Patent Application
20220346652
This application claims priority to U.S. Provisional Application 62/742,209, filed Oct. 5, 2018, entitled “Flexible Iontronic Sensing Wearable Detects Pedal Pulses and Muscular Actives,” which is specifically incorporated by reference herein.
Wearable devices provide an intimate physical and data interface between the human body and mobile computing devices. As medical technology has developed, several body signals have been monitored for data such as gestures, footsteps, general activity levels, and other events have been measured by body sensors and converted into electronic signals that are translated into physiologic data and displayed to a user. Additionally, cardiovascular events such as pulse, blood volume variations, blood oxygen level, electrocardiograph (ECG), electroencephalograph (EEG), and respiration have been successfully measured and analyzed through the wearable sensor interfaces.
One challenge of the wearable device is to identify an ideal body location to receive a relatively long-term or repeated attachment. Researchers have tested a number of sites on the human body (e.g. head, chest, upper arm, wrist, waist, fingers, and feet) to attempt to locate an ideal site that is both useful for monitoring data and for which sensors can be readily attached without interfering with normal day-to-day activities. For example, the chest has been used to acquire ECG, heart rate and respiration signals with a chest band-based heart monitor. Finger rings have been used as a simple healthcare monitor to detect pulse, blood oxygenation, and blood volume variations. Wrist-based sensors incorporated into a wristwatch type device have likely been the most investigated site due to the traditional historic practice of wearing wrist watches in the ability to acquire cardiac signals including pulse and blood volume variations from arteries that are readily detected in the wrist.
Despite these advances, a number of limitations are still present. Poor long-term wearability is the major challenge that prevents many wearable technologies from being widely adopted because the sensor arrays that are available to measure large amounts of data, and to measure them accurately, are bulky and simply uncomfortable to wear. The sensors and the materials holding the sensors in place can be rigid, restrictive, irritating and unable to secure a long-term attachment. As a result, an ideal match between body location and long-term comfort is still being sought.
Human feet contain abundant vasculature, strong tendons and muscles, and characteristic bone structures , from which valuable health information and muscular activities can be continuously collected and precisely extracted. However, although human feet have been considered as an ideal location to place wearable sensor devices, the existing footwear technologies focus on activity tracking and measurement of mechanical contact forces, such as steps and weight bearing, together with some pulse and artery monitoring in clinical environments, and often requiring measurement at other locations on the body. Specifically, microchip-enabled accelerometers have been used in shoes or mounted as shoe clips to record steps during exercises, or detect kinematic changes of gait cycles. These devices and method have been used in the ports footwear industry to develop smart athletic shoes (e.g., Nike+®, Adidas Micropacer®, and Under Armour First Run®.
In terms of the mechanical contact force measurements, the ground reaction force (GRF) can be detected by the flexible pressure sensors embedded in the insole and has been used to map the plantar pressure distribution, as well as to conduct gait analysis. Recently, researchers group illustrated a method to apply ballistocardiogram (BCG) for heart rate detection by incorporating piezoelectric films into the insole, although the BCG signals were rather weak to be detected until the participants did a special stepping training to enhance their cardiac outputs See Chen M. Liu, F. Jiang, H. Jiang, S. Ye, and H. Chen, “Low-power, noninvasive measurement system for wearable ballistocardiography in sitting and standing positions,” Comput. Ind., vol. 91, pp. 24-32, Oct. 2017.
Therefore, although attempts have been made to use sensors around the human foot to measure important physiological parameters, limitations on the sensor arrays and the ability to incorporate sensors into wearable footwear have prevented significant advances in sensor arrays incorporated into footwear and used to measure a wide range of physiological parameters.
This invention includes a foot-based wearable system that can detect cardiac signals, cardiac data, physiologic data, motion artifacts, artifacts of muscle and tendon motion, and signals that reflect muscular activities that can be translated into specific gestures. To achieve such functions in a wearable and comfortable item of footwear, a plurality of flexible iontronic sensing (FITS) pressure sensors are specially configured and integrated into a pre-determined array configuration to create a pressure sensor array comprised of a plurality of sensors that can be integrated into traditional footwear or footwear that has been specially configured to receive the sensor array and optionally accompanying circuit components.
