Methods and systems for assessing health conditions or conditions

Abstract

According to one aspect, a system for evaluating health status includes: a memory containing cardiopulmonary coupling data and heart rate data of a person during a sleep period; one or more processors; and a memory storing instructions. When the instructions are executed by the one or more processors, the system accesses the cardiopulmonary coupling data and the heart rate data from the memory, identifies one or more time periods in the sleep period that are classified as a high-frequency coupling state based on the cardiopulmonary coupling data, evaluates one or more features of a portion of the heart rate data corresponding to the one or more time periods, and evaluates the health status of the person based on the one or more features of a portion of the heart rate data corresponding to the one or more time periods classified as a high-frequency coupling state.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure claims the benefit of and priority to U.S. Provisional Application No. 62 / 686,907, filed on June 19, 2018, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to assessment of health conditions or conditions, and more particularly, to assessment of health conditions or conditions based on cardiopulmonary coupling. Background Art

[0004] Techniques for studying a person's sleep range from simple actigraphy to polysomnography. Actigraphy involves analyzing a person's movements to infer sleep quality and / or sleep disturbances. Polysomnography also involves analysis of sleep quality and / or sleep disturbances, but involves much more than just activity data. Examples of other types of data that may be included in a polysomnogram include blood oxygen levels, brain waves, breathing rate, and eye movements.

[0005] Various other methods have been explored to assess sleep quality and / or sleep disorders. For example, cardiopulmonary coupling is a frequency analysis of the coupling between heart rate variability and respiration during sleep. Various aspects of cardiopulmonary coupling have been disclosed in U.S. Patent No. 7,324,845, U.S. Patent No. 7,734,334, U.S. Patent No. 8,403,848, and U.S. Patent No. 8,401,626. The entire contents of these patents are incorporated herein by reference.

[0006] A wide variety of information can be considered when evaluating sleep analysis. However, the use of this information is limited, except for the analysis of sleep quality or sleep disorders. Summary of the invention

[0007] The present application relates to the assessment of health conditions or health status based on cardiopulmonary coupling data and heart rate data during sleep. The terms "health condition" and "health status" are used interchangeably herein, and it is intended that reference to one term also applies to and encompasses the other term.

[0008] According to one aspect of the present disclosure, a system for evaluating health status includes: a memory containing cardiopulmonary coupling data and heart rate data of a person during a sleep period; one or more processors; and a memory coupled to the one or more processors and storing instructions. When the instructions are executed by the one or more processors, the system: accesses the cardiopulmonary coupling data and the heart rate data from the memory, identifies one or more time periods in the sleep period that are classified as a high-frequency coupling state based on the cardiopulmonary coupling data, evaluates one or more features of the portion of the heart rate data corresponding to the one or more time periods, and evaluates the health status of the person based on the one or more features of the portion of the heart rate data corresponding to the one or more time periods classified as a high-frequency coupling state.

[0009] In various embodiments, when evaluating one or more features, when the instructions are executed by one or more processors, the system classifies the portion of the heart rate data corresponding to one or more time periods as one of the following: a substantial increase, an increase, a flattening, a decrease, or a substantial decrease. In various embodiments, when evaluating a health condition, when the instructions are executed by the one or more processors, the system estimates the health condition as having a potential problem if the heart rate data within one or more time periods is classified as an increase or a substantial increase.

[0010] In various embodiments, when identifying one or more time periods in the sleep period, when the instructions are executed by the one or more processors, the system identifies multiple time periods in the sleep period that are classified as high-frequency coupling states. In various embodiments, when evaluating one or more features, when the instructions are executed by the one or more processors, the system determines an average high-frequency coupling heart rate based on the portion of the heart rate data corresponding to the multiple time periods classified as high-frequency coupling states.

[0011] In various embodiments of the system, when the instructions are executed by the one or more processors, the system is further caused to identify multiple time periods in the sleep period classified as low-frequency coupling states based on the cardiopulmonary coupling data; and determine an average low-frequency coupling heart rate based on the portion of the heart rate data corresponding to the multiple time periods classified as low-frequency coupling states. In various embodiments, when evaluating the health condition, when the instructions are executed by the one or more processors, the system is caused to estimate the health condition as having a potential health problem if the average high-frequency coupling heart rate is greater than the average low-frequency coupling heart rate.

[0012] In various embodiments, when determining the average high-frequency coupling heart rate and determining the average low-frequency coupling heart rate, when the instructions are executed by the one or more processors, the system performs a multivariate linear regression based on a model with heart rate as the response variable, high-frequency coupling state as the explanatory variable, and very-low-frequency coupling state as the explanatory variable.

[0013] In various embodiments, the system also includes a display device, and when the instructions are executed by the one or more processors, the system also displays the beta coefficient of the high-frequency coupling state from the multivariate linear regression model on the display device, and the beta coefficient indicates the difference between the average high-frequency coupling heart rate and the average low-frequency coupling heart rate.

[0014] In various embodiments of the system, when the instructions are executed by the one or more processors, the system is further caused to determine the heart rate trend during the sleep period based on the heart rate data. In various embodiments, when evaluating the health condition, when the instructions are executed by the one or more processors, the system is caused to estimate the health condition as having potential health problems if the heart rate trend during the sleep period does not decrease.

[0015] In various embodiments, when evaluating the at least one characteristic, when the instructions are executed by the one or more processors, the system divides the at least one time period presenting high-frequency coupling into multiple time intervals of equal duration, determines the average heart rate of each of the multiple time intervals based on the portion of the heart rate data corresponding to the multiple time intervals, and identifies pairs of adjacent time intervals in which the increase in the average heart rate in the multiple time intervals exceeds a threshold.

