise program (“Exercise Program 1”) match. Accordingly, the users 4406 “John Doe,” “Jane Smith,” and “Cyndi Thomas,” who selected gardening, may be grouped into a cohort and assigned to “Exercise Program 1.” “Exercise Program 1” may be associated with various other information aside from the exercises and the user energy consumption metrics, such as a physical location where the exercise program takes place, a date and time of the exercise program, a duration of the exercise program, a virtual location where a user can attend remotely, etc. The user interface 18 also presents a coach 4408 “Tyler Morris,” who will coach the users 4406 throughout the exercise program 4402.
Once the exercise program 4402 initiates, the user interface 18 may transition to a live feed of the coach 4408 or a virtual avatar of the coach 4408 instructing the user to perform the Exercise 1. As the user performs the exercises, the user's data may be obtained from sensors and/or self-reported pain levels and the user energy consumption metrics for each exercise may dynamically change, wherein such change may be displayed on the computing device 12. A display screen may be presented in the physical location in which the exercise machines 100 are located. The display screen may present information pertaining to the users 4406 in real-time to encourage the users to compete with each other (e.g., determine who can reach the user energy score the fastest). In some embodiments, to enable privacy, the user's computing device 12 may just present information pertaining to the specific user and not any other user participating in the exercise program.
As depicted, the users selected “Gardening” as “Physical Activity Goal 1”. Based on the types of movements, muscles, body parts, etc. involved, the disclosed embodiments may determine gardening is associated with a user energy score of S1. The machine learning model 60 may be trained to generate “Exercise Program 1,” which has a user energy score of S1. The user energy score may be a summation of the user energy consumption metrics of the exercises included in the exercise program. For example, “Exercise 1” is associated with a user energy consumption metric of 1.10, “Exercise 2” is associated with a user energy consumption metric of 1.20, “Exercise 3” is associated with a user energy consumption metric of 1.15, and “Exercise 4” is associated with a user energy consumption metric of 1.55.
The exercise program 4502 may be generated by the machine learning model 60. In some embodiments, the exercise program 4502 may be generated by the machine learning model 60 based on a target user energy score to be achieved by performing the exercise program 4502 and/or a user energy score associated with a physical activity goal 4504 selected by the user. As depicted, the user energy score (S2) of the physical activity goal (“Riding a bicycle”) and the user energy score (S2) of the exercise program (“Exercise Program 2”) match. Accordingly, the users 4506 “Jim Brown,” and “Peter Albreck,” each of whom selected “Riding a bicycle,” may be grouped into a cohort and assigned to “Exercise Program 2.” “Exercise Program 2” may be associated with various other information aside from the exercises and the user energy consumption metrics, such as a physical location where the exercise program takes place, a date and time of the exercise program, a duration of the exercise program, a virtual location where a user can attend remotely, etc. The user interface 18 also presents a coach 4508 “Betty Miller,” who will coach the users 4506 throughout the exercise program 4502.
Once the exercise program 4502 initiates, the user interface 18 may transition to a live feed of the coach 4508 or a virtual avatar of the coach 4508 instructing the user to perform the Exercise 1. As the user performs the exercises, their data may be obtained from sensors and/or self-reported pain levels and the user energy consumption metrics for each exercise may dynamically change on the computing device 12. A display screen may be presented in the physical location in which the exercise machines 100 are located. The display screen may present information pertaining to the users 4506 in real-time to encourage the users to compete with each other (e.g., determine who can reach the user energy score the fastest). In some embodiments, to enable privacy, the user's computing device 12 may just present information pertaining to the specific user and not any other user participating in the exercise program.
As depicted, the users selected “Riding a bicycle” as “Physical Activity Goal 2”. Based on the types of movements, muscles, body parts, etc. involved, the disclosed embodiments may determine riding a bicycle is associated with a user energy score of S2. The machine learning model 60 may be trained to generate “Exercise Program 2,” which has a user energy score of S2. The user energy score may be a summation, or otherwise mathematically analyzed, processed or manipulated, of the user energy consumption metrics of the exercises included in the exercise program. For example, “Exercise 1” is associated with a user energy consumption metric of 1.10, “Exercise 2” is associated with a user energy consumption metric of 1.55, and “Exercise 3” is associated with a user energy consumption metric of 1.15.
