System and method for remote patient screening and triage

Abstract

A system for short-term screening and triage of subjects includes a smart device carried by the subject and an application on the smart device programmed to instruct the subject to initiate a screening procedure and to record sensor data obtained from sensors mounted within the smart device, the application having a plurality of screening procedures programmed therein.

Description

Technical Field

[0001] The present disclosure relates to remote biosignal monitoring, such as, by way of non-limiting examples, subject monitoring, heart monitoring, and respiratory monitoring. More specifically, the present disclosure relates to systems and methods for short-term and long-term patient screening, taking as examples symptoms of diseases, bacterial or viral infections, heart-related complications, or respiratory-related complications. The present disclosure further relates to triaging patients using such systems and determining risk levels and priorities for further evaluation.

Background Art

[0002] Conventionally, monitoring, diagnosis, and evaluation of patients who may exhibit symptoms of infectious diseases or cardiorespiratory complications have required visits to clinics or hospitals. This process carries the risk of increased exposure to clinical teams and the general public, and during a crisis or pandemic, can overload the capacity of hospital systems for inpatients. Additionally, soldiers on the battlefield may have no caregivers nearby. Currently, the existing healthcare infrastructure is in a situation where there are too many patients in need of monitoring, but there is no large-capacity, accurate remote monitoring tool that can be easily used or deployed without physical access to a caregiver. Also, unlike a persistent condition, an episodic condition that occurs suddenly or intermittently requires an at-home screening solution that can be used immediately and continuously.

Summary of the Invention

[0003] Disclosed herein is a system that uses optical sensors, acoustic sensors, wireless sensors, etc., such as accelerometers, gyroscopes, pressure sensors, load sensors, weight sensors, force sensors, motion sensors, or vibration sensors, to capture mechanical vibrations of the body and physiological movements of the heart and lungs and convert them into biosignal information that can be used for screening and identifying disease states. The systems and methods presented herein can be used by a subject when experiencing symptoms of a complication or condition or, alternatively, can be used when receiving instructions from a physician in a telemedicine application.

[0004] The system is used for short-term and long-term screening. The short-term screening system can include a mobile phone, tablet, wearable watch, or any accessory available to the patient, having one or more sensors capable of capturing mechanical vibrations of the body, heart, and lungs, such as accelerometers, gyroscopes, pressure, load, weight, force, motion, or vibration. Such devices can be placed near physiological sources of the body, such as the heart and lungs, including but not limited to placement on the chest, abdomen, sides, back, etc. The long-term screening system can include sensors installable within or under the legs of a bed capable of capturing mechanical vibrations of the body, heart, and lungs, such as accelerometers, gyroscopes, pressure, load, weight, force, motion, or vibration. Short-term screening is aimed at examinations of limited duration (seconds to minutes), while long-term screening can be used continuously for any period of time (seconds, minutes, days, months, etc.). The short-term and long-term screening systems can operate independently or, alternatively, can operate synchronously as a unit and can exchange data, such as (past) trend data or baseline data of the subject's history.

[0005] The system can be used as a patient triage tool useful for assessing urgency. The system can include the installation of a smartphone application that enables screening and evaluating the patient's condition.

[0006] This disclosure is best understood from the following detailed description when read in conjunction with the accompanying drawings. It is emphasized that, in accordance with common practice, various features of the drawings are not to scale. On the contrary, the dimensions of various features are arbitrarily enlarged or reduced for clarity.

Brief Description of the Drawings

[0007]

Figure 1A

Figure 1B

Figure 1C

Figure 1D

[0008]

Figure 1E

[0009]

Figure 2A

[0010]

Figure 2B

[0011]

Figure 2C

[0012]

Figure 3

[0013]

Figure 4

[0014]

Figure 5

[0015]

Figure 6

[0016]

Figure 7

[0017]

Figure 8

DETAILED DESCRIPTION OF THE INVENTION

[0018] A method is disclosed for deploying a remote screening procedure and triaging a subject's health state using sensor data from such a system. In some implementations, the system can analyze the subject's heart information and determine heart rhythm, morphology, and rate information. The heart rhythm, morphology, and rate information can be utilized to identify the onset or worsening of heart conditions such as atrial fibrillation, atrial flutter, ventricular fibrillation, ventricular flutter, bundle branch block, valvular stenosis, myocardial ischemia, supraventricular tachycardia, etc. In some implementations, the system can analyze the subject's respiratory information and determine respiratory rhythm and rate information. The respiratory rhythm and rate information can be utilized to identify the onset or worsening of respiratory conditions such as shortness of breath, apnea, or hypopnea. In some implementations, the system can analyze the subject's cough information and determine cough rhythm and rate information. The cough rhythm and rate information can be utilized to identify the onset or worsening of cough conditions or respiratory flow such as wheezing, rales, snoring, ronchi, etc. The system can determine the severity of the condition (symptom), and / or the change and trend of the condition (symptom). The system can generate a remote screening report. The system can collect data. The generated report can be stored, transmitted to a physician, accessed by the physician for review, or analyzed using AI technology.

[0019] In some embodiments, the screening system can be utilized to monitor the body's reaction to a bacterial or viral infection. In particular, the sensor data can be utilized to monitor symptoms that are directly or indirectly related to an increase in body temperature. Such symptoms can include changes in respiratory flow and depth, respiratory rate, heart rate, heart rate variability, movement and tremors, body weight, fluid retention (status). The system can further create a patient baseline, continuously track and identify these changes over time, help the patient recognize their body's immune system response, and enable the patient or caregiver to monitor the medical condition.