Although the human foot presents a number of options for measuring pressure and physiologic data, the present invention includes selection of the configuration of the array together with the specific performance parameters of the individual sensors and the overall performance of the sensor array to be specifically targeted to the region of the foot adjacent to the dorsalis pedis artery. Accordingly, the sensor array of the present invention takes advantage of both the orientation of the array around the dorsal artery of the foot together with high performance features of the individual sensors, including high flexibility, comfort, close conformity to the skin, or to a clothing or fabric layer in close conformity to the skin, high pressure-to-capacitance sensitivity, high signal-to-noise ratio, the ability to compensate for background and noise artifacts, fast response time, and sensing range typical to shoe and cardiac pressure. The sensor array is integrated into a wearable article of footwear and the sensor array assembly is integrated into the footwear such that the sensor array is adjacent to the artery such that multiple physiological parameters are detected.
The sensor array has a plurality of individual sensors that, by integration into the footwear assembly, has the capability to measure multiple and independent cardiac functions, physiological functions, muscle and tendon artifacts, and coordinated foot gestures based on characteristic signals received from the pressure signals at individual sensors of the array. The individual sensors can be separately or collectively selected for signal processing that includes comparing individual sensors to a cardiac data profile that identifies an individual sensor in the array that is receiving a stronger signal correlated to cardiac function compared to other sensors in the array. The profile of the cardiac data contains information to match the sensed data, such as time interval, amplitude of the signals and frequency of the signal and a normal heart rate range from a large population. In this configuration, this particular sensor is identified as a primary signal source for cardiac function and is so designated for a single sensing session.
Once a primary cardiac signal is identified, the known spatial configuration of the individual sensors in the array can be used to identify a secondary sensor, separate from the primary cardiac sensor, to sense any of additional cardiac data, separate physiological signals, including muscle and tendon artifacts and essentially any other differential pressure measurement detectable across the individual or plurality members of the array.
Because of the flexible nature of the sensor assembly and the ability to tailor the configuration of the individual sensors in the array to the area adjacent to the dorsalis pedis artery, the sensor array can readily be integrated into an article of footwear. Footwear includes any variety of sock, shoe, or other garment that conforms closely to the foot such that the array can be maintained in a stable position near the dorsalis pedis artery.
Ideal locations for the sensor array include the inner top surface of a shoe or socket, including specifically the tongue of tissue or any structural elements of issue that takes the position equivalent to the tongue in an ordinary athletic shoe. When worn, the sensor array is integrated into the article of footwear and then is maintained in a stable position during the time that the footwear is worn by the user. This garment must produce some pressure to place the sensor into conformal contact with the foot. In some embodiments the pressure is fixed due to the elasticity of a sock, or fly-knitted shoe cover, is manually adjustable through Velcro, straps, and laces, or is automatically adjustable through motorized laces or self-pumped airbag. Notifications can be given to the user to adjust the tension for optimal detection of physiological signals. Because of individual variations, the positioning of the array might vary from time to time depending on individual variations by the user. The array is capable of sensing these variations and the individual sensors can be re-configured electronically to accommodate such variations while collecting data that can be compared with prior uses of the device and integrated into a data set that compares the data secured from the sensor over a large number of individual sessions by the user.
FITS devices offer advantages in pressure sensing, with high sensitivity, excellent mechanical ruggedness, and reliable flexibility, due to the ultrahigh interfacial capacitance and fast polarization of the iontronic materials. The array disclosed herein is made from a solid-state flexible ionic coating in an elastic contact with a conductive electrode array, which can measure pressures in the device at a sensitivity of up to 1 nF/mmHg with a detection range of 1 to 200 mmHg. Resolution is in the range of 0.01 mmHg to 1 mmHg (with 0.01-0.05 mmHg most preferred), and sensitivity is in the ranges of 0.01 nF/mmHg to 1 nF/mmHg (with 0.02-0.1 nF/mmHg most preferred). Device response time is in the range of 0.1 to 10 milliseconds which provides the ability to detect micro fluctuations in blood pressure, rapid changes in cardiac signals, and to precisely determine the time of cardiac events (e.g. systolic peak, dicrotic notch, inter-beat interval). With such a high sensitivity, small variations of blood pressure can be detected at one or more sensors in the array and compared with a stored cardiac profile like data segments, empirical values or reference values to identify an individual sensor producing a primary cardiac signal from the dorsalis pedis artery, which is also known as the pedal pulse waveform, in a gentle contact with the foot or foot covered with a certain fabric layer around the baseline pressure exerted by the shoe or sock of approximately 10 mmHg. Alternatively, the primary cardiac channel can be determined using signature properties from the standard cardiac profile such as the peak-to-peak intensity, onset time, and regular period, which can be characterized through various mathematics and physiological knowledge commonly understood within the art.