[0016] According to one aspect of the present disclosure, a computer-implemented method for assessing health status includes: accessing cardiopulmonary coupling data and heart rate data of a person during a sleep period; based on the cardiopulmonary coupling data, identifying one or more time periods in the sleep period that are classified as a high-frequency coupling state; evaluating one or more features of the portion of the heart rate data corresponding to the one or more time periods; and assessing the health status of the person based on the one or more features of the portion of the heart rate data corresponding to the one or more time periods classified as a high-frequency coupling state.

[0017] In various embodiments of the method, evaluating the feature includes: classifying the portion of the heart rate data corresponding to the one or more time periods as one of: substantially increased, increased, flat, decreased, or substantially decreased. In various embodiments of the method, evaluating the health condition includes: estimating the health condition as having a potential problem when the heart rate data within the at least one time period is classified as increased or substantially increased.

[0018] In various embodiments of the method, identifying one or more time periods in the sleep period includes identifying a plurality of time periods in the sleep period that are classified as high frequency coupling states. In various embodiments of the method, evaluating the one or more features includes determining an average high frequency coupling heart rate based on portions of the heart rate data corresponding to the plurality of time periods classified as high frequency coupling states.

[0019] In various embodiments of the method, the method includes: identifying multiple time periods in the sleep period classified as low-frequency coupling states based on the cardiopulmonary coupling data; and determining an average low-frequency coupling heart rate based on the portion of the heart rate data corresponding to the multiple time periods classified as low-frequency coupling states. In various embodiments of the method, assessing the health condition includes: estimating the health condition as having a potential health problem when the average high-frequency coupling heart rate is greater than the average low-frequency coupling heart rate.

[0020] In various embodiments of the method, determining the average high-frequency coupling heart rate and determining the average low-frequency coupling heart rate include: performing multiple linear regression based on a model with heart rate as a response variable, high-frequency coupling state as an explanatory variable, and very-low-frequency coupling state as an explanatory variable.

[0021] In various embodiments of the method, the method includes displaying a beta coefficient of the high frequency coupling state from a multiple linear regression model on a display device, the beta coefficient indicating a difference between the average high frequency coupling heart rate and the average low frequency coupling heart rate.

[0022] In various embodiments of the method, the method includes determining a heart rate trend within the sleep period based on the heart rate data. In various embodiments of the method, assessing the health condition includes: when the heart rate trend within the sleep period does not decrease, estimating the health condition as having a potential health problem.

[0023] In various embodiments of the method, evaluating the at least one characteristic includes: dividing the at least one time period exhibiting high-frequency coupling into a plurality of time intervals of equal duration; determining an average heart rate for each of the plurality of time intervals based on a portion of the heart rate data corresponding to the plurality of time intervals; and identifying pairs of adjacent time intervals in which an increase in the average heart rate in the plurality of time intervals exceeds a threshold.

[0024] Further details and aspects of exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1is a diagram of an exemplary sleep study measurement system according to some aspects of the present disclosure;

[0026] Figure 2 is a graph of exemplary cardiopulmonary coupling (CPC) data and heart rate data according to some aspects of the present disclosure;

[0027] Figure 3 is a diagram of exemplary CPC data and CPC status classification according to some aspects of the present disclosure;

[0028] Figure 4 exemplary heart rate data and CPC status classification intent for healthy individuals according to aspects of the present disclosure;

[0029] Figure 5 is a graph of exemplary heart rate data and CPC status classifications for individuals with health issues according to aspects of the present disclosure;

[0030] Figure 6 is a graph of exemplary heart rate data over a complete sleep period according to some aspects of the present disclosure;

[0031] Figure 7 is a graph of exemplary heart rate data during a high frequency coupling period according to some aspects of the present disclosure;

[0032] Figure 8 is a graph of another exemplary heart rate data during a high frequency coupling period according to some aspects of the present disclosure;

[0033] Fig. 9 is a graph of exemplary CPC data, heart rate data, and CPC status classifications for a healthy individual according to aspects of the present disclosure;

[0034] Fig.10 is a graph of exemplary CPC data, heart rate data, and CPC status classifications for individuals with health issues according to aspects of the present disclosure;

[0035] Fig.11 is a flowchart of exemplary operations according to some aspects of the present disclosure; and

[0036] Fig.12 is a block diagram of an exemplary system according to some aspects of the present disclosure. DETAILED DESCRIPTION

[0037] The present application relates to the assessment of health conditions or health conditions based on cardiopulmonary coupling data and heart rate data during sleep.

[0038] Cardiopulmonary coupling is a sleep analysis technique that quantitatively analyzes two physiological signals - a sequence of NN intervals consisting of heart rate variability combined with a corresponding direct or derived respiratory signal - to determine the coherent cross power of the two signals. In turn, the coherent cross power provides a measure of cardiopulmonary coupling that can be used for sleep analysis. In various embodiments, the two physiological signals can be derived from various physiological measurements, as described in the U.S. patents listed above.

[0039] Reference now Figure 1 , a schematic diagram of a system 100 according to some aspects of the present disclosure. The system 100 can be attached to a person during sleep to obtain physiological measurements, such as electrocardiogram measurements or other physiological measurements, that can be used to calculate cardiopulmonary coupling ("CPC"). The system 100 also obtains heartbeat and / or heart rate measurements. A person skilled in the art will appreciate the various sensors used to detect physiological signals and heartbeat / heart rate. The physiological measurements can be recorded in a storage medium, such as a disk drive, a flash drive, a solid-state drive, or other storage medium in the system 100. The physiological measurements used to calculate CPC and the heartbeat / heart rate measurements are recorded in parallel. In various embodiments, the recorded data can be marked or associated with a timestamp. By marking the recorded data or associating the recorded data with a timestamp, different types of recorded measurements can be correlated in time. The physiological measurements can be used to determine cardiopulmonary coupling data, and the physiological measurements and / or CPC data can be recorded in a storage medium. In various embodiments, the CPC data can be marked or associated with a timestamp. The disclosed embodiments are exemplary, and it is contemplated that other ways of associating recorded measurements with time may be employed.