[0020] In some embodiments, the screening system can be used in combination with a ventilator and can remotely monitor the effectiveness of the ventilator. Further, the system can be used as a data hub together with other sensors, and can relay local data from a pulse oximeter, a thermometer, a blood pressure monitor, or other sensors to the cloud, enabling remote monitoring of these additional sensors and enhancing the screening.

[0021] In some embodiments, the system includes two-way audio, text, and video for communicating with the patient.

[0022] In some embodiments, the screening system can also be utilized to screen cardiovascular and autonomic nerve indices. For example, the system can be used for a stress test at home, in which case sensor data can be used to monitor indices of heart rate variability to quantify dynamic autonomic regulation or heart rate recovery.

[0023] In some embodiments, the system can be used to create events based on the analysis of vitals. The event can be an audible sound or a message sent to the cloud for a critical condition. In some embodiments, the system enables data convergence between a short-term monitoring system and a long-term monitoring system. For example, one system can use historical data collected by another system to establish a baseline or to determine the progression or deterioration of a disease using such information.

[0024] In some embodiments, additional bed-based sensor data can be combined to cancel or remove common mode or other noise sources. Mobile data acquisition sensors can be used in combination with other monitoring systems having fixed positions, and data from various sources of data monitoring can be combined to increase the overall monitoring range when someone is moving.

[0025] Figures 1A through 1D illustrate an exemplary system for short-term screening and triage using a subject's smart device. Figure 1A is a flowchart of an exemplary method 100 for using a screening and triage system that uses a subject's smart device. In some implementations, the smart device can include a mobile phone, a tablet, a wearable watch, or any accessory available to the subject, having one or more sensors capable of capturing mechanical vibrations of the body, heart, and lungs, such as an accelerometer, a gyroscope, a pressure sensor, a load sensor, a weight sensor, a force sensor, a motion sensor, or a vibration sensor. These figures show a mobile phone as a non-limiting example.

[0026] The start of the screening can be triggered by the onset of symptoms of a disease or condition (e.g., when the subject feels unwell), or can be initiated according to a doctor's request (101). The subject receives instructions for the screening (102). In some implementations, the support can be provided by a doctor. In some implementations, the instructions can be provided by an app (application) installed on a smartphone or other mobile device. The instructions are specific for each condition, and thus, the screening procedure for an infectious disease can be different from that for a heart disease. FIGS. 1B through 1D are examples of screening instructions for the app to instruct the subject to perform the screening procedure (103). In some implementations, the app can instruct the subject to lie still in bed for a predetermined time as shown in FIG. 1B (a counter can be used to display the time or play a countdown sound). In some implementations, the app can instruct the subject to place the mobile phone at various positions on the chest as shown in FIG. 1C. In some implementations, the app can instruct the subject to lie on the side as shown in FIG. 1D. In some implementations, other placements are possible, for example, on the back, stomach, abdomen, etc. The app records sensor data, analyzes the sensor data (104), and generates a health report (105). In some implementations, the report can include physiological measurements such as, for example, heart rate, respiratory rate, heart rate variability, etc. In some implementations, the report can further include a list of identified or suspected problems, and an urgency (severity level). In some implementations, the app can send the data to a doctor or caregiver and propose follow-up examinations using the same or a different system.

[0027] Figure 1E is a graph of a data stream recorded during a short-term screening using a subject's smart device. Figure 1E shows an exemplary data stream when the subject is lying in bed, the smartphone is placed on the chest (as shown in Figure 1B), and the smartwatch is worn on the wrist. The data stream (X, Y, Z) from the accelerometer of the mobile phone is plotted in the upper panel. The data stream (X, Y, Z) from the accelerometer of the smartwatch is plotted in the central panel. The data stream (X, Y, Z) from the gyroscope of the mobile phone is plotted in the lower panel. The accelerometer data of the mobile phone captures the activities of both the heart and respiration. In this example, the respiration signal is strongly seen in the X and Y components, while the heart activity is strongest in the Z direction. The accelerometer data of the smartwatch does not capture the respiration signal. The smartwatch captures only the heart activity. The gyroscope data captures the activities of both respiration and the heart. The effects of respiratory interruptions are visible in the data streams where respiration is recorded (the accelerometer and gyroscope of the mobile phone). Cough episodes are visible in all the data streams that are recorded.

[0028] Figure 2A is a flowchart of an example of a method 200 for short-term screening and triage using a subject's smart device. In some implementations, the smart device can include a mobile phone, a tablet, a wearable watch, or any accessory available to the patient, having one or more sensors capable of capturing mechanical vibrations of the body, heart, and lungs, such as an accelerometer, a gyroscope, a pressure sensor, a load sensor, a weight sensor, a force sensor, a motion sensor, or a vibration sensor.