Because of the ultrathin and flexible construct of the sensing array, the range of 50 μm to 2 mm in thickness. The array can be placed in a position adjacent to the dorsalis pedis artery and in incorporated into an article of footwear in contact with the dorsal area of the foot. Because the array is comprised of a plurality of individual sensors, the orientation of the individual sensors in the array preferably has a known spatial configuration such that individual distances and relative positions from sensor-to-sensor are known as described above. Given the known physiology of the foot, and even taking into account individual person-to-person variations, the signal data obtained from an individual sensor or a plurality of sensors can be used to correlate the spatial relationship of one or more sensors to an additional singular sensor, or set of sensors, so that specific cardiac signal data, or physiological artifact data, can be correlated to additional cardiac or physiological function data obtained from separate sensors and having a distance and direction component based on the known spatial configuration of the array. Calibration procedures conducted by the user can compensate for person-to-person differences and improve the accuracy of physiological feature mapping and parameter determination. A calibration procedure can, for example, consist of conducting various foot gestures in a specified sequence, recording signals for the pressure sensor array, correlating those signals to the specific foot gestures, retaining a collection of signals in device memory to compare with future signals, and correlating the calibrated values against signals from a measured user gesture to identify the gesture.
Because the dorsalis pedis artery region of the dorsal area of the foot has a known configuration, the sensor array advantageously places a number of individual sensors, and a predetermined number of sensors within a defined area. Advantageously, there are at least 2 sensors within 3-10 centimeter squared area, and preferably with a 3, 4, or 5 square centimeter area. Furthermore, at least one or more of the individual sensors is always placed at least 200 pm every adjacent sensor. In a preferred embodiment, a first primary sensor designated as the primary cardiac sensor is a distance at least 200 μm, or at least 300 μm 400 μm or 500 μm no more than 3 mm, 4 mm, or 5 mm, and preferably between 200 um to 2 mm from an adjacent sensor. The adjacent sensor obtains a pressure signal generating a measure of a physiological function detected in the human foot and selected from static pressure distribution, and the group of motion or position artifacts correlated to any of tendons, muscles, bones, cartilage or ligaments, or combinations of any or all of the above. Also, a plurality of sensors or individual sensing arrays can be arranged in a honeycomb (or similar) pattern where rows of sensors are offset such gaps between sensors are covered in the next row of sensors, allowing all locations in the transverse direction of the foot to be covered. As described herein, the sensor obtaining a signal from a cardiac function is sometimes described as the “first” or “primary” sensor while the additional sensors obtaining an additional cardiac signal or physiological signals may be deemed the secondary or second sensor the designation is arbitrary in terms of the order in which the signals are received or analyzed and the placement of the individual sensor in the array. The only operational requirement is that at least one single sensor is analyzed and identified as providing a signal from a cardiac function and so designated in the overall operation of the array.
The invention also includes specific modes and methods of analysis based on the individual signals obtained from one or more pressure sensors in the sensor array. A unique feature of the invention includes the individual selection of pressure sensors in the array in a defined order to individually isolate and compare cardiac function or other physiological parameters signals obtained from the array. In a preferred embodiment, the method includes selection of a first sensor producing a primary cardiac signal, followed by selection of a second, separate sensor yielding a secondary cardiac, or other physiological parameter. Each sensor will detect pressures from different sources including: static/baseline pressure between foot garment and foot, inertial forces from movement causing separation or joining of foot garment and foot, cardiac pressure from the blood pulsing through an artery, or pressures exerted through the movement of tendons, muscles, or bones whether due to contraction or rearrangement during foot gesture. Each pressure signal has a unique signature in intensity, space, and time such that they can be isolated to specific sensors (e.g. primary cardiac signal) or events (e.g. foot gesture). Linear algebraic combinations and other signal processing techniques such as Principle Component Analysis (PCA), adaptive noise cancellation, or machine learning algorithms can be used to isolate pressure signals from each other.