[0040] Figure 22 is a graph of exemplary CPC data and heart rate data associated with each other in time. CPC data 210 is plotted at the top of the graph in units of Hertz (Hz), ranging from 0 Hz to 0.5 Hz. Heart rate data 220 is plotted at the bottom of the graph in units of beats per minute (BPM), ranging from 50 BPM to 80 BPM. In the CPC data portion 210 in the figure, the upper portion reflects high-frequency coupling, while the lower portion reflects low-frequency coupling and / or very low-frequency coupling. High-frequency coupling represents stable sleep, which is a biomarker of comprehensive stable non-rapid eye movement (NREM) sleep, and is associated with periods of stable breathing, high vagal tone, generally non-periodic alternating patterns on the electroencephalogram (EEG), high relative delta power, physiological blood pressure drop (health), and / or a stable arousal threshold. Low-frequency coupling represents unstable sleep, which is a biomarker of comprehensive unstable NREM sleep, with characteristics opposite to stable sleep. Unstable sleep is associated with electroencephalogram (EEG) activity called the Cyclic Alternating Pattern (CAP), periods of fluctuating breathing patterns (tidal volume fluctuations), cyclical variations of heart rate (CVHR), no drop in blood pressure, and / or variable arousal thresholds. Fragmented rapid eye movement (REM) sleep is characterized by low-frequency coupling. Very low-frequency coupling represents REM sleep and arousals.

[0041] As used herein, the term "state" or "CPC state" may refer to a particular type of CPC coupling that is superior to other types of CPC coupling for a period of time, such as a High Frequency Coupling (HFC) state, a Low Frequency Coupling (LFC) state, or a Very Low Frequency Coupling (vLFC) state. Those skilled in the art will recognize that the LFC state may be further subdivided into LFC narrowband or LFC wideband, and the vLFC state may be further subdivided into REM or Wake. With respect to CPC states, the term "period of time" may refer to a period of time during which a CPC state is present. Figure 3 A schematic diagram of cardiopulmonary coupling data is shown, which has been classified into various CPC states 310, including HFC state, LFC state, vLFC state or non-CPC state. In addition, Figure 3The bottom portion of FIG. 3 shows other data 320 that may be used to correlate other physiological characteristics. Such other data 320 may include heart rate (eg, Figure 2 ), elevated low frequency coupling—narrowband, elevated low frequency coupling—wideband, cyclic variation in heart rate (CVHR), various body motion measurements, body position fixation, and / or oxygen saturation, etc. These other data 320 can be temporally correlated with CPC state 310 in the same manner as heart rate data and CPC data are temporally correlated.

[0042] Analysis of CPC data and heart rate data is described below. Some aspects of the present disclosure relate to using this data to assess health status. In various aspects, the present disclosure includes determining one or more of the following: (1) average heart rate in a high frequency coupling (HFC) state, and whether the average heart rate is lower or higher than the average heart rate in a non-HFC state, (2) the trend of heart rate over time during sleep, (3) the response of heart rate to HFC, and / or (4) any spontaneous increase in heart rate during HFC.

[0043] According to some aspects of the present disclosure, a decrease in heart rate during high frequency coupling, when the CPC state is also referred to as stable NREM, may be a biological response indicative of normal health. In contrast, an increase in heart rate during the HFC state may be an indication of a health problem, such as poor cardiovascular function. As used herein, the term "high frequency coupled heart rate (HFC-HR)" refers to high frequency coupled heart rate analysis, which refers to and includes the use of CPC data and heart rate data disclosed herein to assess health conditions. In various embodiments, the high frequency coupled heart rate analyzes the level, trend, and pattern of heart rate during a sleep study period. The analysis may be performed for a complete sleep study period and / or for time periods within a sleep period, which may include time periods classified as HFC states (stable NREM sleep), LFC states (unstable NREM sleep), and / or vLFC states (wake or REM sleep). Unless otherwise stated, the disclosed analyses and operations may be performed by a computing system having a memory for storing data, a memory for storing instructions, and one or more processors for accessing data in the memory and executing instructions from the memory. This document will be combined later in this article. Fig.12 An exemplary computing system is described.

[0044] According to some aspects of the present disclosure, one aspect involves a complete sleep period and performing a multivariate linear regression to fit an ordinary least squares (OLS) model based on CPC data and heart rate data. The model includes heart rate as a response variable and includes heart rate as an explanatory variable / regressor (X HFC ) and includes as an explanatory variable / regressor (X vLFC) of the vLFC state. When CPC data are classified as HFC state, the HFC state explanatory variable X HFC The value of is 1, otherwise it is 0. When CPC data is classified as vLFC status, the vLFC status explanatory variable X vLFC The value is 1, otherwise it is 0. HFC and X vLFC When the values ​​of are all 0, the CPC state corresponds to the low frequency coupling (LFC) state. Therefore, the model is:

[0045] HR=β0+β HFC X HFC +β vLFC X vLFC +ε.

