Radar-based full-time in-bed state monitoring method, system and product

Abstract

The invention provides a radar-based full-time in-bed state monitoring method, system and product, and the method comprises the steps: carrying out the signal preprocessing of a collected radar echo signal, and obtaining target existence information and target physiological information; performing feature extraction on the target existence information and the target physiological information, and adaptively dividing a daytime period and a night period based on a feature extraction result; performing multi-mode sleep state staging based on the target existence information and the target physiological information in the night time period; performing daytime activity state classification based on the target existence information and the target physiological information in the daytime period; and generating a full-time in-bed state monitoring report based on the sleep state staging result and the daytime activity state classification result. According to the method provided by the invention, both long-term bedridden people and non-bedridden people can be considered, the day and night time periods are automatically switched, and the function that single radar equipment synchronously supports night sleep staging and daytime activity classification is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar, in particular to a radar-based all-time in-bed state monitoring method, system and product. BACKGROUND

[0002] With the increasing maturity of sensor technology, in-bed state monitoring as an important part of health monitoring is becoming more and more diversified. With the characteristics of non-contact technology, users do not need to wear equipment and do not interfere with normal life, which shows important value in health management, medical monitoring and other scenarios. At present, radar sensors, visual sensors and pressure sensors are the main sensors for in-bed state monitoring. For example, the technology of detecting body movement by using radar Doppler signal to judge sleep state has the limitations of single data source and not considering user daytime bed activity. The technology of analyzing sleep quality by using infrared camera combined with body sensing device is easily affected by clothing and bedding shielding and has privacy leakage problem. The technology of realizing sleep monitoring through bed body pressure sensor array is limited in monitoring range and cannot distinguish between day and night states. Especially for long-term bedridden population, the physiological signal characteristics of daytime wake-up bed are significantly different from those of night sleep stage. The traditional sleep staging monitoring method is easy to misjudge the wake-up bed state as sleep, resulting in distorted monitoring data and difficulty in meeting the accuracy requirements of all-weather health monitoring. SUMMARY

[0003] The purpose of the present application is to provide a radar-based all-time in-bed state monitoring method, system and product, which can at least solve the problems of limited monitoring range, inability to consider day and night states and poor applicability to bedridden population in related technologies.

[0004] To solve the above technical problems, the first aspect of the embodiments of the present application provides a radar-based all-time in-bed state monitoring method, comprising:

[0005] Signal preprocessing is performed on the collected radar echo signal to obtain target existence information and target physiological information; wherein the target physiological information at least includes body movement index, breathing frequency and heartbeat frequency;

[0006] Feature extraction is performed on the target existence information and the target physiological information, and day and night periods are adaptively divided based on the feature extraction result;

[0007] Multi-modal sleep state staging is performed based on the target existence information and the target physiological information in the night period;

[0008] Daytime activity state classification is performed based on the target existence information and the target physiological information in the daytime period;

[0009] generate a whole-time-in-bed state monitoring report based on the sleep state staging result and the daytime activity state classification result.

[0010] The second aspect of the embodiments of the present application provides a radar-based whole-time-in-bed state monitoring system, comprising:

[0011] a signal processing module, configured to perform signal preprocessing on the collected radar echo signals, to obtain target existence information and target physiological information; wherein the target physiological information at least includes a body movement index, a breathing frequency and a heartbeat frequency;

[0012] a feature extraction module, configured to perform feature extraction on the target existence information and the target physiological information, and to adaptively divide daytime periods and nighttime periods based on the feature extraction result;

[0013] a nighttime sleep staging module, configured to perform multi-modal sleep state staging based on the target existence information and the target physiological information in the nighttime periods, and to calculate a nighttime sleep score according to a sleep state staging result;

[0014] a daytime activity classification module, configured to perform daytime activity state classification based on the target existence information and the target physiological information in the daytime periods, and to calculate a daytime vitality index according to a daytime activity state classification result;

[0015] a report generation module, configured to generate a whole-time-in-bed state monitoring report based on the sleep state staging result and the daytime activity state classification result.

