DEVICE AND METHODS FOR DERIVING A RESPIRATION RATE FROM MULTIPLE BIOMETRIC SOURCES

Abstract

A monitoring system for determining a user's respiration rate includes a wearable monitoring device for attachment to the user's chest wall. The wearable monitoring device includes a plurality of biometric sensors for generating a plurality of separate biometric data types such as motion data, sound data, and EKG data. A controller is included within the wearable monitoring device and/or is communicatively coupled to the device. The controller is configured to receive the biometric data from the biometric sensors, associate the biometric data of each of the separate biometric data types with one another according to a common sampling timeline to generate a set of multiplexed biometric data, and use the multiplexed biometric data to determine the user's respiration rate.

Description

BACKGROUND

Determining a patient's respiration rate is key to monitoring the patient's health. There are many conventional techniques used to measure respiration rate. Such techniques include direct methods such as measuring the expansion of the chest cavity, chest impedance measurements, capnography methods, measuring temperature, or measuring the concentration of carbon dioxide in the patient's expired breath.

Measuring patient respiration rate is challenging because breathing is affected by many other patient activities such as talking or moving. Pain and medications can also affect the depth and rate of respirations. There have been many attempts to find one method that always works in all situations, but such attempts have failed.

In addition, health monitoring when the patient is away from a hospital or doctor's office can be difficult. The expensive and often large equipment is simply not available or feasibly usable by the patient while at home or work or while doing other normal activities. Further, even if it were available, the bulkiness and immobility of the equipment would make it impractical to use effectively.

SUMMARY

The present disclosure describes monitoring systems, monitoring devices, and related methods for determining a user's respiration rate. Certain embodiments described herein include a “wearable” monitoring device that provides improved versatility and comfort relative to conventional means of monitoring respiration rate. In addition, certain embodiments described herein obtain multiple types of biometric data to improve the accuracy of the respiration rate readings.

In one embodiment, a monitoring system configured to determine a user's respiration rate includes a wearable monitoring device and a controller communicatively coupled to the wearable monitoring device. The controller may be attached to or housed within the wearable monitoring device and/or may be separated from but communicatively linked to the monitoring device. The wearable monitoring device may be configured for attachment to the user's chest wall.

The wearable monitoring device includes a plurality of biometric sensors for obtaining multiple different types of biometric data. For example, the wearable monitoring device may include a motion sensor for obtaining biometric motion data (e.g., the rise and fall of the chest wall), a microphone for obtaining biometric sound data (e.g., breath and/or heartbeat sounds), and an EKG sensor for obtaining EKG waveform data.

In certain embodiments, the controller is configured to receive the biometric data obtained by the biometric sensors and associate the biometric data of each of the separate biometric data types with one another. The biometric data may be associated (i.e., multiplexed) based on a sampling timeline common to the separate waveforms, for example.

As explained in further detail below, the multiplexed biometric data allows more versatile and/or more accurate determination of the respiration rate. One type of biometric data may provide a better correlation to respiration rate in one situation, while another type of biometric data may be more useful in another situation. For example, the differential weight given to the separate biometric data types may be different depending on whether the user is walking, standing still, lying down, or talking. By combining multiple types of biometric input, overall accuracy is increased across multiple types of user situations and environments.

Additional features and advantages will be set forth in part in the description that follows. It is to be understood that both the foregoing summary and the following detailed description are exemplary and explanatory only, and are not to be read as limiting the disclosure to any particular set of embodiments.

BRIEF DESCRIPTION OF THE DRAWINGS

A more particular description will be rendered by the embodiments illustrated in the appended drawings. It is appreciated that these drawings depict only exemplary embodiments of the disclosure and are therefore not to be considered limiting of its scope. In the accompanying drawings:

FIG. 1 illustrates an exemplary monitoring system including a torso monitoring device and a limb monitoring device;

FIG. 2 illustrates a detailed view of the torso monitoring device;

FIG. 3 schematically illustrates the torso monitoring device showing its biometric sensors, logical components, and other components;

FIGS. 4A through 4C illustrate examples of motion data, obtained using a motion sensor, that may be utilized to determine an estimated respiration rate;