[0046] For n observations, HR is an n×1 vector of values ​​of the response variable (measured heart rate), and ε is an n×1 vector of error values; X HFC is an n×1 matrix with element values ​​of 0 or 1, where the element value of 0 or 1 corresponds to whether the CPC data is classified as HFC state; vLFC is an n×1 matrix with element values ​​of 0 or 1, where the element value 0 or 1 corresponds to whether the CPC data is classified as vLFC state. Ordinary least squares method is used to estimate the β0 coefficient, β HFC Coefficients and β vLFC Coefficient, where β HFC The coefficients are the coefficients of interest. Those skilled in the art understand various methods of applying multiple linear regression and ordinary least squares. The described methods and models are exemplary, and various variations are also considered to be within the scope of the present disclosure. For example, in various embodiments, the model can be specifically used to control low-frequency coupling. In various embodiments, when the following data are available, the model can be used to control elevated low-frequency coupling broadband, elevated low-frequency coupling narrowband, heart rate cyclical changes, oxygen saturation, body movement, body position fixation and / or other physiological characteristics described herein. Those skilled in the art will understand how to achieve such control and variation.

[0047] Continue to use the above model and refer to Figure 4 , the HFC state, LFC state, and vLFC state are mutually exclusive. Because the two explanatory variables correspond to the HFC state and the vLFC state, the zero value of both explanatory variables corresponds to the LFC state. Therefore, the β0 coefficient corresponds to the average heart rate in the LFC state and can be expressed as β LFC In HFC state, HR = β LFC +β HFC Indicates the average heart rate in the HFC state, β HFCThe coefficient indicates how much lower / higher the average heart rate during the HFC state is than the average heart rate during the LFC state. In the vLFC state, HR = β LFC +β vLFC Indicates the average heart rate in the vLFC state, β vLFC The coefficient indicates how much lower / higher the average heart rate during the vLFC state is than the average heart rate during the LFC state.

[0048] According to some aspects of the present disclosure, HFC The coefficient and / or another or all other coefficients provide a statistical significance / p-value. Those skilled in the art understand how to implement and use statistical significance / p-values. In various embodiments, the p-value can be used in the system to determine whether to display or not display the beta test in a report or display screen. HFC coefficient and / or other coefficients. If β is displayed HFC The coefficients and / or other coefficients, as described in the examples below, can be used by a clinician to determine or estimate a health condition or health problem using the p-values ​​of these coefficients.

[0049] Continue to refer Figure 4 , Figure 4 A graphical representation of the CPC classification and heart rate data for an individual during sleep is shown. Based on the application of multiple linear regression, the average heart rate during the LFC state was 60.3 BPM (β LFC =60.3), the average heart rate during the HFC state was 56.2 BPM (β LFC +β HFC =56.2). Therefore, on average, the mean heart rate in the HFC state is 4.1 BPM lower than the mean heart rate in the LFC state (β HFC =-4.1). In the absence of other indicators, these results may indicate a healthy condition.

[0050] In addition, the p-value for each value is zero (β HFC The p-value equals 0) indicates that the confidence of the average BPM reported by each CPC state is very high. In various embodiments, if the p-value is less than 0.05, the corresponding beta coefficient may be displayed. Therefore, in this example, the beta coefficient is displayed. HFC The displayed coefficients and / or other coefficients can be used by a clinician to learn the average heart rate of an individual and to make a diagnosis for the individual.

[0051] Figure 5 Figure 2 is a graphical representation of CPC classification and heart rate data during sleep for morbidly obese individuals. Based on the application of multiple linear regression as described above, the average heart rate during the LFC state was 62.6 BPM (β LFC =62.6), and the average heart rate during the HFC state was 64.8 BPM (βLFC +β HFC =64.8). Therefore, on average, the heart rate in the HFC state is 2.2 BPM higher than that in the LFC state (β HFC =2.2). In addition, the p-value for each value is zero (0), which indicates that the confidence in the average BPM reported for each CPC state is very high. Therefore, in this example, β HFC coefficients and / or other coefficients, the clinician can use the displayed coefficients to learn the average heart rate of the individual and make a diagnosis for the individual. Figure 5 These results in individuals with DM may indicate health problems or suboptimal health, such as poor cardiovascular function.

[0052] In various embodiments, clinicians can compare Figure 4 and Figure 5 The heart rate during the HFC state may be the most reliable indicator of resting heart rate, and the average resting heart rate difference between the two individuals mentioned above was 8.6 BPM. Therefore, the resting heart rate of obese individuals is 8.6 BPM higher than that of healthy individuals, which may be Figure 5 Individuals with a low p-value may have evidence of a health problem that places a significant negative strain on the cardiovascular system. A low p-value provides high confidence in the result and diagnosis.

[0053] According to some aspects of the present disclosure, the above analysis results of multiple individuals or part or all of the population can be collected and recorded. The recorded data can be used to compare the results of a specific individual with the results of other individuals, or with the results of part or all of the population.

[0054] Thus, one aspect related to complete sleep periods is described above for performing multiple linear regression to fit an ordinary least squares (OLS) model to CPC data and heart rate data.

[0055] Another aspect of the present disclosure also relates to complete sleep sessions, and to analyzing heart rate trends during sleep sessions. Figure 6 6 shows a graph of exemplary heart rate data over a complete sleep period. In various embodiments, linear regression and / or polynomial regression may be applied to the heart rate data to determine a line or curve 610 that best fits the data. If the fitted line or curve 610 has a downward sloping trend from the beginning of the sleep period to the end of the sleep period, such as Figure 6 As shown, the feature may indicate a health condition. However, if the fitted straight line or curve does not have a downward slope from the beginning of the sleep period to the end of the sleep period, the feature may indicate a health problem.

[0056] The following section will describe aspects of the time period categorized as a high frequency coupling (HFC) state in the present disclosure.