[0029] The start of screening can be triggered by the onset of symptoms of a disease or condition (e.g., when the subject feels unwell), or can be initiated according to a physician's request (201). After the start of screening, sensor data is acquired from the sensor (202). The sensor data is analyzed (203). The subject's condition is identified (determined) using the analyzed criteria (204). In some implementations, changes in respiratory flow and depth, respiratory rate, heart rate, heart rate variability, movement and sway, body weight, and body fluid retention are analyzed to quantify the body's response to a bacterial or viral infection. Values outside the normal range can define an out-of-standard condition, or, if the system has access to the patient's baseline, a sharp change compared to the baseline can be detected as an out-of-standard condition. In some implementations, a condition within the normal range, or a condition outside the normal range, can be relative to data representing the general (population) group. When a condition is identified (determined), it is determined whether immediate action is required (208). If necessary, immediate action is performed (209). In some implementations, the immediate action can be notification to the patient, notification to the patient's physician (attending physician), a call to a medical center, etc. If immediate action is not required, it is determined whether a follow-up examination is required to confirm the results or to provide new insights (210). If a follow-up is required, additional or new sensor data is collected and the aforementioned process starts anew. Otherwise, the screening process is terminated (211). In some implementations, the sensor data, the analyzed data, and the identified (determined) data can be stored locally in the local database 206 or in the cloud database 207 for future access (205).

[0030] Figure 2B shows an exemplary method 220 for short-term and long-term screening and triage. Short-term screening can utilize the subject's smart device. In some implementations, the smart device can include a mobile phone, a tablet, a wearable watch, or any accessory available to the patient, having one or more sensors capable of capturing mechanical vibrations of the body, heart, and lungs, such as an accelerometer, a gyroscope, a pressure sensor, a load sensor, a weight sensor, a force sensor, a motion sensor, or a vibration sensor. Long-term screening can use sensors installable within or under the legs of a bed (an example of a substrate on which the subject can be placed) capable of capturing mechanical vibrations of the body, heart, and lungs, such as accelerometers, gyroscopes, pressure, load, weight, force, motion, or vibration. If it is determined (229) that long-term screening is necessary based on short-term screening, the system can recommend the addition of a long-term screening system. Long-term screening can be added to enhance the tracking of continuous biosignals.

[0031] The data exchange process 231 enables data exchange between short-term screening and long-term screening, and the trend data (i.e., the subject's baseline data and historical data) can be accessed by any process. The data exchange process 231 also enables long-term screening to access short-term screening sensor data, synchronize it, and add it to the data stream set to enhance monitoring. In some implementations, data from sensor data acquisition 222, data from sensor data acquisition 232, data from data storage 226 including local-based and cloud-based storage, and data from data storage 236 including local-based and cloud-based storage can be input into the data exchange process 231. In some implementations, the data exchange process 231 outputs data to obtain trend data 224 and obtain trend data 234. That is, both initial and processed short-term and long-term data can be exchanged between short-term screening and long-term screening.

[0032] Figure 2C is a system 250 and system architecture for performing short-term and long-term screening and triage. System 250 includes one or more devices 260 that are connected to or communicating with (collectively "connected to") a computing platform 270. In some implementations, a machine learning training platform 280 may be connected to the computing platform 270. In some implementations, a user may access data via a connection device 290, which may receive data from the computing platform 270 or device 260. The connections between the one or more devices 260, the computing platform 270, the machine learning training platform 280, and the connection device 290 may be wired, wireless, optical, combinations thereof, and the like. System 250 is exemplary and may include more (additional), fewer, or different devices, entities, etc., designed similarly or differently without departing from the scope of this specification and the claims. Further, the illustrated devices may perform other functions without departing from the scope of this specification and the claims.

[0033] In some implementations, system 250, the sensors, and the data processing may be as described, for example, in U.S. Patent Application No. 16 / 777,385, filed January 30, 2020, U.S. Patent Application No. 16 / 595,848, filed October 8, 2019, and U.S. Provisional Patent Application No. 62 / 804,623, filed February 12, 2019 (collectively "the applications"). The entire disclosure content of these applications is incorporated herein by reference.

[0034] In some implementations, device 260 may include one or more sensors 261, a controller 262, a database 263, and a communication interface 264. In some implementations, the device 260may include a classifier 265 for applicable and appropriate machine learning techniques as described herein. One or more sensors 261 may detect and capture sensor data of vibration, pressure, force, weight, presence, and motion, related to a subject.

[0035] In some implementations, controller 262 may apply the processes and algorithms described herein in connection with FIGS. 1A, 2A, 2B, and 4 through 8 to sensor data to determine biometric signal information and data for short-term and long-term screening as described herein. In some implementations, classifier 265 may apply the processes and algorithms described herein in connection with FIGS. 1A, 2A, 2B, and 4 through 8 to sensor data to determine biometric signal information and data for short-term and long-term screening. In some implementations, classifier 265 may be implemented by controller 262. In some implementations, the captured sensor data and the biometric signal information and data for short-term and long-term screening may be stored in database 263. In one implementation, the captured sensor data and the biometric signal information and data for short-term and long-term screening may be transferred or transmitted to computing platform 270 via communication interface 264 for processing, storage, and / or combinations thereof. Communication interface 264 may be any interface and may use any communication protocol to communicate or transfer data between a source endpoint and a destination endpoint. In one implementation, device 260 may be any platform or structure that uses one or more sensors 261 to collect data from a subject for use by controller 262 and / or computing platform 270 as described herein. Device 260 and its internal elements may include other elements that may be desirable or necessary to implement the devices, systems, and methods described herein. However, since such elements and processes are well known in the art and do not serve to facilitate a better understanding of the disclosed embodiments, descriptions of such elements and processes may not be provided herein.