FIG. 5 illustrates example EKG data, obtained using an EKG sensor, that may be utilized to determine an estimated respiration rate;

FIGS. 6A and 6B illustrate an exploded view and a cross-sectional view, respectively, of an exemplary contact microphone including a piezo element that may be utilized to obtain sound data for use in determining an estimated respiration rate;

FIG. 7 schematically illustrates a controller (i.e., computer system) included as part of the monitoring system and that may be communicatively coupled to the monitoring device to receive multiple biometric data types and determine a respiration rate;

FIG. 8 is a flowchart illustrating an exemplary method of determining a respiration rate using multiple biometric data sources; and

FIG. 9 is a flowchart illustrating an exemplary method of using a monitoring system as described herein to determine a respiration rate.

DETAILED DESCRIPTION

FIG. 1 illustrates a wearable monitoring system 100 configured to monitor one or more vital signs while worn by a user. As described in further detail herein, the monitoring system 100 is configured to gather multiple different types of biometric data and use that data to effectively derive a respiration rate reading. The wearable monitoring system 100 includes a torso monitoring device 102 and a limb monitoring device 104. The torso monitoring device 102 is preferably positioned on the user's chest wall as shown. In the context of this disclosure, the limb monitoring device 104 is optional. When utilized, the limb monitoring device 104 is preferably positioned on the user's upper arm as shown.

The limb monitoring device 104 may be used, for example, to monitor blood pressure and/or other vital signs in addition to or in conjunction with the torso monitoring device 102. The limb monitoring device 104 may include one or more position sensors that enable determination of the spatial position of the limb monitoring device 104 relative to the torso monitoring device 102. This spatial position data can be utilized to calibrate blood pressure readings obtained using the limb monitoring device 104.

In FIG. 1, the torso monitoring device 102 is shown placed on the torso wall at the fourth intercostal space in a mid-axillary position. The torso monitoring device 102 is also tilted relative to the transverse plane of the user in order to better receive the electrocardiogram signal from the user. This position also provides an effective baseline level of the height of the heart, which is useful when the torso monitoring device 102 is used in conjunction with the limb monitoring device 104.

While this represents a typical placement of the torso monitoring device 102, it will be understood that placement may vary according to particular application requirements and according to particular user preferences, needs, and/or anatomy. Further, while the components of the monitoring system 100 are shown here as being worn on the user's right side, one or both components may be worn on the left side (e.g., if placement is desired closer to the heart).

FIG. 2 illustrates the torso monitoring device 102 in greater detail. The torso monitoring device 102 includes an interior side 103 and an exterior side 105. The interior side 103 is generally flat and is configured to be placed against the user's skin. Placement may be performed using a medically acceptable adhesive and/or gel pad, for example. Additionally, or alternatively, one or more straps/belts may be attached to the torso monitoring device 102 and sized to allow a user to wrap the one or more straps around the torso to secure the device in position.

As explained in more detail below, the torso monitoring device 102 may contain an integrated battery or plurality of batteries (either disposable or reusable), one or more electrodes for receiving an EKG signal, a thermal isolation component and associated temperature sensor interface, and a microphone interface (e.g., a hole with interface to a piezo pickup for sensing breathing sounds).

The torso monitoring device 102 may include one or more compartments, such as the illustrated medial compartment 107 and side compartments 109, to house internal componentry described in more detail below. While the illustrated design has been found to make efficient use of space and keeps the overall dimensions of the device within manageable, wearable sizes, other embodiments may situate components differently, may use differently sized compartments, and/or may use greater or fewer number of compartments.

The compartments include rounded outer surfaces that beneficially improves the wear ability of the device. For example, the user's clothing is less likely to catch on the rounded surfaces and the rounded surfaces are likely to be more comfortable when the user's arm or another object rubs against it.

The surfaces of the torso monitoring device 102, and in particular the interior side 103, may be in contact with the skin for relatively long durations and thus are preferably compatible with long-term contact with the skin. The device housing may be molded or otherwise formed out of a flexible material, and the inner electronic components are also preferably flexible (e.g., flexible electronics and/or flexible PCB components) for comfort and durability.