[0016] The third aspect of the present application provides a radar, comprising a memory and a processor, wherein the processor is configured to execute a computer program stored in the memory, and the processor, when executing the computer program, implements each step of the whole-time-in-bed state monitoring method described in the first aspect of the embodiments of the present application.

[0017] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, and the computer program, when executed by a processor, implements each step of the whole-time-in-bed state monitoring method described in the first aspect of the embodiments of the present application.

[0018] From the above, the embodiment of the present application first carries out signal preprocessing on the collected radar echo signal, obtains target existence information and target physiological information, wherein the target physiological information at least includes a body movement index, a breathing frequency and a heartbeat frequency, then carries out feature extraction on the target existence information and the target physiological information, and adaptively divides day time periods and night time periods based on the feature extraction result, next carries out multi-modal sleep state staging based on the target existence information and the target physiological information in the night time period, and carries out day time activity state classification based on the target existence information and the target physiological information in the day time period, and finally generates a full time period in bed state monitoring report based on the sleep state staging result and the day time activity state classification result. The full time period in bed state monitoring method proposed in the present application can take into account long-term bedridden and non-bedridden populations, automatically switch day and night periods, realize the function of synchronously supporting night sleep staging and day activity classification by a single radar device, and the method also has the characteristics of non-contact, does not need to wear any device, has high penetration, is not affected by clothes and bedding, and can effectively increase the monitoring range.

[0019] It should be understood that the content described in this part is not intended to identify key or important features of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the related art or the embodiments of the present application, the drawings needed to be used in the description of the related art or the embodiments of the present application will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and not all embodiments. Those skilled in the art can also obtain other drawings according to these drawings without creating any creative labor.

[0021] Figure 1 The flowchart of the full time period in bed state monitoring method provided by the embodiment of the present application is shown in the figure.

[0022] Figure 2 The state transition rule diagram of the sleep period in the full time period in bed state monitoring method provided by the embodiment of the present application is shown in the figure.

[0023] Figure 3 The day time activity state transition rule diagram in the full time period in bed state monitoring method provided by the embodiment of the present application is shown in the figure.

[0024] Figure 4 The detailed flowchart of the full time period in bed state monitoring method provided by the embodiment of the present application is shown in the figure.

[0025] Figure 5A program module schematic diagram of the whole period in-bed state monitoring system provided by the embodiment of the present application is provided.

[0026] Figure 6 A module block diagram of the radar provided by the embodiment of the present application is provided.

[0027] Figure 7 A module block diagram of the computer readable storage medium provided by the embodiment of the present application is provided. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present application more obvious and easy to understand, the present application will be described clearly and completely below in combination with the embodiments of the present application and the drawings, wherein the same or similar reference signs represent the same or similar elements or elements with the same or similar functions throughout. It should be understood that the embodiments of the present application described below are only used to explain the present application and do not limit the present application, that is, based on the various embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application. In addition, the technical features involved in the various embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0029] With the increasing maturity of sensor technology, in-bed state monitoring as an important part of health monitoring is becoming more and more diversified. With the characteristics of non-contact technology, users do not need to wear equipment and do not interfere with normal life, which shows important value in health management, medical monitoring and other scenes. At present, radar sensors, visual sensors and pressure sensors are the main sensors for in-bed state monitoring. For example, the technology of detecting body movement by using radar Doppler signal to judge sleep state has the limitations of single data source and not considering the user's daytime bed activity. The technology of analyzing sleep quality by using infrared camera combined with body sensing device is easily affected by clothing and bedding shielding and has the problem of privacy leakage. The technology of realizing sleep monitoring through bed body pressure sensor array is limited in monitoring range and cannot distinguish between day and night states.