[0057] As mentioned above, the HFC state indicates that high frequency coupling dominates the CPC data. The time period corresponding to the HFC state can have different durations, such as Figure 3-5 As shown. In various embodiments, only HFC states with a sufficiently long duration are analyzed. In various embodiments, the duration threshold may be ten minutes. In various embodiments, the duration threshold may be other durations, such as longer than ten minutes. Generally speaking, a longer duration threshold will increase the certainty of the analysis because the analysis will be based on more data. In various embodiments, in certain circumstances, a duration threshold of less than 10 minutes may be used if it would produce a statistically significant output.

[0058] According to one aspect of the present disclosure, and with reference to Figure 7 , linear regression and / or polynomial regression, and / or any other regression model with the best fit can be used to fit various curves (linear curves and / or quadratic curves and / or other curves) according to the heart rate data in the HFC time period with a sufficiently long duration to determine the response of the heart rate to entering the HFC state. In various embodiments, the heart rate data can be used to fit any number of curves, and any type of curve can be used, including n-order polynomials, where n is an arbitrary number. For example, the heart rate data can be used to fit linear curves, concave curves, convex curves, and third-order polynomials. In various embodiments, other numbers of curves and combinations of other types of curves can be used. For all curves with a fit higher than a threshold value (such as a mean square error value), the best fit curve is selected. In the case where no curve has a good enough fit, for example, when the heart rate has an oscillating pattern, the response is marked as unknown. Figure 7 7 shows a curve 710 that may not meet the goodness-of-fit threshold because the heart rate data 720 oscillates within the HFC time period. Figure 7 The heart rate data 720 shown may be marked as having an unknown response to entering the HFC state. Figure 8 is an example of a curve with a better fit that meets the goodness of fit threshold.

[0059] Continue to refer Figure 8, when it is determined that curve 810 has the best fit effect, curve 810 is classified as a substantial decrease, decrease, flat, increase or a substantial increase. Curve 810 indicates the response of the heart rate to entering the HFC state. In various embodiments, based on the statistical significance / p-value of the parameter describing the slope or trend of curve 810, the response can be classified as a substantial increase or a substantial decrease. In various embodiments, p-value <0.05 (95% confidence) can be used as a threshold for statistical significance. In various embodiments, other p-values ​​can be used, such as less than 0.05. If the p-value of the parameter used to describe the slope / trend of the curve is less than 0.05, curve 810 will be considered to be a substantial increase or a substantial decrease. Otherwise, curve 810 will be classified as a decrease, flat or increase. In various embodiments, if the slope or trend of the curve is within the nominal range of zero, the curve can be marked as flat. In various embodiments, if the estimated slope is statistically no different from zero, the curve can be classified as flat without having a slope or trend of exactly zero. For example, in some cases, an estimated slope of 0.5 with a p-value of 0.3 may not be statistically different from a slope of zero, and in certain circumstances, the estimated slope of 0.5 may be classified as flat. For another example, in some cases, for certain circumstances, an estimated slope of 0.0000001 (or any other practically small number) may be statistically significant, and therefore, the estimated slope of 0.0000001 may not be classified as flat. In various embodiments, the slope coefficient and p-value may be provided to the user, and the user may consider it appropriate for diagnostic purposes. In various embodiments, if the best fit curve 810 reflects an increase in heart rate in response to entering the HFC state, such as Figure 8 As shown, this feature may indicate a health problem.

[0060] In various embodiments, the number of HFC time periods that exhibit an increase in heart rate in response to entering the HFC state may be counted and divided by the total number of HFC time periods in the sleep period to provide a measure of a suboptimal heart rate response. For example, if there are five HFC time periods that exhibit an increase in heart rate due to entering the HFC state, then the measure of a suboptimal heart rate response is 5 / 6 of the total six HFC time periods in the sleep period. The measures described herein are exemplary, and other measures of a suboptimal heart rate response may be considered.

[0061] According to one aspect of the present disclosure, HFC periods of sufficient duration may be analyzed to determine the spontaneous increase in heart rate during the HFC period as mentioned herein. As described above, the duration of multiple HFC periods may vary. According to some aspects of the present disclosure, and with reference to Figure 7, an HFC time period is divided into time intervals or time windows 730 of equal duration. In various embodiments, the duration of each time interval / time window 730 may be thirty seconds. In various embodiments, durations other than thirty seconds may be used. The average heart rate 720 is calculated for each time interval / time window 730 in the HFC time period. For each time window 730, if the increase in the heart rate average value from the current interval / window to the next interval / window exceeds a threshold, the time window is marked and the counter value is incremented. For example, referring to Figure 7 , the window of time t is marked as a spontaneous increase in heart rate 720. In various embodiments, the ratio between the count result and the total number of time windows 730 within the HFC time period can be calculated as a measure of the spontaneous increase in heart rate within the HFC time period. For example, if an HFC time period includes 24 time windows, and the value of the counter obtained from the above analysis is 13, then the calculated measure of the spontaneous increase in heart rate within the HFC time period will be 13 / 24. In various embodiments, if the calculated measure of spontaneous increase is greater than a threshold, the feature may indicate the presence of a health problem, such as indicating autonomous arousal. In various embodiments, when the measure of spontaneous increase is greater than a threshold, the measure of spontaneous increase can be displayed on a display screen. In various embodiments, even if the measure of spontaneous increase is less than a threshold, the measure of spontaneous increase can be displayed on a display screen.

[0062] Thus, the above description is some aspects related to complete sleep periods and some aspects related to time periods classified as HFC states.The described aspects and embodiments may be combined in various ways, including combinations described above or below, and combinations not explicitly described herein.

[0063] In various embodiments, the analysis results can be displayed in a report on a screen or paper. The report can include the average heart rate during the complete sleep period and / or the average heart rate (β) during the HFC state. LFC +β HFC ).