The torso monitoring device 102 is sized for effective wearable placement on the user. An overly large or bulky device would be uncomfortable and difficult to use as a true “wearable” device. In particular, the torso monitoring device 102 is configured in size and shape to minimize any significant limitations to the mobility of the user, to be wearable under typical clothing, and to minimize potentially uncomfortable protrusions when worn.

In view of the foregoing benefits of effective sizing, the device 102 may have a length of about 50 mm to 150 mm, or more preferably about 70 mm to about 120 mm. The device 102 may have a width of about 12 mm to about 50 mm, or more preferably about 25 mm to about 40 mm. The device may have a height of about 4 mm to about 25 mm, or more preferably about 6 mm to about 15 mm. For a typical implementation, dimensional features within the foregoing ranges better enable the device to be wearable and readily utilized by the user. However, larger or smaller dimensions may also be utilized according to particular application needs and/or user preferences/requirements.

The illustrated torso monitoring device 102 may also include a tab (not shown) disposed near the edge of the device and/or partially on or within an adhesive element of the device. The tab may extend beyond an end of the of the device 102 and provides a surface for the user to grip while attaching and/or while removing the device. For example, when a user desires to remove the device, the user can grip the tab between his/her fingers to gain sufficient leverage to lift and peel the device 102 off of their skin. As described above, some embodiments may alternatively utilize one or more straps/belts to hold the device 102 in contact against the torso.

FIG. 3 schematically illustrates the biometric sensors and other operational components of the torso monitoring device 102. As shown, the device 102 may include a motion sensor 110, a microphone 112 and an EKG sensor 114. As used herein, a “microphone” can refer to a conventional microphone but can also refer to other sound-sensing devices such as a piezo-based vibration sensor. The torso monitoring device may also include one or more additional biometric sensors, such as a temperature sensor. Although shown schematically here within a single, shared compartment, the operational compartments may be divided among multiple compartments such as by moving one or more components into side compartments shown in FIG. 2.

In one embodiment, the motion sensor 110 is a 9-axis sensor. The 9-axis sensor includes a 3-axis gyroscope, a 3-axis accelerometer, and a 3-axis magnetometer (sometimes referred to as a 3-axis compass). Other embodiments may utilize other motion sensors, such as a 6-axis sensor including a 3-axis gyroscope and a 3-axis accelerometer. The motion sensor 110 tracks chest wall movement when the torso monitoring device 102 is placed on the user's chest.

The microphone 112 is configured to detect biometric sound data related to the user's respiration, including breath sounds and heart-beat sounds. Any suitable microphone device known in the art may be utilized. In a preferred embodiment, the microphone 112 includes a piezo-based vibration sensor. The microphone/sound-sensing device may be coupled to the patient via gel or other acoustic coupling media. The EKG sensor 114 is coupled to leads (not shown) which are arranged on the interior side of the device 102 so as to contact the user's skin when the device 102 is worn. The EKG sensor 114 is configured to detect EKG signals to enable processing of the R to R signals using the Pan Tompkins algorithm and/or other such processes known in the art.

The illustrated monitoring device 102 also includes a logic device 116. The logic device 116 may be any device capable of performing logic functions. For example, the logic device 116 can perform Boolean logic or can produce a pre-determined output based on input. The logic device 116 can include ROM memory, programmable logic device (PLD), programmable array logic (PAL), generic array logic (GAL), complex programmable logic device (CPLD), field programmable gate arrays (FPGA), logic gates, processors, or any other logic function. The logic device 116 may control the functions of the other components of the monitoring device 102. In particular, the logic device 116 can include and/or be in communication with a timer to ensure that the components of the monitoring device 102 perform their desired function at the appropriate time and in the appropriate manner to correctly detect biometric data.

The illustrated monitoring device 102 also includes memory 118. The memory 118 may include any device capable of storing data in computer readable form. The memory 118 may include volatile memory and/or non-volatile memory. The monitoring device 102 also includes a battery 122 to power the various components of the device. The monitoring device also includes a timer 124 for measuring time and coordinating sampling rate of the sensors, for example.