The various aspects, embodiments, implementations or features of the described embodiments can be used separately or in any combination. The embodiments disclosed herein are modular in nature and can be used in conjunction with or coupled to other embodiments, including both statically-based and dynamically-based equipment. In addition, the embodiments disclosed herein can employ selected equipment such that they can identify individual users and auto-calibrate threshold multiple-of-body-weight targets, as well as other individualized parameters, for individual users.
Consistent with the above disclosure, the examples of systems and methods enumerated in the following clauses are specifically contemplated and are intended as a non-limiting set of examples.
Clause 1. A method for generating, by an artificial intelligence engine, an exercise program comprising a first user energy score, wherein the method comprises:
generating, by the artificial intelligence engine, the exercise program comprising an exercise plan including a plurality of exercises, wherein each respective one of the plurality of exercises is associated with one or more user energy consumption metrics based at least on a metabolic equivalent of task (MET) value, and based on the one or more user energy consumption metrics, the first user energy score is associated with the exercise program;
receiving data pertaining to a plurality of users, wherein the data includes physical activity goals the plurality of users desires to achieve;
determining second user energy scores for the physical activity goals; and
based on the first and second user energy scores, assigning, by the artificial intelligence engine, at least a subset of the plurality of users to the exercise program.
Clause 2. The method of any clause herein, further comprising transmitting the exercise plan to computing devices.
Clause 3. The method of any clause herein, wherein the physical activity goals comprise one or more levels of attainment.
Clause 4. The method of any clause herein, wherein the one or more levels of attainment comprise at least one of a range of motion, strength, endurance, balance, pliability, proprioception, cardiovascular health, intelligence, neurological responsiveness, health measurement criteria, performance measurement of physical health, emotional well-being, and mobility.
Clause 5. The method of any clause herein, further comprising:
Clause 6. The method of any clause herein, further comprising:
Clause 7. The method of any clause herein, wherein the second user energy consumption metric is further generated based on at least one selected from the group consisting of heartrate, step count, blood pressure, perspiration, blood oxygen levels, progress of the exercise plan, weight, height, range of motion measurement relating to any body part, and body temperature.
Clause 8. The method of any clause herein, further comprising generating one or more machine learning models trained to perform the generating of the exercise program and to perform the assigning of the subset of the plurality of users to the exercise program.
Clause 9. The method of any clause herein, further comprising:
transmitting a signal to a plurality of exercise apparatuses, wherein the plurality of users performs on the exercise apparatuses at least one of the subset of the plurality of exercises included in the exercise plan; and
in response to the exercise apparatuses receiving the signal, adjusting, based on at least one operating parameter specified in the exercise plan, at least one portion of the exercise apparatuses.
Clause 10. A system for generating, by an artificial intelligence engine, an exercise program comprising a first user energy score, wherein the system comprises:
a memory device storing instructions; and
a processing device communicatively coupled to the memory device, wherein the processing device is configured to execute the processing device to:
generate, by the artificial intelligence engine, an exercise program comprising an exercise plan including a plurality of exercises, wherein each respective one of the plurality of exercises is associated with one or more user energy consumption metrics based at least on a metabolic equivalent of task (MET) value, and based on the one or more user energy consumption metrics, the first user energy score is associated with the exercise program;
receive data pertaining to a plurality of users, wherein the data includes physical activity goals the plurality of users desires to achieve;
determine second user energy scores for the physical activity goals; and
based on the first and second user energy scores, assign, by the artificial intelligence engine, at least a subset of the plurality of users to the exercise program.
Clause 11. The system of any clause herein, wherein the processing device is further to transmit the exercise plan to computing devices.
Clause 12. The system of any clause herein, wherein the physical activity goals comprise one or more levels of attainment.
Clause 13. The system of any clause herein, wherein the one or more levels of attainment comprise at least one of a range of motion, strength, endurance, balance, pliability, proprioception, cardiovascular health, intelligence, neurological responsiveness, health measurement criteria, performance measurement of physical health, emotional well-being, and mobility.
Clause 14. The system of any clause herein, wherein the processing device is further to:
Clause 15. The system of any clause herein, wherein the processing device is further to:
Clause 16. The system of any clause herein, wherein the second user energy consumption metric is further generated based on at least one selected from the group consisting of heartrate, step count, blood pressure, perspiration, blood oxygen levels, progress of the exercise plan, weight, height, range of motion measurement relating to any body part, and body temperature.