[0030] Especially for long-term bedridden population, the physiological signal characteristics during daytime wake-up bed are significantly different from those during nighttime sleep stage. The traditional sleep staging monitoring method is easy to misjudge the wake-up bed state as sleep, resulting in distorted monitoring data and difficulty in meeting the accuracy requirements of all-weather health monitoring. Therefore, there is an urgent need in the technical field for a millimeter wave radar whole period in-bed state monitoring method to solve the problems of limited monitoring range, inability to consider day and night states and poor applicability to bedridden population in the prior art.

[0031] The present application provides a whole period in-bed state monitoring method, specifically, please refer to Figure 1 , Figure 1A flowchart of a radar-based whole-time in-bed state monitoring method provided in an embodiment of the present application is shown in FIG. 1. The whole-time in-bed state monitoring method includes the following steps 101-105.

[0032] In step 101, the radar echo signals collected are preprocessed to obtain target existence information and target physiological information. The target physiological information includes at least a body movement index, a breathing frequency, and a heartbeat frequency.

[0033] In the embodiment of the present application, the radar echo signals collected are preprocessed to obtain target existence information and target physiological information, including: preprocessing the original radar echo signals collected to obtain a first discrete echo signal; performing fast Fourier transform on the first discrete echo signal in a fast time dimension to obtain a second discrete echo signal; performing range dimension clutter suppression and coherent accumulation on the second discrete echo signal to obtain a third discrete echo signal; performing target detection based on the third discrete echo signal to obtain the target existence information; and obtaining the target physiological information based on the second discrete echo signal and the target existence information.

[0034] Specifically, first, the radar continuously emits electromagnetic wave signals into space, which are reflected by objects and then received by the radar receiver. After being amplified by a signal amplifier, mixed by a mixer, and sampled by an ADC, a discrete echo signal containing a range dimension, a velocity dimension, and an angle dimension, i.e., a first discrete echo signal, is obtained.

[0035] The first discrete echo signal can be represented as y(m, n, k), where m is a slow time dimension, representing the mth linear frequency modulation continuous wave signal; n is a fast time dimension, representing the nth sampling point; and k is an antenna dimension, representing the received signal of the kth channel.

[0036] Then, based on the first discrete echo signal, fast Fourier transform is first performed on the fast time dimension to convert the signal from the time domain to the frequency domain, obtaining a second discrete echo signal. The second discrete echo signal, i.e., a slow time-range-antenna dimension signal, can be represented as y'(m, r, k), where r ∈ [1, N] represents a range unit sample.

[0037] Further, range dimension clutter suppression and coherent accumulation are performed on the second discrete echo signal. By using a sliding average algorithm or a time domain accumulation method, environmental noise and non-target echo interference are filtered out, obtaining an enhanced signal after clutter suppression and coherent accumulation, i.e., a third discrete echo signal, which can be represented as

[0038] Finally, the signal y'(m, r, k) is used to perform target detection to obtain the target existence information. The human target detection is performed on the distance dimension to obtain target existence information and a location where the target is located. Then, the signal y'(m, r, k) is further processed to obtain physiological signals such as a body movement index, a breathing frequency, and a heartbeat frequency.

[0039] Further, in the embodiment of the present application, the target detection based on the third discrete echo signal comprises: obtaining a Doppler power spectrum of the third discrete echo signal; performing peak extraction on the Doppler power spectrum based on a constant false alarm rate algorithm, and performing target detection on the extracted peak points to obtain effective target points.

[0040] Specifically, the third discrete echo signal contains distance dimension r, velocity dimension v, and antenna dimension k information, and is taken as The Doppler power spectrum can be obtained by taking the power spectrum of the signal. The Doppler power spectrum can directly reflect whether there is a target in the radar detection range and the distance and velocity information of the target by analyzing the Doppler power spectrum of the signal. Then, a peak value in the power spectrum is extracted by using a constant false alarm rate (CFAR) algorithm. The constant false alarm rate algorithm specifically includes a cell average constant false alarm rate CA-CFAR, an ordered statistics constant false alarm rate OS-CFAR, and the like. By setting a detection threshold value matched with the intensity of environmental clutter, the peak value corresponding to the real target is screened out, noise and interference are excluded, and effective target points are obtained, so as to determine the existence, preliminary location, and velocity parameter of the target.

