Health monitoring systems and methods

Abstract

Systems, methods and devices for reducing noise in health monitoring including monitoring systems, methods and/or devices receiving a health signal and/or having at least one electrode or sensor for health monitoring.

What is claimed is: 
     
         1 . A method for determining pulse transit time comprising:
 generating one or more PPG frames from heartbeat information and one or more PPG signals;   creating a weighted, n-sample moving-average frame; and,   determining the pulse transit time by finding the shift in the frame signal with respect to the heartbeat using the n-sample moving-average frame.   
     
     
         2 . A method according to  claim 1  wherein the pulse transit time is used to determine blood pressure. 
     
     
         3 . A method according to  claim 1  including using heartbeat information and one or more PPG signals to generate one or more PPG frames. 
     
     
         4 . A method according to  claim 3  where each frame contains the PPG signal of a single wavelength for the duration of one corresponding beat of the heart. 
     
     
         5 . A method according to  claim 1  including using ECG to generate heartbeat timing information by detecting the R-wave or other ECG feature from each beat. 
     
     
         6 . A method according to  claim 1  including performing a PPG signal quality estimate. 
     
     
         7 . A method according to  claim 6  wherein the PPG signal quality estimate uses one or more of PPG variance, estimated PPG signal-to-noise ratio, PPG signal saturation, patient motion accelerometer or gyroscope data, an ECG or impedance measurement noise estimate, or other information about the PPG signal quality. 
     
     
         8 . A method according to  claim 6  wherein the signal quality estimate is used in conjunction with the heartbeat timing information to generate the gain for each frame. 
     
     
         9 . A method according to  claim 8 , where lower signal quality results in a lower gain. 
     
     
         10 . A method according to  claim 6  wherein the signal quality estimate is a constant used for the gain information. 
     
     
         11 . A method according to  claim 8  wherein the gain information is used with the frame information to create a weighted, n-sample moving-average frame. 
     
     
         12 . A method according to  claim 11 , where the PPG signal that is correlated with the heartbeat timing is reinforced while the uncorrelated noise is reduced. 
     
     
         13 . A method according to  claim 11  wherein one or more of:
 the number of samples included in the frame (n) are adapted to reduce noise or decrease response time; 
 the frames are additionally weighted by time in order to increase the contribution of recent or near-future frames with respect to frames that are further away and potentially less-relevant; and, 
 the additional weighting by time is implemented using an IIR or FIR filter. 
 
     
     
         14 . A method according to  claim 11  further comprising determining the pulse transit time by finding the shift in the frame signal with respect to the heartbeat. 
     
     
         15 . A method according to  claim 11  further including one or more of:
 finding the sample index where the signal is at a minimum or maximum and comparing it with the frame boundary (heartbeat timing) to determine the pulse transit time; 
 interpolating the signal using a spline or polynomial fit around the minimum or maximum values, allowing the minimum or maximum to be determined with greater precision than the sample rate; or, 
 comparing the frame to a reference frame template, where the average frame is shifted with respect to the template; wherein the shift with the highest correlation between the average frame and the template indicates the transit time; wherein:
 the reference template may be a predetermined signal, or 
 it may be allowed to adapt by using a long-term frame average with a known transit time. 
 
 
     
     
         16 . A method according to  claim 1  including one or both of performing a PPG signal quality estimate; and the signal quality estimate is used in conjunction with the heartbeat timing information to generate the gain for each frame; and, wherein
 one or both of the PPG signal quality estimate and the gain information is used with the frame information to create a weighted, n-sample moving-average frame. 
 
     
     
         17 . A method according to  claim 1 , where the PPG signal that is correlated with the heartbeat timing is reinforced while the uncorrelated noise is reduced. 
     