[0064] Fig. 9 and Fig.10 Reflects another combination of analyses. Fig. 9 is a graph showing CPC data 910, heart rate data 920, and CPC states 930. As described above, one analysis involves a complete sleep period and applying linear regression and / or polynomial regression to the heart rate data 920 to determine the line or curve that best fits the data. Fig. 9A fitted line 925 is shown having a downward sloping trend from the beginning of the sleep period to the end of the sleep period, which is characteristic of a healthy condition. As described above, another analysis involves each HFC period and determines the curve that best fits the heart rate data for each HFC period. Fig. 9 A best fit curve 928 is shown which exhibits a concave shape and slope in response to entering the HFC state, which is indicative of a healthy condition. Fig. 9 Both analyses in the are indicative of a healthy condition.

[0065] Fig.10 The same analysis in provides different results. Applying linear regression and / or polynomial regression to the heart rate data 1020 over the complete sleep period, Fig.10 A trend line 1025 is shown that does not decrease from the beginning of the sleep period to the end of the sleep period. In addition, the best fit curve 1028 for each HFC period shows that many HFC periods do not show a decrease in heart rate in response to entering the HFC state. These two analyses show that Fig.10 Individuals may have health problems, such as the development or progression of underlying cardiometabolic disease. Fig. 9 and Fig.10 The embodiments are exemplary, and other combinations of the various aspects described herein are contemplated.

[0066] Reference now Fig.11 , Fig.11A flow chart of a process according to some aspects of the present disclosure is shown. At block 1102, the process accesses cardiopulmonary coupling data and heart rate data from a memory. At block 1104, the process identifies multiple time periods in a sleep period that are classified as a high frequency coupling (HFC) state based on the cardiopulmonary coupling data. In various embodiments, the HFC state may already be included in the CPC data. In various embodiments, the CPC data may not indicate the CPC state, and the CPC state needs to be determined based on the CPC data. At block 1106, the process evaluates at least one feature of a portion of the heart rate data corresponding to the multiple time periods. In various embodiments, the feature may include whether the slope or trend of the heart rate data increases during the HFC time period, whether there is a decreased heart rate response to entering the HFC state, and / or whether the heart rate spontaneously increases during the HFC time period, as well as other contents described herein. At block 1108, the process evaluates the health status of the person based on the features of the portion of the heart rate data corresponding to the time period classified as the HFC state. For example, as previously described, if the slope or trend of the heart rate data increases during the HFC time period, this feature may be an indication of a health problem. If the heart rate drops when entering the HFC state, this may be an indication of a healthy condition. If the heart rate increases spontaneously during the HFC period, this may be an indication of a health problem, such as autonomous arousal. At box 1110, the process may evaluate at least one feature of the heart rate data corresponding to the complete sleep period. For example, as described above, the feature may include whether there is a decreasing heart rate trend or slope from the beginning of the sleep period to the end of the sleep period, and / or the difference between the heart rate average value in the HFC state and the heart rate average value in the LFC state, as well as other features described herein. At box 1112, the process may evaluate the health status of the time period based on the features corresponding to the complete sleep period. For example, if there is a decreasing heart rate trend or slope from the beginning of the sleep period to the end of the sleep period, the feature may be an indication of a healthy condition. For another example, if the heart rate average value in the HFC state is higher than the heart rate average value in the LFC state, this feature may be an indication of a health problem, such as poor cardiovascular function. Fig.11 The various operations of the present invention are exemplary and need not be performed in the order shown. In various embodiments, the operations may be performed in another order, or the various operations may be performed in parallel.

[0067] refer to Fig.12, the various aspects, embodiments, processes and operations described herein can be performed by a system 1200, which includes a memory 1210, a memory 1220, a processor 1230 and a display device 1240. The system 1200 can be a local computing system or a cloud computing system, or a part of the system can be local and another part of the system can be operated in the cloud. The memory 1210 includes data for the operations described herein, such as cardiopulmonary data, heart rate data, body motion recording data, posture data and / or oxygen saturation data, etc. The memory 1210 can be any kind of storage device, such as a magnetic memory, a solid-state memory, or an optical memory, etc. The memory 1220 includes machine instructions configured to perform the operations described herein. The memory 1220 can also include software for providing software as a service and / or software for enabling medical devices as a service. The memory 1220 can be any kind of electronic memory and can be a random access memory. In various embodiments, the data or instructions in the memory 1210 or the memory 1220 can also reside in other components. The processor 1230 accesses data from the storage 1210 and instructions / software from the memory 1220 to perform the operations and processes described herein. The processor 1230 can be any type of computing device, such as a CPU (Central Processing Unit), a GPU (Graphics Processor Unit), a microcontroller, a digital signal processor, a programmable logic device, a field programmable gate array or other computing device, or a combination of these devices. The processor 1230 can provide various analysis results to the display device 1240 to display these analysis results on the display device. In various embodiments, the display device can be an independent display device that is separated from other components, or the display device can be integrated into a single device together with other components. Fig.12 The system is exemplary and various variations are considered to be within the scope of the present disclosure.

[0068] The embodiments disclosed herein are examples of the present disclosure and can be implemented in various forms. For example, although some embodiments herein are described as separate embodiments, each embodiment herein can be combined with one or more other embodiments herein. The specific structural and functional details disclosed herein should not be interpreted as restrictive, but as the basis of the claims, and as a representative basis for teaching those skilled in the art to use the present disclosure in various ways with virtually any appropriate detailed structure. In the description of the entire drawings, similar reference numerals may refer to similar or identical elements.