For example, the static pressure on a reference sensor signal can be subtracted from the primary cardiac signal to determine the pure cardiac pressure signal. Similarly, common inertial forces can be subtracted to denoise the cardiac pressure signal during movement. Simple analysis of the primary cardiac signal (or other physiological feature) places the location directly at the center of the sensing element limited in spatial resolution to the distance between sensing elements (pitch). Mathematical techniques can allow for locating cardiac signal at a higher precision than the sensing pitch using a combination of the location and intensity of the primary, second, and any other detectable cardiac signals.
In one embodiment of the invention where the system is worn on each foot, the common origin of the cardiac signal produces synchronous responses in both systems. This effect can used to produce motion-resistant pure cardiac signals by detecting signal on the system with the minimum noise or combining signals using aforementioned signal analysis techniques. For example, during standard walking, stance and swing phases alternate. By selecting the system in the swing phase (with one foot airborne) as the primary cardiac signal, the motion artifacts are significantly reduced.
Specifically, after one sensor is selected from the array as yielding a primary cardiac signal based on pulse waveform-and is preferably analyzed against a stored or accessed cardiac profile, critical additional signals measuring additional or repeated cardiovascular parameters are collected from an additional, separate sensor. The designation of a sensor as first, second or separate does not necessarily require that these individual sensors were the first or second sensors in order analyzed for signal or measurement, but rather that these were designated as first or second based on substantive analysis of the signal detected and so designated for further signal processing. Accordingly, in, for example, a ten-sensor array, the 9th signal analyzed in sequence may be the signal that is matched to a cardiac signal and designated as a “first” signal for subsequent analysis and comparison with other signals. The “first” signal is sufficient to determine several cardiac functional parameters, such as upstroke time, augmentation index, blood pressure trend, as well as derived vital signs, such as heart rate (HR), heart rate variability (HRV), and respiratory estimation.
The additional, separate or second sensor is similarly selected based on the measured signal of an additional physiologic parameter, including additional cardiac functional parameters, such as upstroke time and augmentation index, as well as derived vital signs, such as heart rate (HR), heart rate variability (HRV), pulse wave velocity, and pulse transit time, and respiratory estimation. The second, separate signal obtained from any literal sequence or order of analysis, is designated and identified as yielding a secondary or separate signal that can be further processed and compared with both the primary cardiac signal as well as additional tertiary or quaternary signals from the array. As noted above, because the spatial relationship between each of the sensors of the array is predetermined and known relative to one another in advance of the signal detection sequences, all of the measurements from the first, second, or additional pressure sensors can be correlated to specific physiological structures that are unique to each of the individual sensors, the remaining sensors in the array, and the orientation relative and proximal to the dorsalis pedis artery. Separate signals correlated to specific physiological structures, such as tendons, can allow for individual tendon contract/relaxation analysis to determine foot gesture, gait patterns, or other biomechanical parameters of foot status and motion.
The preferred embodied of the invention uses FITS devices for detection of high quality primary cardiac signals. The signal quality allows for detection of aforementioned cardiac functional parameters such as HRV and can provide tolerance for person-to-person variability. The high accuracy measurement of HRV can be used to derive several physiological-related parameters for stress evaluation. For example, frequency analysis parameters such as the integrity of the high frequency band (HF), low frequency band (LF) ultra low frequency band (ULF), and ratio of LF to HF as described in Kim,Hye-Geum et al., “Stress and heart rate variability: a meta-analysis and review of the literature.” Psychiatry Investigation 15.3 (2018): 235. Other flexible pressure sensors may be used detect the primary cardiac signal at lesser quality and used to calculate noise-tolerant derived vital signs such as heart rate, whereas low magnitude signals such those used for augmentation index are challenging without proper signal-to-noise ratio and resolution. Examples of other flexible pressure sensors include: piezoresistive, piezoelectric, capacitive, and pneumatic manometry systems. Combination with non-pressure sensors such as accelerometers, gyroscopes, magnetometers, temperature sensors, and humidity sensors can provide additional data useful in noise reduction or physiological parameters.