The monitoring device 102 also includes a communications module 120. The communication module 120 enables the monitoring device to communicate with one or more other computing devices, such as an external base station (not shown) and/or healthcare provider's computer system. The communications module 120 may include Bluetooth functionality for exchanging data over short distances (using short-wavelength radio transmissions in the ISM band from 2400-2480 MHz) from fixed and mobile devices, creating personal area networks (PANs) with high levels of security. Other communication means may also be utilized, such as the Internet, cellular RF networks, and/or other wired or wireless networks such as, but not limited to, 802.xx networks.

The external base station may be a user's mobile phone or personal computer, for example and/or a dedicated unit. In embodiments that include a base station, processing may be divided between the monitoring device 102 and the base station to offload at least a portion of the data processing requirements from the monitoring device 102. In some circumstances, raw data may be fed to the base station for substantially all of the data processing. Some embodiments may further relay raw data or partially processed data to a cloud network to which the base station is connected for further processing and/or storage. Offloading processing and/or storage requirements from the monitoring device 102 (by sending to the base station and/or further to the cloud from the base station) can reduce power consumption and prolong battery life in the monitoring device 102.

FIGS. 4A through 4D illustrate an example of biometric motion data obtained using a motion sensor such as motion sensor 110. The biometric motion data may be processed using various signal processing techniques known in the art. In one embodiment, Principal Component Analysis (PCA) is used to decompose the multidimensional motion data into orthogonal components using eigenvalues and eigenvectors derived from the covariance matrix. A bandpass filter may be applied to the orthogonal component data to generate the “resampled” data shown. A Fast Fourier Transform (FFT) may then be applied to the filtered PCA data to determine the primary frequency of the data, as shown in FIG. 4D.

FIG. 5 illustrates an example of biometric electrical data obtained using an EKG sensor such as EKG sensor 114. The measurement of the interval between adjacent R peaks in an EKG is termed the R to R interval. FIG. 5 shows the R to R interval in seconds, showing how the time interval between R primes increases as the user inhales. As shown in this example, 27 breaths were taken during a 5-minute sample period.

FIGS. 6A and 6B illustrate an exploded view and a cross-sectional view, respectively, of an exemplary contact microphone 500 that may be utilized to obtain sound data for use in determining an estimated respiration rate. The contact microphone includes a piezo element 502 embedded within conduction media 504 and is optionally contained within a housing 506. The piezo element 502 is preferably centered within the conduction media 504. The piezo element 502 may be any material with piezo characteristics such as piezo crystals bonded to a metal diaphragm or polyvinylidene difluoride (PVDF) film with contacts.

The piezo element 502 may be connected via electrical contacts (not shown) to a buffer and/or analog-to-digital converter. The buffer may be configured to convert the voltage signal close to the source and to match the electrical impedance of the input to the analog-to-digital converter. The buffer may have analog filtering as needed. The buffer may be integrated and built into the housing or may be a physically separate component.

The conduction media 504 may be a gel or other material configured to match the acoustic impedance of tissue of the body. This allows the microphone 500 to pick up respiration sounds very well and minimizes interference from sounds traveling through the air. The interface 508 of the conduction media 504 is preferably a domed or curved surface that is contacted against the patient's skin to allow conduction of respiration sounds directly to the interface 508 for measurement. As with the motion data, filtering can be used to isolate the proper frequency range and count the breaths in a given time period.

FIG. 7 schematically illustrates a controller 230 included as part of the overall monitoring system and that may be communicatively coupled to the monitoring device 102. The controller 230 may be included internally within the monitoring device 102 (e.g., as part of the logic device 116), may be distributed between an external base station and the monitoring device 102, or may be fully included within the external base station.

The controller 230 operates by receiving multiple types of biometric data and using the biometric data to determine a respiration rate 232. As shown here, biometric motion data 210, biometric sound data 212, and biometric electrical data 214 are received by the controller 230. The various data processing functions described above in FIGS. 4A through 6 may be respectively performed using the motion-based respiration rate module 240, the electrical-based respiration rate module 244, and the sound-based respiration rate module 242. Because the different biometric data types have been obtained across a common sampling timeline, the controller 230 operates to align and associate the multiple biometric data types according to the common timeline and thereby form a multiplexed set of biometric data.

The controller 230 may also include a weighting module 246 and/or a filtering module 248. The weighting module 246 is configured to assign differential weights to the different biometric data types and/or to different time segments of the biometric data types. For example, the statistical weight of one type of biometric data may be reduced (or the data type even completely ignored) where the estimated respiration values of the data type are sufficiently different from the values associated with the other types.