[0069] Each of the phrases “in one embodiment,” “in an embodiment,” “in various embodiments,” “in some embodiments,” or “in other embodiments” may refer to one or more of the same or different embodiments according to the present disclosure. Phrases in the form “A or B” mean “(A), (B), or (A and B).” Phrases in the form “at least one of A, B, or C” mean “(A); (B); (C); (A and B); (A and C); (B and C); or (A, B, and C).”

[0070] Any method, procedure, algorithm or code described herein may be converted into or expressed in a programming language or computer program. As used herein, the terms "programming language" and "computer program" include any language used to specify instructions to a computer, and include (but are not limited to) the following languages ​​and their derivatives: assembly language, BASIC (Beginner's All-purpose Symbolic Instruction Code) language, batch file, BCPL (Basic Combined Programming Language) language, C language, C language, C++, Delphi language, Fortran (Formula Translation) language, Java language, JavaScript language, machine code, operating system command language, Pascal (Philips automatic sequence calculator) language, Perl (Practical Extraction and Reporting Language) language, PL1 (Programming Language No.1) language, scripting language, Visual Basic language, metalanguage that specifies the program itself, and all first, second, third, fourth, fifth or next generation computer languages. Databases and other data models, as well as any other metalanguages ​​are also included. No distinction is made between languages ​​that are interpreted, compiled, or using both compiled and interpreted methods. There is no distinction between a compiled version of a program and a source code version. Thus, a reference to a program, where the programming language may exist in more than one state (e.g., source state, compiled state, object state, or linked state) is a reference to any and all such states. A reference to a program may include the actual instructions and / or the intent of those instructions.

[0071] It should be understood that the foregoing description is merely an illustration of the present disclosure. Without departing from the present disclosure, those skilled in the art may devise various substitutions and modifications. Therefore, the present disclosure is intended to include all of these substitutions, modifications, and variations. The embodiments described with reference to the accompanying drawings are only used to illustrate certain examples of the present disclosure. Other elements, steps, methods, and techniques that are not substantially different from the elements, steps, methods, and techniques described above and / or in the appended claims are also intended to be included within the scope of the present disclosure.

[0072] The systems described herein may also utilize one or more controllers to receive various information and convert the received information to generate output. The controller may include any type of computing device, computing circuit, or any type of processor or processing circuit capable of executing a series of instructions stored in a memory. The controller may include multiple processors and / or multi-core central processing units (CPUs), and may include any type of processor, such as a microprocessor, a digital signal processor, a microcontroller, a programmable logic device (PLD), a field programmable gate array (FPGA), etc. The controller may also include a memory for storing data and / or instructions, which, when executed by one or more processors, causes one or more processors to execute one or more methods and / or algorithms.

[0073] Any method, program, algorithm or code described herein may be converted into or expressed in a programming language or computer program. The terms "programming language" and "computer program" used herein include any language used to specify instructions to a computer, and include (but are not limited to) the following languages ​​and their derivatives: assembly language, BASIC language, batch file, BCPL language, C language, C language, C++, Delphi language, Fortran language, Java language, JavaScript language, machine code, operating system command language, Pascal language, Perl language, PL1 language, scripting language, Visual Basic language, metalanguage that specifies the program itself, and all first, second, third, fourth, fifth or next generation computer languages. Also included are databases and other data models, and any other metalanguage. There is no distinction between languages ​​that are interpreted, compiled, or use both compilation and interpretation methods. There is no distinction between the compiled version and the source code version of the program. Therefore, a reference to a program, in which a programming language can exist in more than one state (e.g., source state, compiled state, object state, or link state) is a reference to any and all such states. A reference to a program may include the actual instructions and / or the intent of those instructions.

[0074] It should be understood that the foregoing description is merely an illustration of the present disclosure. Without departing from the present disclosure, those skilled in the art may devise various substitutions and modifications. Therefore, the present disclosure is intended to include all of these substitutions, modifications, and variations. The embodiments described with reference to the accompanying drawings are only used to illustrate certain examples of the present disclosure. Other elements, steps, methods, and techniques that are not substantially different from the elements, steps, methods, and techniques described above and / or in the appended claims are also intended to be included within the scope of the present disclosure.

Claims

1. A system for assessing health status, the system comprising: a memory, comprising cardiopulmonary coupling data and heart rate data of a person during a sleep period; one or more processors; as well as A memory coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to: accessing the cardiopulmonary coupling data and the heart rate data from the memory, identifying at least one time period in the sleep period that is classified as a high frequency coupling state based on the cardiopulmonary coupling data, evaluating at least one feature of the portion of the heart rate data corresponding to the at least one time period, the at least one feature comprising a feature of a curve fitted to the portion of the heart rate data, the feature comprising a slope or a trend, and Based on the at least one feature of the portion of the heart rate data corresponding to the at least one time period classified as a high frequency coupling state, a health condition of the person is assessed.

2. The system according to claim 1, wherein: When evaluating the at least one characteristic, the instructions, when executed by the one or more processors, cause the system to: determining a curve that fits the portion of the heart rate data by applying a curve fitting technique to the portion of the heart rate data; and The curve is classified as one of: a large increase, an increase, a flat, a decrease, or a large decrease as the at least one feature of the portion of the heart rate data corresponding to the at least one time period classified as a high frequency coupling state.

3. The system according to claim 2, wherein: When evaluating the health condition, when the instructions are executed by the one or more processors, the system estimates the health condition as having potential problems if the curve fitting the heart rate data within the at least one time period is classified as increasing or significantly increasing.

4. The system according to claim 1, wherein: When identifying the at least one time period in the sleep period, when the instructions are executed by the one or more processors, the system identifies a plurality of time periods in the sleep period that are classified as being in a high frequency coupling state.