In one particular structure and method of the invention, the array of the invention yields a signal pressure set of signals, at least one of which is been compared with a standard 2-lead electrocardiogram (ECG) that has been stored as a reference or is generated in real time from the sensor array of the invention. Furthermore, the linear array with multiple sensing units is adjacent to the dorsalis pedis artery thereby covering the transverse plane on the dorsum, making the capture of pedal pulse signals without any special alignment step, while the location of pedal artery provides a spatial anatomic reference to muscular activities. Finally, muscular responses are collected with a sufficient resolution to track individual tendon activities, providing a highly integrated way for foot gesture classification, see J. Alexander, T. Han, W. Judd, P. Irani, and S. Subramanian, “Putting Your Best Foot Forward: Investigating Real-world Mappings for Foot-based Gestures,” in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, New York, N.Y., USA, 2012, pp. 1229-1238, gait analysis M. W. Whittle, “Clinical gait analysis: A review,” Hum. Mov. Sci., vol. 15, no. 3, pp. 369-387, June 1996, as well as body status tracking.
A highly sensitive and flexible pressure sensing array is enabled by the novel iontronic sensing principle, sensor array and method of sensing as described herein. The array has been fabricated and oriented in combination with support structures to simultaneously acquire body vital signals and track pedal skeletomuscular activities. The unique characteristics of the array allow it to be seamlessly integrated in a footwear format, such as a shoe of any type, a sock or any article of clothing that maintains the sensor array adjacent to the dorsalis pedis artery. The device illustrates that a wearable device can capture high-resolution peripheral arterial pulse waveforms, from which both heart rates and respiratory patterns can be extracted within a medical-standard precision.
The pressure sensor array is electrically connected to a data module that may be placed separately on an article of clothing or incorporated into a support structure that is also wearable on the foot. The data or circuit module contains or is connected to a processor that contains instructions for processing signal data, including shoes and associated articles.
Moreover, the high-spatial resolution of the sensing array allows alignment-free capture of pulse signals as well as provides a spatial reference to the pedal structures. It further enables tracking of individual pedal tendon movements, from which the majority of foot gestures can be assessed in real-time. The device operates as a personal mobile platform to acquire and analyze the human health and activity information in a comfortable and unnoticeable fashion and that is integrated into ordinary articles of clothing without great expense or interference with ordinary functions.
Human feet contain rich vasculature and characteristic bones and muscles that enable a wide range of motion and activities that sustain daily living. The foot also withstands incredible forces and retains strength and flexibility through a wide range of functions such as standing and walking. Because the foot is so intimately involved in a wide range of human activities, it is an ideal location to measure the cardiac and physiological functions that accompany these activities. Two major arterial branches passing towards the foot, i.e., dorsalis pedis artery and posterior tibial artery. The pulsations of both arteries can be directly detected and are known as an indicator of peripheral vascular health. The present invention takes advantage of the ability to locate a sensor array adjacent to the physiologic structures as shown in
A portion of the array component 10 is comprised of a segment of substrate material 5 wherein individual electronic leads 6 are disposed to electrically connect the individual sensors 3 to a connector 7 that facilitates electrical connection to a separate a circuit module (not shown—see
Referring to
The description of the array component 10 as being “adjacent” to the dorsalis pedis artery 25 means that the sensor array component 10 is placed in as area, generally designated at ‘B’ in
Referring specifically to the physiology shown in
Based on the ability of the sensor array component 10 to provide signal data including motion detection from the anatomical structures described above, a novel detection methods have also been discovered based on the combination of the ability to use a wearable item of clothing such as traditional footwear to capture both cardiovascular pressures adjacent dorsalis pedis artery 25 and individual skeleto-muscular responses from the dorsum region of the foot. These methods become enabled once the sensing array component 10 as described herein has been embedded into, for example, the tongue of traditional footwear and positioned in the upper dorsum region proximate to the dorsalis pedis artery 25, as shown in
Referring specifically to
Continuing to refer to
where r and t stand for the radius and thickness of the sensing membrane, respectively; indicates the thickness of the spacer layer; D represents the flexural rigidity of the deforming substrate.