Whether the particular data type is “significantly different” may be determined based on a predetermined threshold relationship, such as whether a respiration rate calculated using just the suspected outlier data type is more than some chosen percentage different than an average of respiration rates calculated using the other data types. This would indicate that the particular outlier data type is a less accurate representation of the user's actual respiration rate.

Because each of the different biometric data types provide different methods for determining respiration rate, they each have their own strengths and limitations. One technique may be superior in one particular situation while another may be superior in a different situation. The most accurate technique may therefore depend on whether or not the user was talking, walking, holding still, or sitting, for example. By combining multiple different types of biometric data into a multiplexed biometric data set, a more versatile data set may be obtained. At least one of the multiple biometric data types is likely to be superior for any given user circumstance, and a more accurate determination of respiration rate may therefore be made.

The filtering module 248 may be used to remove or differentially weigh certain portions of the biometric data corresponding to certain time periods. For example, the controller 230 may detect a disruption event in the biometric data, and the filtering module 248 may operate to ignore or differentially weigh portions of the biometric data that occur during the time of the disruption event.

In one example, a disruption event may be a detected sound event, such as a sound event within the sound data 212 indicative of the user talking. Because readings taken while the user is talking may be less representative of a baseline respiration rate, the time segments corresponding to the user talking may be weighted as less relevant for determining the respiration rate. Such filtering may include removing or differentially weighing just the sound data 212 associated with these identified time segments, or may additionally include removing or differentially weighing the motion data 210 and/or electrical data 214 associated with these identified time segments.

In another example, a disruption event may be related to motion data 210 indicative of the user walking, running, or otherwise moving to a relatively large degree. This may indicate that at least the motion data corresponding to the same time period should be ignored or given less weight.

The controller 230 may also be configured to detect coughing, wheezing or other respiratory distress. For example, such conditions may include various detectable markers in the sound data and/or other data types. When these types of distresses are detected, the controller 230 can coordinate a distress message to one or more connected devices, such as to a device associated with a healthcare provider.

FIG. 8 illustrates an exemplary method 300 for determining a respiration rate using multiple biometric data sources. The method 300 may utilize any of the system and/or device embodiments described herein. As shown, the method 300 includes a step of receiving biometric data from a plurality of biometric sensors, the biometric data including a plurality of biometric data types (step 302). As described above, the biometric sensors and biometric data types may include a motion sensor for detecting motion data (particularly chest wall motion data), a microphone for detecting sound data (particularly breath and/or heartbeat sound data), and an electrical sensor for detecting electrical data (particularly EKG voltage data).

The method 300 also includes a step of associating the biometric data of each of the separate biometric data types with one another according to a common sampling timeline to generate a set of multiplexed biometric data (step 304). This may be accomplished using the controller 230, for example, which may be operationally located at the monitoring device 102, at an external base station, or divided between both.

The method 300 then uses the multiplexed biometric data to determine respiration rate (step 306). As described above, this may include differential weighing of different biometric data types and/or different time segments within the biometric data. It may also be based on the detection of one or more disruption events, such as a user talking, moving excessively, and the like.

FIG. 9 illustrates a method 400 of using a monitoring system as described herein to determine a respiration rate. The method 400 may utilize any of the system of device embodiments described herein. The method 400 includes the step of providing a monitoring system including a wearable monitoring device, the wearable monitoring device having a plurality of biometric sensors for obtaining a plurality of different biometric data types (step 402). The wearable monitoring device is placed on a user in an operational position (step 404). This is preferably on the chest wall of the user, as described above.

The method 400 also includes the step of operating the wearable monitoring device to obtain a plurality of biometric data types from the plurality of biometric sensors (step 406). The system may then use the biometric data to determine a respiration rate (step 408). As described above, this determination may include one or more filtering, weighing, and/or other adjustment processes to more accurately calculate the respiration rate.

Terms such as “approximately,” “about,” and “substantially,” as used herein represent an amount or condition close to the stated amount or condition that still performs a desired function or achieves a desired result. For example, the terms “approximately,” “about,” and “substantially” may refer to an amount or condition that deviates by less than 10%, or by less than 5%, or by less than 1%, or by less than 0.1%, or by less than 0.01% from a stated amount or condition.