[0041] It can be understood that the embodiment of the present application selects a multi-channel millimeter wave radar for bed state monitoring in a second-level full-time period. The radar can be a single-transmit multi-receive radar or a multi-transmit multi-receive radar, which is not limited in the present application.

[0042] In step 102, feature extraction is performed on the target existence information and the target physiological information, and day and night time periods are adaptively divided based on the feature extraction result.

[0043] In the traditional time division, division is usually performed according to fixed rules, for example, 6:00 to 18:00 is divided into daytime, and 0:00 to 6:00 and 18:00 to 24:00 are divided into night. However, the living and working habits of each individual are not the same, and the living and working habits of the same individual may also have large differences every day. Therefore, the present scheme performs feature extraction based on the target existence information and the target physiological information, obtains the living and working characteristics of the individual through the feature extraction result, dynamically divides the day and night time periods through logical calculation, and realizes adaptive matching to the living and working habits of the user.

[0044] In the embodiment of the present application, the target presence information and the target physiological information are subjected to feature extraction, and the day time period and the night time period are adaptively divided based on the feature extraction result, including: minute-level feature extraction is performed on the target presence information and the target physiological information; the falling asleep time and the waking up time of the day are identified through a preset determination condition based on the minute-level feature extraction result; the falling asleep time and the waking up time of the previous day are obtained, and the time period is adaptively divided in combination with the falling asleep time and the waking up time of the day, to obtain the day time period and the night time period of the day.

[0045] Specifically, the target presence information indicates whether the target exists, and includes position information corresponding to the target if the target exists, and the target physiological information at least includes a body motion index, a breathing frequency, and a heartbeat frequency. Minute-level features of these information are extracted, and the falling asleep time and the waking up time of the day are identified through a preset determination condition. When the minute-level features of the target position, the body motion index, the breathing frequency, and the heartbeat frequency are all lower than the corresponding threshold value, the falling asleep time is determined; when the above features are all higher than the corresponding threshold value, the waking up time is determined. The falling asleep time and the waking up time of the previous day are further obtained, and the falling asleep / waking up time of the day and the preset minimum starting time of the day (such as 6 am) and the maximum starting time of the night (such as 10 pm) are combined, and the time period is adaptively divided through logical calculation: the maximum value of the waking up time of the previous day and the minimum starting time of the day is taken as the starting time of the day time period of the day, the minimum value of the falling asleep time of the previous day and the maximum starting time of the night is taken as the ending time of the day time period and the starting time of the night time period, and the maximum value of the waking up time of the day and the minimum starting time of the day is taken as the ending time of the night time period, so as to realize dynamic adaptive division of the day and night time periods, effectively adapt to the work and rest habits of different users, and improve the accuracy of whole-time monitoring.

[0046] In detail, the falling asleep time T ss is identified based on the minute-level features:

[0047] T ss = min{t|[P s (t)<T Ps ]∧[M s (t)<T Ms ]∧[R s (t)<T Rs ]∧[H s (t)<T Hs ]}

[0048] Wherein, ∧ represents logical and judgment, P s (t), M s (t), R s (t) and H s (t) are the minute-level features of the target position, the body motion index, the breathing frequency and the heartbeat frequency, respectively.