Referring specifically to the operation of the FITS device 40 in
Specifically, 0.5 g PVA (341584, Sigma-Aldrich) was dissolved in 10 g distilled water. Then, this PVA solution was mixed with 0.25 g [EMMI][TCM] (IOLITEC Inc.) and stirred at 50° C. for 2 h. To form a uniform thin ionic coating layer 42, the solution was poured and spin-coated onto the surface of a polyimide film (Kapton, Dupont) at 600 rpm for 30 s, using a commercial spinner (WS-400-6NPP, Laurell). The resulted polyimide film with the ionic coating was baked at 120° C. for 2 h on a hot plate in ambient air. Subsequently, the coated polyimide film was trimmed with the designed layout using a UV laser etching (SAMURAI UV Making System, DPSS Lasers) to complete the fabrication of the top membrane of the individual sensors 3. To form the electrode 43, a 50 nm film of Au was sputter-coated onto another polyimide sheet with a sputtering machine (AUTO108, Cressington), followed by the UV laser etching to form the interdigital single-sided electrodes. Finally, a double-sided tape (467 MP, 3M) was used to create the spacer layer trimmed and applied as the spacer layer 44 to assemble both the top sensing membrane 41 and the assembly of the electrode 43 in the bottom membrane 45 layers together.
To characterize the sensor performance and optimize the sensor dimension for the application of pedal pulses sensing and muscular activities detecting, a custom test setup was constructed using a pneumatic airbag design controlled by a high-resolution manometer (840080, Sper Scientific) to apply a uniform pressure load on the FITS device 40. The capacitive change of a single sensor as assessed in real-time by an LCR meter (4284A, Agilent), while the pressure change was recorded by the manometer. Several sensor evaluation tests were conducted, including a device sensitivity test, a performance evaluation on a curved surface as well as a temperature stability test. Device sensitivity test was to investigate the relationships between the loaded pressure and the generated interfacial capacitance by adjusting different geometric parameters of the FITS device 40 sensors, i.e., radius of the sensing chamber, sensing membrane thickness, and height of the spacer layer. For the bending stability test, the sensor was attached to convex surfaces with various radii of curvatures (flat, 25 mm, 50 mm, 100 mm). The temperature stability test was operated on a hot plate and a thermocouple probe was used to monitor the temperature of the sensor, varying from 15 to 60° C.
To continuously collect and transmit the signal data from the optimized sensing array, a custom circuit was designed and built on a printed circuit board to detect the capacitive changes comprising five-unit iontronic pressure sensing array comprised of an analog front 51, five low-voltage operational amplifiers (LMV324, Texas Instruments), an 8-bit MCU with ADC component 52, a Bluetooth low energy (BLE) module 54 (CC2541, Texas Instruments), and the power management module 53 with a standard rechargeable Lithium-ion battery.
Moreover, a custom graphic user interface (GUI) was programmed in MATLAB to receive, process, and display the signals.
As shown in
The structural curvature has also been investigated on its influence over sensor performance. The measured capacitive values have only marginal changes on the surfaces with a radius of curvature varying from 25 mm to an infinite flat. This result implies that consistent device performances is obtained under different surface topologies and additional calibration steps can be optionally assessed under specified conditions and may be bypassed even in variations from use to use and from person to person and this bypass preserves power and aids in ease of use. Finally, temperature variation was studied across temperature rises from 15° C. to 60° C., at no pressure, pressure of 100 mmHg, and pressure of 200 mmHg and revealed to be less than 5%, indicating substantial immunity of the sensors to environmental temperature fluctuations.
Referring to
The sensor component 10 may also be removable, for example by a single user, and introduced into different items footwear 30 that are worn by the user. The sensor component number 10 and the circuit case 42 may be integrated or completely separable by virtue of the plug 7 (see
Within the article of footwear 30 for example within the tongue 41 and enclosure may be created that has a border to engage the outer portion of the sensor array 4 or sensor component 10 to hold the individual sensors static during a sensing session. In this configuration, the outer border of the sensor component 10 would be tailored to the inner border of the compartment integrated into the article of footwear 30 so that the positioning of the sensors is reproducible. In this configuration, the flexible polymer substrate in which the sensor component 10 is disposed is positioned between two layers of an article of footwear 30 and held in close conforming engagement with the physiology of the foot adjacent to the dorsalis pedis artery 25. Although the flexibility of the sensor system does not require precise positioning of the sensor component ten during any sensing session, the orientation of the sensor component ten within the article of footwear 30 that places the largest number of individual sensors 3 of the sensor array 4 adjacent to the dorsalis pedis artery 25 improves the overall performance of the device.