Specific elements or components described in relation to any particular embodiment described herein may be substituted for or combined with elements described in relation to any other embodiment described herein. For example, any of the system of device embodiments described herein may be utilized to perform any of the methods described herein.

Claims

1-20. (canceled)

21. A monitoring system configured to determine a user's respiration rate, the monitoring system comprising: a wearable monitoring device including a plurality of biometric sensors, the biometric sensors being configured to generate biometric data including a plurality of separate biometric data types each generated by a different biometric sensor type; anda controller communicatively coupled to the wearable monitoring device, the controller including one or more processors and one or more hardware storage devices having stored thereon computer-executable instructions which are executable by the one or more processors to cause the controller to receive the biometric data from the plurality of biometric sensors;associate the biometric data of each of the separate biometric data types with one another according to a common sampling timeline to generate a set of multiplexed biometric data; anduse the multiplexed biometric data to determine respiration rate.

22. The monitoring system of claim 21, wherein the plurality of biometric sensors includes at least two biometric sensor types selected from the group consisting of a motion sensor type, a sound sensor type, and an electrical sensor type.

23. The monitoring system of claim 21, wherein the wearable monitoring device is a torso monitoring device configured in size and shape for wearable placement on the user's torso.

24. The monitoring system of claim 21, wherein the wearable monitoring device has an adhesive-backed interior side enabling attachment to the user's skin.

25. The monitoring system of claim 21, wherein the wearable monitoring device includes one or more rounded exterior surfaces.

26. The monitoring system of claim 21, wherein the wearable monitoring device has a height of about 4 mm to about 25 mm, or about 6 mm to about 15 mm.

27. The monitoring system of claim 21, wherein the plurality of biometric sensors includes at least one sensor from each of the following biometric sensor types: motion sensor, microphone, and electrical sensor.

28. The monitoring system of claim 27, wherein the electrical sensor is an EKG sensor.

29. The monitoring system of claim 27, wherein the microphone is a contact microphone including a piezo element.

30. The monitoring system of claim 21, wherein the computer-executable instructions are further configured to cause the controller to differentially weigh at least one of the biometric data types of the multiplexed biometric data.

31. The monitoring system of claim 30, wherein the differential weighing comprises reducing the weight of a particular biometric data type where a difference between the particular biometric data type and at least two other biometric data types surpasses a predetermined threshold.

32. The monitoring system of claim 21, wherein the computer-executable instructions are further configured to cause the controller to: identify a disruption event within a particular biometric data type;identify a time segment within the sampling timeline at which the disruption event occurs; andadjust at least a portion of the biometric data associated with the identified time segment.

33. The monitoring system of claim 32, wherein adjusting at least a portion of the biometric data includes adjusting at least one other biometric data type at the identified time segment.

34. The monitoring system of claim 33, wherein biometric data from the at least one other biometric data type at the identified time segment is ignored or is given reduced weight in determining the respiration rate.

35. The monitoring system of claim 32, wherein the disruption event includes a sound event within a sound biometric data type.

36. The monitoring system of claim 35, wherein the sound event includes a time segment where it is determined that the user is talking.

37. The monitoring system of claim 33, wherein the at least one other biometric data type is a motion biometric data type or an electrical biometric data type.

38. A method of determining respiration rate using multiple biometric data sources, the method comprising: providing a monitoring system as in claim 1;placing the wearable monitoring device upon a user in an operational position;operating the wearable monitoring device to obtain a plurality of biometric data types from the plurality of biometric sensors; andthe monitoring system using the biometric data to determine respiration rate.

39. The method of claim 38, wherein the wearable monitoring device is placed upon the chest wall of the user.

40. A computer-implemented method of determining respiration rate using multiple biometric data sources, the method comprising: receiving biometric data from a plurality of biometric sensors, the biometric data including a plurality of biometric data types each generated by a different biometric sensor type;associating the biometric data of each of the separate biometric data types with one another according to a common sampling timeline to generate a set of multiplexed biometric data; andusing the multiplexed biometric data to determine respiration rate.