5. The system according to claim 4, wherein: When evaluating the at least one characteristic, when the instructions are executed by the one or more processors, the system determines an average high-frequency coupling heart rate based on the portion of the heart rate data corresponding to the multiple time periods classified as high-frequency coupling states.

6. The system according to claim 5, wherein: When the instructions are executed by the one or more processors, the system is further caused to: Based on the cardiopulmonary coupling data, identifying a second plurality of time periods in the sleep period that are classified as low frequency coupling states; and An average low frequency coupling heart rate is determined based on portions of the heart rate data corresponding to the second plurality of time periods classified as being in a low frequency coupling state.

7. The system according to claim 6, wherein: When evaluating the health condition, when the instructions are executed by the one or more processors, the system estimates the health condition as having potential health problems if the average high-frequency coupled heart rate is greater than the average low-frequency coupled heart rate.

8. The system according to claim 6, wherein: When determining the average high-frequency coupling heart rate and determining the average low-frequency coupling heart rate, when the instructions are executed by the one or more processors, the system performs a multiple linear regression based on a model with heart rate as a response variable, high-frequency coupling state as an explanatory variable, and very-low-frequency coupling state as an explanatory variable.

9. The system according to claim 8, further comprising a display device, in, When the instructions are executed by the one or more processors, the system also displays the beta coefficient of the high-frequency coupling state from the multivariate linear regression model on the display device, wherein the beta coefficient indicates the difference between the average high-frequency coupling heart rate and the average low-frequency coupling heart rate.

10. The system according to claim 1, wherein: When the instructions are executed by the one or more processors, the system is further caused to determine a heart rate trend within the sleep period based on the heart rate data.

11. The system according to claim 10, wherein: When evaluating the health condition, when the instructions are executed by the one or more processors, the system estimates the health condition as having potential health problems if the heart rate trend during the sleep period does not decrease.

12. The system of claim 1, wherein: When evaluating the at least one characteristic, the instructions, when executed by the one or more processors, cause the system to: dividing the at least one time period exhibiting high frequency coupling into a plurality of time intervals of equal duration; determining an average heart rate for each of the plurality of time intervals based on a portion of the heart rate data corresponding to the plurality of time intervals; A pair of adjacent time intervals in which an increase in average heart rate in the plurality of time intervals exceeds a threshold is identified.

13. A computer-implemented method for assessing health status, the method comprising: Access to a person's cardiorespiratory coupling data and heart rate data during sleep; identifying, based on the cardiopulmonary coupling data, at least one time period in the sleep period that is classified as a high frequency coupling state; evaluating at least one feature of the portion of the heart rate data corresponding to the at least one time period, the at least one feature comprising a feature of a curve fitted to the portion of the heart rate data, the feature comprising a slope or a trend; as well as Based on the at least one feature of the portion of the heart rate data corresponding to the at least one time period classified as a high frequency coupling state, a health condition of the person is assessed.

14. The computer-implemented method of claim 13, wherein: Assessing the at least one characteristic comprises: Determining a curve that fits the portion of the heart rate data by applying a curve fitting technique to the portion of the heart rate data; and classifying the curve as one of: a substantial increase, an increase, a flattening, a decrease, or a substantial decrease as at least one feature of the portion of the heart rate data corresponding to the at least one time period classified as a high frequency coupling state.

15. The computer-implemented method of claim 14, wherein: Assess the health conditions, including: When the curve fitting the heart rate data within the at least one time period is classified as increasing or greatly increasing, the health condition is estimated to have a potential problem.

16. The computer-implemented method of claim 13, wherein: Identifying the at least one time period in the sleep period includes: A plurality of time periods in the sleep period that are classified as being in a high frequency coupling state are identified.

17. The computer-implemented method of claim 16, wherein: Assessing the at least one characteristic comprises: An average high frequency coupling heart rate is determined based on portions of the heart rate data corresponding to the plurality of time periods classified as being in a high frequency coupling state.

18. The computer-implemented method of claim 17, further comprising: identifying, based on the cardiopulmonary coupling data, a second plurality of time periods in the sleep period that are classified as low-frequency coupling states; as well as An average low frequency coupling heart rate is determined based on portions of the heart rate data corresponding to the second plurality of time periods classified as being in a low frequency coupling state.

19. The computer-implemented method of claim 18, wherein: Assess the health conditions, including: When the average high-frequency coupled heart rate is greater than the average low-frequency coupled heart rate, the health condition is estimated as having potential health problems.

20. The computer-implemented method of claim 18, wherein: Determining the average high-frequency coupled heart rate and determining the average low-frequency coupled heart rate comprises: A multivariate linear regression was performed based on a model with heart rate as the response variable, high-frequency coupling status as the explanatory variable, and very-low-frequency coupling status as the explanatory variable.

21. The computer-implemented method of claim 20, further comprising: A beta coefficient of the high frequency coupling state from the multiple linear regression model is displayed on a display device, wherein the beta coefficient indicates a difference between the average high frequency coupling heart rate and the average low frequency coupling heart rate.

22. The computer-implemented method of claim 13, further comprising: Based on the heart rate data, a heart rate trend during the sleep period is determined.

23. The computer-implemented method of claim 22, wherein: Assessment of health conditions includes: When the heart rate trend within the sleeping period does not decrease, the health condition is estimated as having a potential health problem.

24. The computer-implemented method of claim 13, wherein: Evaluating the at least one characteristic comprises: dividing the at least one time period exhibiting high frequency coupling into a plurality of time intervals of equal duration; determining an average heart rate for each of the plurality of time intervals based on a portion of the heart rate data corresponding to the plurality of time intervals; A pair of adjacent time intervals in which an increase in average heart rate in the plurality of time intervals exceeds a threshold is identified.

Images