Referring to
Referring to
Continuing to refer to
As noted above and below, assignment of an individual sensor 3 from within the sensor array 4 for detection of the cardiac signal for analysis 70 as opposed to the motion artifacts for analysis 68 is arbitrary and the respective signals may be obtained from a different individual sensor 3 depending on the user, the particular article of footwear 30, even when the particular article of footwear 30 is used in a different sensing sessions from an individual user. Additionally, the step progression in the analytical flowchart of
Measurement results can also provide the information about respiratory patterns together with cardiac function, as shown in
Referring to
where tx and Px indicate the corresponding time and pressure value of each individual feature point i, respectively, respectively. The high-resolution pulse waveform of
Moreover, a custom HR-detection algorithm has been implemented by using a peak-detection method. Specifically, following the collection of the pedal pulse signals from the device, any signal or collection of signals are fed through the MATLAB algorithm consisting of a band pass filter from 0.1 Hz to 5 Hz and a peak detection module to detect the systolic peaks of the pulse waveform in the time domain, from which the real-time HR can be computed continuously every second. Meanwhile, a standard ECG electrode pair has been placed on the right arm, left arm and left leg of the same testing subject and the cardiac bioelectrical signal has been simultaneously acquired, which is considered as a golden standard for HR detection in clinic practices, to calculate the accuracy of the HR from the real-time pulse waveform from the device of the invention.
Moreover, a Bland-Altman analysis reveals a low mean error of 0.81 bpm between ECG and the output of the invention with a 95% confidence interval from −0.74 BPM to 2.4 BPM. These result indicates that when analyzing measurements from these two devices (ECG vs sensor array) in a static condition, high accuracy can be acquired. This level of accuracy is within the published standard for heart rate measurement, the standard for cardiac monitors, heart rate meters, and alarms (ANSI/AAMI EC13:2002). These measurement results show the potential of the HR measurement from the device to be within a medical-grade precision as confirmed by ECG and pedal pulse waveforms normalized at the same scale and the correlation (r=0.97) between the heart rate calculated from the sensor array 4 of the invention and the heart rate from ECG and based on a Bland-Altman plot of heart rate HR comparison between a value calculated from the sensor array 4 and a value from ECG.
In addition to the vital signal extractions and arterial pulse waveform measurements, wearable tracking of skeletomuscular activities can be implemented using the flexible sensing array, in which individual tendon movements have been investigated. According to human anatomy adjacent to the dorsalis pedis artery 25 as shown in
Referring to
A highly sensitive and flexible pressure sensing array enabled by the novel iontronic sensing principle has been fabricated, to simultaneously acquire body vital signals as well as track pedal skeletomuscular activities, which has been seamlessly integrated in a footwear format such as a shoe. The device demonstrates that a foot wearable device can capture high-resolution peripheral arterial pulse waveforms, from which both heart rates and respiratory patterns can be extracted within a medical-standard precision. Moreover, the high-spatial resolution of the sensing array allows alignment-free capture of pulse signals as well as provides a spatial reference to the pedal structures. It further enables tracking of individual pedal tendon movements, from which the majority of foot gestures can be assessed in real-time. The device also enables a valuable personal mobile platform to acquire and analyze the human health and activity information in a comfortable and unnoticeable fashion.