[0049] Based on minute-level features on wake-up time T se Perform recognition:

[0050] T se = max{t | [P s (t)>T Pw ]∧[M s (t)>T Mw ]∧[R s (t)>T Rw ]∧[H s (t)>T Hw ]}

[0051] Based on sleep-wake time markers T ss and T se , define the day, night period demarcation points as follows:

[0052] T1 = max{T' se , T mn}

[0053] T2 = min{T' ss , T en}

[0054] T3 = max{T se , T mn}

[0055] Wherein, T1 is the starting time of the day period, T2 is the end time of the day period and the starting time of the night period, T3 is the end time of the night period, T' se , T se are the wake-up time of the previous day and the next morning respectively, T' ss is the sleep-in time of the previous night, T mn is the minimum allowed starting time of the day period, T en is the maximum allowed starting time of the night period.

[0056] Step 103, based on the target presence information and the target physiological information in the night period, multi-modal sleep state staging is performed.

[0057] In the embodiment of the present application, the multi-modal sleep state staging based on the target presence information and the target physiological information in the night period comprises: acquiring first sleep feature information based on the target presence information and the target physiological information in the night period; wherein the first sleep feature information comprises body movement frequency, feature window length, body movement index mean in a feature extraction window, medium frequency and high frequency body movement threshold; dividing the night period into wake period and sleep period based on the first sleep feature information and first decision rule; acquiring second sleep feature information based on the target presence information and the target physiological information in the sleep period; wherein the second sleep feature information comprises basic respiration rate, basic body movement index, basic respiration rate variance, observation window information, decision threshold, body movement index in the observation window, respiration rate variance in the observation window and respiration frequency in the observation window; dividing the sleep period into light sleep period, deep sleep period and REM period based on the second sleep feature information and second decision rule.

[0058] Specifically, when the multi-modal sleep state staging is performed based on the target position in the target presence information and the body movement index, respiration frequency and heartbeat frequency in the target physiological information in the night period, first sleep feature information is acquired, including body movement frequency, feature window length, body movement index mean in a feature extraction window, and medium frequency and high frequency body movement threshold. Through the first decision rule, the night period is divided into wake period and sleep period in combination with feature combinations such as standard deviation of body movement index and medium frequency body movement cumulative value.

[0059] When in the sleep period, second sleep feature information is further acquired, including basic respiration rate, basic body movement index, basic respiration rate variance, observation window range, decision threshold, and body movement index, respiration rate variance and respiration frequency in the observation window. Based on the second decision rule, the sleep period is dynamically divided into light sleep period, deep sleep period and rapid eye movement (REM) period by analyzing multi-modal features such as the ratio of body movement index to basic body movement index, the ratio of respiration rate variance to basic respiration rate variance, and respiration frequency fluctuation amplitude, so as to realize accurate sleep state staging based on multi-dimensional physiological signals.

[0060] In detail, for the night period, when the target presence is detected, features are first extracted for wake period and sleep period judgment:

[0061]

[0062]

[0063] wherein m i is the body movement frequency at the i moment, t is the current moment, n is the feature window length, μ t is the body movement index mean in the feature extraction window, ∈1 and ∈2 are medium frequency and high frequency body movement threshold respectively, is the Heaviside step function, is the Kronecker delta function, defined as:

[0064]

[0065] When the features satisfy one of the following transition conditions, then transition from sleep to wake:

[0066] Condition 1:

[0067] (F std >∈3)∧(F mid <∈4)

[0068] Condition 2:

[0069] (F max >∈5)∧(F std >∈6)∧(F mid <∈7)

[0070] Condition 3:

[0071] F std >∈8

[0072] Condition 4:

[0073] F mid <∈9

[0074] Condition 5:

[0075] (F std >∈3)∧(F min <∈4)∧(F max <∈ 10 )

[0076] where ∈3-∈ 10 are reference thresholds.

[0077] If the current is sleep, the effective features are extracted for dynamic judgment and conversion of sleep, which is divided into light sleep, deep sleep and REM period. The extracted features are as follows:

[0078]

[0079] where ρ is the observation window, b(l) is the body motion index in the observation window, ξ1 is the judgment threshold, g(l) is the respiratory rate variance in the observation window, is the basic respiratory rate variance, h(l) is the respiratory frequency in the observation window, is the basic respiratory rate, is the basic body motion index.