[0036] In some implementations, computing platform 270 may include a processor 271, a database 272, and a communication interface 273. In some implementations, computing platform 270 may include a classifier 274 for applicable and appropriate machine learning techniques as described herein. Processor 271 may obtain sensor data from sensor 261 or controller 262, and may apply the processes and algorithms described herein in connection with FIGS. 1A, 2A, 2B, and 4 through 8 to the sensor data to determine biometric signal information and data for short-term and long-term screening as described herein. In certain implementations, processor 271 may obtain biometric signal information and data for short-term and long-term screening as described herein from controller 262 and store it in database 272 for temporary and other types of analysis. In certain implementations, classifier 274 may apply the processes and algorithms described herein in connection with FIGS. 1A, 2A, 2B, and 4 through 8 to the sensor data to determine biometric signal information and data for short-term and long-term screening as described herein. Classifier 274 may apply classifiers to the sensor data and may determine biometric signal information and data for short-term and long-term screening as described herein through machine learning. In certain implementations, classifier 274 may be implemented by processor 271. In certain implementations, the captured sensor data and the biometric signal information and data for short-term and long-term screening may be stored in database 272. Communication interface 273 may be any interface and may use any communication protocol to communicate or transfer data between a source endpoint and a destination endpoint. In certain implementations, computing platform 270 may be a cloud-based platform. In certain implementations, processor 271 may be a cloud-based computer or an off-site controller.Computing platform 270 and the elements within it may include other elements that may be desirable or necessary to implement the devices, systems, and methods described herein. However, since such elements and processes are well known in the art and do not contribute to a better understanding of the disclosed embodiments, descriptions of such elements and processes may not be provided herein.

[0037] In certain implementations, machine learning training platform 280 may access and process sensor data and may train and generate multiple classifiers. The multiple classifiers may be transferred or transmitted to classifier 265 or classifier 274.

[0038] In FIG. 2B, sensor data is obtained from the sensor (232). In some implementations, the sensor data may be analyzed (233), for example, as shown to the application. Through data exchange process 231, instantaneous or substantially instantaneous data from short-term processing is obtained by long-term processing (234). Using the analyzed data and the obtained data, a state related to the subject is identified (235). The identified state and data are stored in local or cloud-based storage (236). As described herein, the identified state and data are also input into data exchange process 231. A determination is made as to whether the identified state requires immediate action (treatment) (237). If not, the scheduled long-term processing continues. If immediate action is required, a response action is executed (238).

[0039] In FIG. 2B, The method starts from step 221, sensor data is obtained from the sensor ( 222)。In some implementations, sensor data can be analyzed, for example, as presented to an application (223). Through data exchange process 231, trend data from long-term processing is obtained by short-term processing (224). Using the analyzed data and the obtained data, a state related to the subject is identified (225). The identified state and data are stored in local or cloud-based storage (226). As described herein, the identified state and data are also input into data exchange process 231. A determination is made as to whether the identified state requires an immediate action (treatment) (227). If an immediate action is required, a response action is executed (228). If an immediate action is not required, a determination is made as to whether long-term processing is required (229). If long-term processing is required, long-term processing is executed. If long-term processing is not required, a determination is made as to whether short-term processing is required (230). If short-term processing is not required, the current short-term processing ends. If short-term processing is required, data is obtained and another short-term processing is executed.

[0040] Figure 3 is an accelerometer data , gyroscope data , a processing pipeline 300 for obtaining sensor data such as pressure, load, weight, force, motion, or vibration (not limited to these). An analog sensor data stream 302 is received from sensor 301. A digitizer 303 digitizes the analog sensor data stream into a digital sensor data stream 304. A framer 305 generates a digital sensor data frame 306 from the digital sensor data stream 304. It includes all digital sensor data stream values within a fixed or adaptive time window. The processing pipeline 300 shown in FIG. 3 is exemplary and may include any, all, or none of the blocks or modules shown in FIG. 3, or combinations thereof. The order of the processing shown in FIG. 3 is exemplary, and the order of processing may be changed without departing from the scope of this specification or the claims.

[0041] Figure 4 is a preprocessing pipeline 400 for processing sensor data. The preprocessing pipeline 400 processes a digital sensor data frame 401. A noise reduction unit 402 removes or attenuates noise sources that can affect each sensor at the same level or different levels. The noise reduction unit 402 can utilize various techniques including, but not limited to, subtraction, combination of input data frames, adaptive filtering, wavelet transform, independent component analysis, principal component analysis, and / or other linear or non-linear transforms. A signal enhancement unit 403 can improve the signal-to-noise ratio of the input data. The signal enhancement unit 403 can be implemented as a linear or non-linear combination of the input data frame. For example, the signal enhancement unit 403 can combine signal deltas and increase the signal strength for higher resolution algorithm analysis. Subsampling units 404, 405, and 406 sample the digital enhanced sensor data and can include downsampling, upsampling, or resampling. The subsampling can be implemented as multistage sampling or polyphase sampling and can use the same or different sampling rates for heart analysis 407, respiratory analysis 408, and cough analysis 409. The order of the processing shown in Figure 4 is illustrative, and the order of the processing can be changed without departing from the scope of this specification or the claims.