Individual datasets are assembled from the individual sensors 3 in the sensor array 4 and stored for processing or analyzed in real time against stored sensor data. These datasets include a first signal dataset from a first pressure sensor in the array. As noted above, the designation of an individual sensor as first, second, or third etc. is arbitrary and only refers to individual sensors 3 of the sensor array 4 that are analyzed in a sensing session and assigned a status of first, second, third etc. in the pressure signal processing steps, the individual datasets may include each of discrete cardiac datasets, static datasets, inertial datasets and combinations of each. Each dataset from an individual sensor may be assigned as yielding any of the cardiac, static, or inertial datasets or combinations thereof. Furthermore, the datasets may be correlated to spatial orientation based on the known spacing, orientation, and distances between the individual sensors 3 in the sensor array 4. Based on this orientation, individual pressure signals may be comprised of a component that indicates the spatial relationship to any individual sensor 3 in the sensor array 4 or any combination thereof. The data processing system may contain stored data values for a sample cardiac profile that includes the artifacts described above for comparison with the sensed pressure signal from any component of the sensor array 4 including combinations of individual sensors 3 in the array 4. A cardiac profile may also be created from prior sessions wherein a user creates a cardiac profile by wearing the item of footwear 30 together with the sensor array 4 during a static or sensing shut session as described in
Because of the ability to correlate pressure sensor data with physiological structures based on the known orientation between the dorsalis pedis artery 25 and the major muscles, bones, and tendons of the foot, as shown in
The methods of the invention enable the detection and measurement of a number of cardiac and physiological parameters resulting from pressure sensor signals generated by the array when the sensor component generated by the array is placed proximate to the dorsalis pedis artery. Pressure sensor data is obtained from at least two of the individual sensors in the array based on changes in pressure sensed proximate to the dorsalis pedis artery. In the data processing methodology, a first pressure sensor is identified from amongst the individual sensors in the array and one sensor is identified as the source of a primary cardiac data signal. A second signal is identified as providing additional sensor data that may be supplemental cardiac data or may be a result of sensing pressure differences from physiological changes proximate to the dorsalis pedis artery. These physiological changes result from the motion of muscles, tendons, bone, interstitial tissue, cartilage and other structures that can be translated based on changes in the pressure sensor signals.
The method of the invention includes placing the sensor component into an article of footwear and making a connection to a circuit module that enables detection storage and processing of the sensor data as described above. The data processing includes using the known spatial orientation of the individual sensors in the array to identify particular physiological structures as described in connection with the description of
The methodology of the invention also includes identifying a particular individual sensor as the primary cardiac signal as described above through comparison with a stored cardiac data profile including a stored cardiac data profile from prior uses of the sensor array by the same user. The methodology of the invention also includes determining a specific distance between a pressure sensor identified as the primary cardiac sensor and at least one additional sensor in the array based on a known specific distance between two sensors based on the orientation and spatial relationship of individual sensors within the overall sensor array as fixed in the sensor component. The orientation and spatial relationship of sensors in the array can also be correlated with the physiological profiles of the human foot and individual physiological structures can be assigned as being measured by the second or supplemental pressure signals detected by the second or supplemental sensors in the array following identification of the primary cardiac signal.
As described in connection with
Data processing techniques that are unique to the sensor array of the invention include using the first second, or any sensor as a noise detection function to separate out extraneous pressure sensor input from the primary cardiac and additional sensor inputs. The calculation of any value resulting from the sensory input can be obtained by addition, subtraction, summation, multiplication or other manipulation of the data to yield an output perceived by the user. The output includes a measurement of all of pulse, heart rate variability, blood pressure, foot gesture, respiration and respiration patterns, cardiovascular system parameters such as blood pressure, pulse, pulse flow, including arterial system parameters such as arterial integrity, arterial patency, arterial flexibility, vascular system parameters, detection of heart valve operation, including valve patency and related parameters, and combinations thereof. Also, cardiac abnormalities such as arrhythmia and tachycardia can be detected. Any of the foregoing parameters can also be detected and upper and lower limits established wherein touching either the upper or lower limits generates a separate signal to the user.
In addition to sensing foot gestures, data output from the sensor component can be translated into interpretations of the activity level of the user indicating rest or recline, activities such as sitting, walking, and running and can be used to correlate any activity mode to any of the measured cardiac or physiological mode such that characteristic motion, cardiac, and respiratory patterns can be assigned to different activities by the user such as sitting, walking, and running.
Calibration of the overall system as well as the individual sensor component can be achieved by repeated use by a single user wherein the data from a single sensing session is compared to a stored data profile such that future sensing sessions for an individual user can be correlated according to detection of input from the primary cardiac sensor the secondary physiological sensors and combinations of the above. Calibration sessions can also be performed between individual sessions by the user and correlated to changes in any of the cardiac or physiological data as measured.
Although the disclosed examples have been fully described with reference to the accompanying drawings, it is to be noted that various changes and modifications will become apparent to those skilled in the art. Such changes and modifications are to be understood as being included within the scope of the disclosed examples as defined by the following claims.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2019/027754 | 4/16/2019 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 62742209 | Oct 2018 | US |