[0080] The state transition rule in the sleep period is as follows:

[0081]

[0082] wherein, ξ2-ξ7 are reference thresholds, and α1-α7 are decision thresholds. H1 represents transition to the wake period, H2 represents transition to the light sleep period, H3 represents transition to the deep sleep period, and H4 represents transition to the REM period.

[0083] In the embodiment of the present application, the method further comprises: obtaining third sleep feature information based on the target existence information and the target physiological information in a preset sliding time window; wherein the third sleep feature information comprises a body movement index, a respiratory frequency, a change trend coefficient of a heartbeat frequency, and a length of the preset sliding time window in the preset sliding time window; and dividing the sleep period into a normal sleep state and a dozing state based on the third sleep feature information and a third decision rule.

[0084] Specifically, as shown in Figure 2 the present scheme further performs in-depth analysis on the sleep period state based on a preset sliding time window. By setting a sliding time window, the daily average change trend coefficient of the body movement index, the respiratory frequency, and the heartbeat frequency in the window is calculated. The coefficient represents the relative change amplitude of the current value and the starting value of the window. When at least τ days in the time window meet one of the following conditions, it is determined that the dozing state is entered: the daily average of the body movement index is lower than the preset threshold and the change trend coefficient is negative, the daily average of the respiratory frequency is outside the normal range and the trend coefficient is continuously decreasing, and the daily average of the heartbeat frequency is abnormal and the trend coefficient is decreasing. Conversely, when at least τ days in the window meet the conditions that the daily average of the body movement index rises above the threshold and the trend coefficient is positively increasing, the respiratory / heartbeat frequency returns to the normal range and the trend coefficient stabilizes, it is determined that the normal sleep state is restored, thereby realizing dynamic monitoring and early warning of the abnormal state of the sleep period.

[0085] In detail, first, the change trend coefficient of the body movement index, the respiratory frequency, and the heartbeat frequency in the past period of time is calculated through the sliding time window:

[0086]

[0087] wherein, the daily average of the body movement index, the respiratory frequency, and the heartbeat frequency, respectively, and k is the length of the time window.

[0088] When one of the following conditions is met, the dozing state is entered:

[0089] Condition 1: At least τ days in the time window meet m1 and and

[0090] Condition 2: at least τ days within the time window satisfy r1 Condition 3: at least τ days within the time window satisfy and

[0091] Condition 2: at least τ days within the time window satisfy h1 Condition 3: at least τ days within the time window satisfy and

[0092] When the following conditions are met simultaneously, the normal sleep state is resumed:

[0093] Condition 1: at least τ days within the time window satisfy m2 Condition 2: at least τ days within the time window satisfy and

[0094] Condition 2: at least τ days within the time window satisfy r2 Condition 3: at least τ days within the time window satisfy and

[0095] Condition 3: at least τ days within the time window satisfy h2 Condition 2: at least τ days within the time window satisfy and

[0096] Step 104, daytime activity state classification is performed based on the target presence information and the target physiological information within the daytime period.

[0097] For long-term bedridden people, they are usually awake during the day but may remain in a continuous bedridden state, at which time their physiological signal characteristics such as respiratory rate, heart rate fluctuation, and body movement amplitude will be significantly different from the sleep stage at night. If the traditional sleep staging monitoring method is used, it is easy to misjudge such an awake bedridden state as sleep, resulting in distorted monitoring data. Therefore, an independent daytime state detection system needs to be constructed to accurately distinguish between different states such as awake in bed, a little activity, frequent activity, etc., to meet the accuracy requirements of all-weather health monitoring.

[0098] In the embodiments of the present application, the daytime activity state classification based on the target presence information and the target physiological information within the daytime period includes: quantifying the body movement index within the daytime period to obtain a body movement amplitude level; and dividing the daytime activity state into no one state, rest state, a little activity state, and frequent activity state based on the target presence information, the body movement amplitude level, and a fourth determination rule.