[0042] FIG. 5 is an exemplary process 500 for cardiac analysis 407 using pre - processed and subsampled data 501. Filtering is used (502) to remove unwanted components of the input sensor data and / or to retain content useful for cardiac processing. In some implementations, the filtering can be an infinite impulse response (IIR) filter, a finite impulse response (FIR) filter, or a combination thereof. The filter can be a low - pass, high - pass, band - pass, band - stop, notch, or a combination thereof. In some embodiments, the filtering can include sources from other sensors, can remove common - mode or other noise, and can use adaptive filtering techniques to remove unwanted signals. The filtered sensor data is transformed (503) to enhance cardiac components by modeling the input signal as a set of waveforms of a specific form (sine waves of a Fourier transform, mother wavelets of a wavelet transform, and / or periodic basis functions of a periodicity transform). In some embodiments, the process can be a Fourier transform, a wavelet transform, a cosine transform, or a mathematical operation such as root mean square, absolute value, moving average, moving median, etc.

[0043] Envelope (signal envelope) detection is performed on the transformed sensor data. This provides an output corresponding to the outline of the input data, which is described by receiving a relatively high - frequency amplitude - modulated signal as input and connecting all local peaks of the signal ( 530 ). In some embodiments, envelope detection can use a low - pass filter, a Hilbert transform, or other envelope detection methods. Peak detection is performed ( 540 ) to find the maximum and minimum points of the input signal. In some embodiments, peak detection can return all peaks, all valleys, or only the most dominant ones (all).

[0044] Correlation analysis is performed (504) to measure the strength of the relationship between different segments of an input signal using linear and non-linear methods. The correlation analysis and peak positions can be used to identify individual beats of the input signal (505). The identified individual beats are enhanced (506). In some implementations, this may include applying a window, coefficients, or transform to enhance specific characteristics of the signal. Analysis in the time domain, frequency domain, or time-frequency domain can be performed, and the enhanced individual beats can be used to determine the heart rate (507). Also, analysis in the time domain, frequency domain, or time-frequency domain can be performed, and the enhanced individual beats can be used to determine a heart rate variability metric (508). In some implementations, the heart rate variability metric may include metrics such as SDNN, RMSSD, PNN50, LF, HF, and LF / HF. Also, analysis in the time domain, frequency domain, or time-frequency domain can be performed, and the heart beat components can be determined (509). In the case of a cardiac signal, the beat components can be the P, Q, R, S, and T waveforms, or atrial / ventricular depolarization and repolarization. An irregular rate or rhythm can be detected within the cardiac data using HR, HRV, beat components, and subject trend data 510 (i.e., baseline data and history data) (511).

[0045] The order of the processes shown in FIG. 5 is illustrative, and the order of the processes can be changed without departing from the scope of this specification or the claims.

[0046] FIG. 6 is an exemplary process 600 for respiratory analysis 408 using pre - processed and subsampled data 601. Filtering is used (602) to remove unwanted components of the input sensor data and / or to retain content useful for respiratory processing. In some implementations, the filter may use IIR, FIR, or a combination thereof. In some implementations, the filter may be a low - pass, high - pass, band - pass, band - stop, notch, or a combination thereof. In some embodiments, the filter may include sources from other sensors, may remove common - mode or other noise, and may use adaptive filtering techniques to remove unwanted signals. The filtered data is transformed (603) to enhance the respiratory component by modeling the input signal as a set of waveforms of a specific form (sine waves of Fourier transform, mother wavelets of wavelet transform, and / or periodic basis functions of periodic transform). The transformation can be a Fourier transform, wavelet transform, cosine transform, or mathematical operations such as root - mean - square, absolute value, moving average, moving median, etc. Peak detection is performed (605) to find the maximum and minimum points of the input signal. In some embodiments, the peak detection may return all peaks, all valleys, or only the most dominant ones (all).

[0047] Correlation analysis measures the strength of the relationship between different segments of the input signal using linear and non - linear methods (604). The correlation analysis and peak positions can be used (606) to identify individual breaths of the input signal. The identified individual breathing are enhanced (607). In some embodiments, this may include applying a window, coefficients, or a transformation to enhance specific characteristics of the signal.

[0048] Analysis in the time domain, frequency domain, or time-frequency domain can be utilized to determine the respiratory rate (breathing rate) using enhanced individual breaths (608). Also, analysis in the time domain, frequency domain, or time-frequency domain can be utilized to determine a respiratory rate variability metric using enhanced individual breaths (609). In some implementations, the respiratory rate variability metric can include metrics of DNN, RMSSD, PNN50, LF, HF, and LF / HF. Also, analysis in the time domain, frequency domain, or time-frequency domain can be utilized to determine a respiratory component (610). In the case of a respiratory signal, the respiratory component can be inhalation (inspiration) and exhalation. An irregular rate or rhythm can be identified within the respiratory data using RR, RRV, respiratory components, and subject trend data 611 (i.e., baseline data and history data) (612).

[0049] The order of the processes shown in FIG. 6 is illustrative, and the order of the processes can be changed without departing from the scope of this specification or the claims.