     
         18 . A method according to  claim 1  wherein one or more of:
 the number of samples included in the frame (n) are adapted to reduce noise or decrease response time; 
 the frames are additionally weighted by time in order to increase the contribution of recent or near-future frames with respect to frames that are further away and potentially less-relevant; and, 
 the additional weighting by time is implemented using an IIR or FIR filter. 
 
     
     
         19 . A method according to  claim 1  further comprising determining the pulse transit time by finding the shift in the frame signal with respect to the heartbeat. 
     
     
         20 . A method according to  claim 1  further including one or more of:
 finding the sample index where the signal is at a minimum or maximum and comparing it with the frame boundary (heartbeat timing) to determine the pulse transit time; 
 interpolating the signal using a spline or polynomial fit around the minimum or maximum values, allowing the minimum or maximum to be determined with greater precision than the sample rate; or, 
 comparing the frame to a reference frame template, where the average frame is shifted with respect to the template; wherein the shift with the highest correlation between the average frame and the template indicates the transit time; wherein:
 the reference template may be a predetermined signal, or 
 it may be allowed to adapt by using a long-term frame average with a known transit time. 
 
 
     
     
         21 . A system using the method according to  claim 1  wherein one or more of the signals are obtained from one or more of:
 a wearable health monitoring device configured to be adhered to the skin of a subject for the physiological parameter monitoring; the device comprising: 
 a substrate; 
 a conductive sensor connected to the substrate, and 
 a double-sided composite adhesive having:
 at least one conductive adhesive portion, and 
 at least one non-conductive adhesive portion; 
 
 the double-sided composite adhesive being attached to the substrate and the conductive sensor; 
 the at least one conductive adhesive portion being disposed in conductive communicative contact with the conductive sensor, and 
 being configured to be conductively adhered to the skin of the subject for conductive signal communication from the subject to the conductive sensor. 
 
     
     
         22 . A device using the method of  claim 1 . 
     
     
         23 . A system using the device according to  claim 22  the device being a wearable health monitoring device. 
     
     
         24 . A method for health monitoring comprising:
 determining from a user's ECG signal when heart beats occur;   time averaging each of two photoplethysmogram signals correlated to the beat locations;   generating ensemble averages;   using a linear regression of the ensemble averages to determine the linear gain factor between the two signals;   determining from the gain factor the patient oxygen saturation.   
     
     
         25 . A method according to  claim 24  further comprising:
 recording ECG data in time-concordance with two or more photoplethysmographs of different light wavelengths; 
 detecting the heart beats in the ECG signal, these heart beats allowing for definition of a frame of photoplethysmogram data for the time between two adjacent heart beats; 
 averaging two or more of these frames together at each point in time to create an average frame for the time interval; wherein the photoplethysmograph signal is reinforced by this averaging because the photoplethysmogram is correlated with the heartbeat, and any motion artifact or other noise source that is uncorrelated in time with the heartbeat is diminished; 
 interpreting the signal-to-noise ratio of the average frame as typically higher than that of the individual frames; 
 using linear regression to estimate the gain between the two average frame signals; 
 estimating from this gain value one or more of the blood oxygen saturation or other components present in the blood such as hemoglobin, carbon dioxide or others. 
 
     
     
         26 . A method according to  claim 25  wherein the operations are repeated for additional light wavelengths in order to estimate from the gain value one or more of the blood oxygen saturation or other components present in the blood such as hemoglobin, carbon dioxide or others. 
     
     
         27 . A method comprising:
 selecting a possible gain value,   multiplying the average frame signal by it, and   determining the residual error with respect to an average frame of a different wavelength.   
     
     
         28 . A method according to  claim 27  comprising:
 repeating the operations of the method  claim 27  for a number of potential gain values. 
 
     
     
         29 . A method according to  claim 27  wherein this method includes finding local minima; wherein if it is likely that the global minimum represents correlation caused by motion artifact, venous blood movement or another noise source, it may be ignored, and a local minimum may be selected instead. 
     