Clause 17. The system of any clause herein, wherein the processing device is further to generate one or more machine learning models trained to perform the generating of the exercise program and to perform the assigning of the subset of the plurality of users to the exercise program.
Clause 18. The system of any clause herein, wherein the processing device is further to:
transmit a signal to a plurality of exercise apparatuses, wherein the plurality of users performs on the exercise apparatuses at least one of the subset of the plurality of exercises included in the exercise plan; and
in response to the exercise apparatuses receiving the signal, adjust, based on at least one operating parameter specified in the exercise plan, at least one portion of the exercise apparatuses.
Clause 19. A tangible, non-transitory computer-readable medium storing instructions that, when executed, cause a processing device to:
generate, by an artificial intelligence engine, an exercise program comprising an exercise plan including a plurality of exercises, wherein each respective one of the plurality of exercises is associated with one or more user energy consumption metrics based at least on a metabolic equivalent of task (MET) value, and based on the one or more user energy consumption metrics, the first user energy score is associated with the exercise program;
receive data pertaining to a plurality of users, wherein the data includes physical activity goals the plurality of users desires to achieve;
determine second user energy scores for the physical activity goals; and
based on the first and second user energy scores, assign, by the artificial intelligence engine, at least a subset of the plurality of users to the exercise program.
Clause 20. The non-transitory computer-readable medium of any clause herein, wherein the physical activity goals comprise one or more levels of attainment, and wherein the one or more levels of attainment comprise at least one of a range of motion, strength, endurance, balance, pliability, proprioception, cardiovascular health, intelligence, neurological responsiveness, health measurement criteria, performance measurement of physical health, emotional well-being, and mobility.
No part of the description in this application should be read as implying that any particular element, step, or function is an essential element that must be included in the claim scope. The scope of patented subject matter is defined only by the claims. Moreover, none of the claims is intended to invoke 25 U.S.C. § 104(f) unless the exact words “means for” are followed by a participle.
The foregoing description, for purposes of explanation, use specific nomenclature to provide a thorough understanding of the described embodiments. However, it should be apparent to one skilled in the art that the specific details are not required to practice the described embodiments. Thus, the foregoing descriptions of specific embodiments are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the described embodiments to the precise forms disclosed. It should be apparent to one of ordinary skill in the art that many modifications and variations are possible in view of the above teachings.
The information below may provide a guideline of the rules employed in delivering an effective, well-structured exercise program for rehab, conditioning, and/or long-term fitness adapted to the capabilities of each user. The information below is for explanatory purposes and the subject matter of the present disclosure is not limited to the examples provided below.
2. Determining a User's Exercise Level
Exercise levels may range from 1 to 5. Placing a user in an exercise level may be a function of measuring the individual's Range-of-Motion (ROM) and establishing the Degree-of-Knee-Pain they are experiencing.
How the Combination of Pain and ROM Test Results Defines Levels:
How the Combination of Level and ROM Test Results Define Resistance:
3. Exercise Sessions
4. Exercise Adaptations
5. In-Session Exercise Switching
6. Counting Reps & Sets
7. Completing an Exercise or an Exercise Session
8. Advancing Out of Levels
9. Exercise Session Example for Level-1
Additional Algorithm Rules
Each exercise may target at least one body part. The body part targets may be tagged in the data structures as follows:
Each session may include at least one exercise that targets each of the body parts listed.
The exercises per session must not exceed a percentage (e.g., 5, 10, 20, 30, 40, 50, etc.) of one particular body part target.
The number of exercises assigned to each body part may be tabulated. The selection of an exercise for any body part, should not result in total exercises for that body part exceeding the total of any other body part by more than one.
Resistance (Cycling)
Bands
The above discussion is meant to be illustrative of the principles and various embodiments of the present invention. Once the above disclosure is fully appreciated, numerous variations and modifications will become apparent to those skilled in the art. It is intended that the following claims be interpreted to embrace all such variations and modifications.
This application claims priority to U.S. Provisional Patent Application Ser. No. 63/211,193, filed on Jun. 16, 2021, the entire disclosure of which is incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| 63211193 | Jun 2021 | US |