[0050] FIG. 7 is an exemplary process 700 for cough analysis 409 that uses pre - processed and subsampled data 701. Filtering is used (702) to remove unwanted components of the input sensor data and / or to retain content useful for cough processing. In some implementations, the filter can be an IIR, FIR, or a combination thereof. In some implementations, the filter can be a low - pass, high - pass, band - pass, band - stop, notch, or a combination thereof. In some embodiments, filtering can include sources from other sensors, can remove common - mode or other noise, and can use adaptive filtering techniques to remove unwanted signals. The filtered sensor data is transformed (703) to enhance the cough component by modeling the input signal as a set of waveforms of a specific form (sine waves of a Fourier transform, mother wavelets of a wavelet transform, and / or periodic basis functions of a periodicity transform). In some embodiments, the transformation can be a Fourier transform, wavelet transform, cosine transform, or mathematical operations such as root - mean - square, absolute value, moving average, moving median, etc. Peak detection is performed (605) to find the maximum and minimum points of the input signal. Envelope (signal envelope) detection can be performed (704) to receive a relatively high - frequency amplitude - modulated signal as input and provide an output corresponding to the outline of the input data described by connecting all the local peaks of the signal. In some embodiments, envelope detection can use a low - pass filter, Hilbert transform, or other envelope detection methods. It can be detected (705) that the patterns in the processed sensor data match the morphology (form) or spectral signature of a cough.

[0051] To measure the level of change of data compared to a baseline, variance analysis can be performed (706). In some implementations, this can be done by evaluating the standard deviation, coefficient of variation, etc. Variance analysis and cough signatures can be used to identify individual cough episodes within an input signal (707). Analysis in the time domain, frequency domain, or time-frequency domain can be utilized to determine the cough rate (708). Also, analysis in the time domain, frequency domain, or time-frequency domain can be utilized to determine the cough severity (709). Irregular coughs can be determined using the cough rate, cough severity, and subject trend data 710 (i.e., baseline data and historical data) (711).

[0052] The order of the processes shown in FIG. 7 is exemplary, and the order of the processes can be changed without departing from the scope of this specification or the claims.

[0053] FIG. 8 is an exemplary process for short-term screening and triage based on a machine learning classifier. Swimlane diagram 800 includes a device 801 that includes a first device set 806 and a second device set 807, a local database 802, a cloud server 803, a classifier factory 804, and a configuration server 805.

[0054] The first device set 806 receives (808), stores (809), and generates sensor data received by the cloud server 803 (811). The cloud server 803 searches for sensor data (812), and the classifier factory 804 generates or retrains a classifier (814). The generated or retrained classifier is stored by the classifier factory 804 (815). The generated or retrained classifier is used by the classifier factory 804 to classify sensor data (816) and automatically detect various arrhythmias, diseases, or abnormal conditions. The classified data is stored (813), and the subject trend data is stored (810). The configuration server 805 obtains the generated or retrained classifier and generates an update for the device 801 (817). In some implementations, the update can be an update to the app of the smart device or an update to the software of the remote device. The configuration server 805 sends the update to both the first device set 806 and the second device set 807 (818). Here, the second device set 807 can be a new device. Since the system has more data inputs available from more devices, it can be used to provide new or updated classifiers to older devices (such as the first device set 806). The system can also be used to provide software updates with improved accuracy, can learn personalized patterns, and can increase (enhance) the personalization of classifiers and data.

[0055] The order of the processes shown in FIG. 8 is illustrative, and the order of the processes can be changed without departing from the scope of this specification or the claims.

[0056] As a whole, a system for at least short-term screening and triage of a subject comprises a smart device owned by the subject and an application provided on the smart device, wherein a plurality of screening procedures are programmed in the application, and the application and the smart device are configured to provide an instruction to the subject to initiate a screening procedure, record short-term sensor data obtained from one or more sensors located within the smart device, and compare the short-term data, trend data, and overall population data to perform screening and triage of the subject.

[0057] In some implementations, the application and the smart device are further configured to analyze the short-term sensor data, obtain trend data from long-term screening, identify a state based on the analyzed short-term sensor data and the trend data, and execute an operation in response to the identified state. In some implementations, the operation includes one or more of generating an audible tone for a critical state, sending a message to a cloud entity for a critical state, and sending the sensor data and the identified state to an entity other than the subject. In some implementations, the application and the smart device are further configured to initiate long-term screening to generate the trend data in response to the identified state, the long-term screening system including one or more sensors installed proximate to a substrate on which the subject is placed, each sensor configured to capture mechanical vibrations from the subject's movement relative to the substrate, the mechanical vibrations indicative of the subject's biometric signal information. In some implementations, the long-term screening system is further configured to access the short-term sensor data and the identified state. In some implementations, the application and the smart device are further configured to initiate further short-term screening in response to the identified state. In some implementations, the application and the smart device are further configured to analyze the subject's heart information, determine heart rhythm and rate information from the heart information, determine the subject's health state from the heart rhythm and rate information, and identify the onset or progression of heart symptoms. In some implementations, the application and the smart device are further configured to analyze the subject's breathing information, determine breathing rhythm and rate information from the breathing information, determine the subject's health state from the breathing rhythm and rate information, and identify the onset or progression of breathing symptoms.In some implementations, the application and the smart device are further configured to analyze the cough information of the subject, determine the cough rhythm and rate information from the cough information, determine the health status of the subject from the cough rhythm and rate information, and identify the onset or progression of cough symptoms or respiratory flow. In some implementations, it is configured to determine the severity or progression of the condition. In some implementations, the smart device is one of a mobile phone, a tablet, a smartwatch, or an accessory, having one or more of an accelerometer, a gyroscope, a pressure sensor, a load sensor, a weight sensor, a force sensor, a motion sensor, a microphone, or a vibration sensor. In some implementations, the screening procedure includes an instruction for the subject to assume a certain posture, an instruction to stay in that posture for a specified time, and an instruction to place the smart device at one or more positions on the subject's body.