     
         30 . A method according to  claim 27  whereby the gain between red and infrared (IR) frame signals are found by:
 averaging the two frames together first to provide a signal with reduced noise; 
 performing linear regression of the red versus combined and IR versus combined and then finding the ratio of these two results. 
 
     
     
         31 . A method according to  claim 27  for determining depth and/or rate of respiration. 
     
     
         32 . A method according to  claim 31  wherein one or more of ECG data, PPG data, pulse oximeter data and/or accelerometer data is used to determine respiration rate and/or depth. 
     
     
         33 . A method according to  claim 32  including generating, using the one or more of ECG data, PPG data, pulse oximeter data and/or accelerometer data, a respiration waveform. 
     
     
         34 . A method according to  claim 31  where Red or IR values over time are used. 
     
     
         35 . A method according to  claim 31  including measuring IR or red reflection by the photodiode to estimate depth and/or rate of respiration. 
     
     
         36 . A method according to  claim 31  where one or both of maximum values and minimum values of a curve or waveform of the IR or red data represent the difference between the maximum and minimum values related to the depth of breath in an individual being monitored, and/or over time, the rate of respiration can be evaluated from the curve of maximum and minimum values over time. 
     
     
         37 . A method according to  claim 31  using PPG signals. 
     
     
         38 . A method according to  claim 37  including
 generating the PPG signals when the chest expands and contracts during breathing, the motion hereof presenting as a wandering baseline artifact on PPG signals. 
 
     
     
         39 . A method according to  claim 37  including
 isolating the respiration signal by filtering out the PPG data to focus on the breathing/respiration signal. 
 
     
     
         40 . A method according to  claim 37  wherein the PPG is chest-mounted. 
     
     
         41 . A method according to  claim 31  using accelerometer signals. 
     
     
         42 . A method according to  claim 41  including
 generating the accelerometer signals when one or both the chest expands and contracts, or the chest accelerates up and down, as measured by the accelerometer. 
 
     
     
         43 . A method according to  claim 41  including
 isolating the respiration signal may by filtering out the accelerometer data to focus on the breathing/respiration signal. 
 
     
     
         44 . A method according to  claim 41  wherein the accelerometer is chest-mounted. 
     
     
         45 . A method according to  claim 44  wherein the user is lying on his/her back. 
     
     
         46 . A device using the method according to  claim 27  wherein one or more of the signals are obtained from one or more of:
 a wearable health monitoring device having: 
 a substrate; 
 a conductive sensor connected to the substrate, and 
 a double-sided composite adhesive having:
 at least one conductive adhesive portion, and 
 at least one non-conductive adhesive portion; 
 
 the double-sided composite adhesive being attached to the substrate and the conductive sensor; 
 the at least one conductive adhesive portion being disposed in conductive communicative contact with the conductive sensor, and 
 being configured to be conductively adhered to the skin of the subject for conductive signal communication from the subject to the conductive sensor. 
 
     
     
         47 . A method for determining respiration rate and/or depth comprising:
 generating one or more of ECG, PPG and/or accelerometer data from respiration of a user;   creating a respiration waveform therefrom; and,   determining the respiration rate and depth using the respiration waveform.   
     
     
         48 . A method according to  claim 47  wherein one or both the respiration rate and/or depth are estimated from the respiration signal using time-domain and/or frequency domain methods. 
     
     
         49 . A method according to  claim 31  wherein any one or more of PPG and/or accelerometer, methods are used discretely or in combination with each other and/or with the above-described ECG-based respiration estimation techniques. 
     
     
         50 . A method according to  claim 31  wherein using multiple methods improves accuracy when compared to estimates based on a single method. 
     
     
         51 . A system using the device according to  claim 46  further including a computer. 
     
     
         52 . A device using the method of  claim 47 . 
     
     
         53 . A device according to  claim 52  using a wearable health monitoring device. 
     
     
         54 . A method according to  claim 27  using software and computer hardware to determine the oxygen saturation.