[0058] Overall, a system for screening and triage of a subject includes a smart device provided with an application, a substrate equipped with a plurality of sensors, and a processor connected to the sensors. The smart device is configured to instruct the subject to start a screening procedure, record short-term sensor data obtained from at least one sensor located within the smart device, and identify a state from the short-term sensor data and the resulting trend data. The plurality of sensors of the substrate are configured to capture mechanical vibrations from the subject's movements relative to the substrate, the mechanical vibrations indicating the subject's biosignal information. The processor is configured to capture sensor data from the plurality of sensors in response to the state identified by the smart device, identify a state from the sensor data captured from the plurality of sensors and the short-term sensor data obtained, and execute an operation based on the identified state.

[0059] In some implementations, the application and the smart device are further configured to analyze the short-term sensor data, obtain trend data from storage associated with the processor, and perform an operation in response to a state identified by the smart device. In some implementations, the application and the smart device are further configured to transfer the short-term sensor data to an entity for analyzing trend data, obtain an analysis result, and perform an operation in response to a state identified by the entity. In some implementations, the operation in response to a state identified by the smart device, or the operation in response to a state identified by the entity, includes one or more of generating an audible tone for a critical state, sending a message to a cloud entity for a critical state, and sending the short-term sensor data and the identified state to an entity other than the subject. In some implementations, the application and the smart device are further configured to initiate further screening by the smart device in response to a state identified by the smart device. In some implementations, the application and the smart device are further configured to analyze the subject's heart information, determine heart rhythm and rate information from the heart information, determine the subject's health state from the heart rhythm and rate information, and identify the onset or progression of heart symptoms, or analyze the subject's respiratory information, determine respiratory rhythm and rate information from the respiratory information, determine the subject's health state from the respiratory rhythm and rate information, and identify the onset or progression of respiratory symptoms, or analyze the subject's cough information, determine cough rhythm and rate information from the cough information, determine the subject's health state from the cough rhythm and rate information, and identify the onset or progression of cough symptoms or respiratory flow.

[0060] Overall, a method for at least short-term screening and triage of a subject comprises instructing the subject to initiate a screening procedure via a smart device; recording short-term sensor data from the subject by sensors on the smart device when the subject follows the screening procedure; analyzing the short-term sensor data; obtaining trend data from a long-term screening device; identifying the subject's condition based on the analyzed short-term sensor data and the trend data; and performing an action in response to the identified condition.

[0061] In some implementations, the method comprises initiating capture of sensor data in the long-term screening device in response to the identified condition. In some implementations, the method comprises analyzing sensor data by the long-term screening device, obtaining data from the smart device, identifying the subject's condition based on the analyzed sensor data by the long-term screening device, overall population data, and the short-term sensor data, and performing an action in response to the condition identified by the long-term screening device. In some implementations, the method comprises transmitting the short-term sensor data, the condition identified by the smart device, the sensor data by the long-term screening device, and the condition identified by the long-term screening device, respectively, to at least one entity other than the subject. In some implementations, the method comprises initiating further smart device-based screening in response to the identified condition.

[0062] Although the present disclosure has been described in connection with specific embodiments, it is to be understood that the disclosure is not limited to the disclosed embodiments. On the contrary, the disclosure is intended to cover various modifications and equivalent configurations included within the scope of the appended claims. The claims are to be given the broadest interpretation so as to encompass all such modifications and equivalent structures as is permitted under the law.

Claims

1. A system for at least short-term screening and triage of a subject, comprising: a smart device owned by the subject; an application provided on the smart device; The application has a plurality of screening procedures programmed therein. The application and the smart device are configured to: provide an instruction to the subject to start a screening procedure; record short-term sensor data obtained from one or more sensors located within the smart device; compare the short-term sensor data, trend data, and overall population data to perform screening and triage of the subject; analyze the short-term sensor data; obtain the trend data from a long-term screening system; identify a condition based on the analyzed short-term sensor data and the trend data; execute an operation in response to the identified condition. The long-term screening system includes one or more sensors installed in proximity to a substrate on which the subject is placed. Each sensor is configured to capture mechanical vibrations from the subject's movements relative to the substrate. The mechanical vibrations indicate the subject's biometric signal information. A system characterized by the above.

2. The operations include one or more of: generating an audible tone for a critical condition; sending a message to a cloud entity for a critical condition; sending the short-term sensor data and the identified condition to an entity other than the subject. The system according to claim 1, characterized by the above.

3. The application and the smart device are further configured to start a long-term screening to generate the trend data in response to the identified condition. The system according to claim 1, characterized by the above.

4. The long-term screening system is further configured to access the short-term sensor data and the identified condition. The system according to claim 1, characterized by the above.

5. The application and the smart device are further configured to start additional short-term screening in response to the identified condition. The system according to claim 1, characterized by the above. ​ ​ ​ ​ ​ ​ ​ The system according to claim 1, characterized in that

6. The application and the smart device further analyze the cardiac information of the subject, determine the rhythm and rate information of the heartbeat from the cardiac information, determine the health status of the subject from the rhythm and rate information of the heartbeat, identify the onset or progression of cardiac symptoms and are configured to The system according to claim 1, characterized in that

7. The application and the smart device further analyze the respiratory information of the subject, determine the rhythm and rate information of the respiration from the respiratory information, determine the health status of the subject from the rhythm and rate information of the respiration, identify the onset or progression of respiratory symptoms and are configured to The system according to claim 1, characterized in that

8. The application and the smart device further analyze the cough information of the subject, determine the rhythm and rate information of the cough from the cough information, determine the health status of the subject from the rhythm and rate information of the cough, identify the onset or progression of cough symptoms or respiratory flow and are configured to The system according to claim 1, characterized in that

9. The application and the smart device further determine the severity or degree of progression of the identified condition and are configured to The system according to claim 1, characterized in that

10. The smart device is one of a mobile phone, a tablet, a smart watch, or an accessory, having one or more of an accelerometer, a gyroscope, a pressure sensor, a load sensor, a weight sensor, a force sensor, a motion sensor, a microphone, or a vibration sensor The system according to claim 1, characterized in that

11. The screening procedure includes an instruction for the subject to assume a certain posture, an instruction to stay in that posture for a specified time, and an instruction to place the smart device at one or more positions on the subject's body and includes The system according to claim 1, characterized in that

12. A system for screening and triage of a subject, comprising a smart device provided with an application, a substrate equipped with a plurality of sensors, a processor connected to the sensors, and comprising The smart device collectively instructs the subject to start a screening procedure Record short-term sensor data obtained from at least one sensor located within the smart device, and Identify the state from the short-term sensor data and the obtained trend data It is configured as The plurality of sensors on the substrate are configured to capture mechanical vibrations from the movements of the subject with respect to the substrate, The mechanical vibrations indicate the biometric signal information of the subject, The processor In response to the state identified by the smart device, capture sensor data from the plurality of sensors, Identify the state from the sensor data captured from the plurality of sensors and the obtained short-term sensor data, Execute an operation based on the identified state It is configured as A system characterized by this.

13. The application and the smart device further Analyze the short-term sensor data, Obtain trend data from the storage associated with the processor, Execute an operation in response to the state identified by the smart device The system according to claim 12, which is configured as

14. The application and the smart device further Transfer the short-term sensor data to an entity for analyzing trend data, Obtain the analysis result, Execute an operation in response to the state identified by the entity The system according to claim 13, which is configured as

15. The operation in response to the state identified by the smart device, or the operation in response to the state identified by the entity, Generating an audible tone for a critical state, Sending a message to a cloud entity for a critical state, Sending the short-term sensor data and the identified state to an entity other than the subject, Including one or more of The system according to claim 14, characterized by this.

16. The application and the smart device further Start screening by a further smart device in response to the state identified by the smart device It is configured as The system according to claim 12, characterized by this.

17. The application and the smart device further Analyze the cardiac information of the subject, determine the rhythm and rate information of the heartbeat from the cardiac information, determine the health status of the subject from the rhythm and rate information of the heartbeat, and determine whether it is configured to identify the onset or progression of cardiac symptoms, Analyze the respiratory information of the subject, determine the rhythm and rate information of breathing from the respiratory information, determine the health status of the subject from the rhythm and rate information of breathing, and determine whether it is configured to identify the onset or progression of respiratory symptoms, or, Analyze the cough information of the subject, determine the rhythm and rate information of the cough from the cough information, determine the health status of the subject from the rhythm and rate information of the cough, and determine whether it is configured to identify the onset or progression of cough symptoms or respiratory flow The system according to claim 12, characterized in that.

18. A method of operating a system for at least short-term screening and triage of a subject, comprising: Instructing the subject to initiate a screening procedure via a smart device; When the subject follows the screening procedure, recording short-term sensor data from the subject by a sensor on the smart device; Analyzing the short-term sensor data; Obtaining trend data from a long-term screening device; Identifying the state of the subject based on the analyzed short-term sensor data and the trend data; Performing an operation in response to the identified state; Comprising The long-term screening system includes one or more sensors installed in proximity to a substrate on which the subject is placed, Each sensor is configured to capture mechanical vibrations from the movement of the subject relative to the substrate, The mechanical vibration indicates the biosignal information of the subject A method characterized by that.

19. A method of operating a system according to claim 18, further comprising starting to capture sensor data in the long-term screening device in response to the identified state The method of operating a system according to claim 18, characterized in that it further comprises.

20. Analyzing sensor data from a long-term screening device; Obtaining data of the smart device; Identifying the state of the subject based on the analyzed sensor data from the long-term screening device, the overall population data, and the short-term sensor data; performing an operation in response to the specified state of the subject; The method of operating a system according to claim 19, further comprising the above. **Claim 21** transmitting short-term sensor data, a state identified by a smart device, long-term sensor data by a long-term screening device, and a state identified by a long-term screening device, respectively, to at least one entity other than the subject The method of operating a system according to claim 20, further comprising the above. **Claim 22** starting screening by a further smart device in response to the specified state The method of operating a system according to claim 18, further comprising the above.

Images