[0099] Specifically, for the daytime period, the target presence detection result and the body movement index are used to classify four states of getting out of bed, resting, a little activity, and frequent activity. First, the body movement index is quantified, and the body movement amplitude is divided into three levels of small, medium, and large and is labeled as 0, 1, and 2, respectively:

[0100]

[0101] wherein m is a body motion index, is a body motion index quantization result, T m1 , T m2 and T m3 are thresholds for dividing three body motion amplitude levels of small, medium and large.

[0102] Four states of no one, rest, a little activity and frequent activity are divided, and the conversion logic between states is defined to realize daytime activity classification. The activity state conversion rules are shown in the following figure. Among them, is the sum of body motion index quantization in the last n seconds, T s1 , T s2 , T s3 and T s4 are state conversion thresholds. As shown in the following figure, the conversion logic between states is: Figure 3

[0103] (1) When the target existence detection result of the previous moment is "no one" and the current moment is "someone", the activity state is converted from "off bed" to "a little activity"; otherwise, the original state is maintained.

[0104] (2) When the activity state of the previous moment is "a little activity", if the target existence detection result of the current moment is "no one", the activity state is converted from "a little activity" to "off bed"; if M s1 , the activity state is converted from "a little activity" to "rest"; if M≥T s3 , the activity state is converted from "a little activity" to "rest"; otherwise, the original state is maintained.

[0105] (3) When the activity state of the previous moment is "rest", if M≥T s2 , the activity state is converted from "rest" to "a little activity"; otherwise, the original state is maintained.

[0106] (4) When the activity state of the previous moment is "frequent activity", if the target existence detection result of the current moment is "no one", the activity state is converted from "frequent activity" to "off bed"; if M s4 , the activity state is converted from "frequent activity" to "a little activity"; otherwise, the original state is maintained.

[0107] Step 105, generating a whole period in-bed state monitoring report based on the sleep state staging result and the daytime activity state classification result.

[0108] ​In the embodiments of the present application, the generating of the whole period in-bed state monitoring report based on the sleep state staging result and the daytime activity state classification result comprises: calculating a night sleep score based on the sleep state staging result and the target physiological information, and calculating a daytime vitality index based on the daytime activity state classification result and the target physiological information; and generating the whole period in-bed state monitoring report based on the night sleep score and the daytime vitality index.

[0109] Specifically, the night sleep score is composed of 12 indexes: respiratory frequency, heartbeat frequency, body movement index, target position, activity state, deep sleep duration, light sleep duration, REM duration, sleep duration of wakefulness, sleep duration of wakefulness, turning over times, and getting out of bed times. Based on the weighted integration analysis of multi-dimensional sleep indexes, the standardized night sleep score is calculated.

[0110] Similarly, the daytime vitality index is composed of 12 indexes: respiratory frequency, heartbeat frequency, body movement index, target position, light activity duration, light activity duration ratio, frequent activity duration, frequent activity duration ratio, light activity times, frequent activity times, large body movement index, and total in-bed duration. Based on the weighted integration analysis of multi-dimensional vitality indexes, the standardized daytime vitality index is calculated.

[0111] Finally, the night sleep score, the daytime vitality index, and corresponding sleep staging details, activity classification duration statistics, and other information are integrated to generate the whole period monitoring report, which intuitively presents the day and night health status data to the user, and provides a quantitative health evaluation basis for health management institutions.

[0112] From the above, the embodiment of the application first performs signal preprocessing on the collected radar echo signal to obtain target existence information and target physiological information, wherein the target physiological information at least includes a body movement index, a breathing frequency and a heartbeat frequency, then performs feature extraction on the target existence information and the target physiological information, and adaptively divides day time periods and night time periods based on the feature extraction result, next performs multi-modal sleep state staging based on the target existence information and the target physiological information in the night time period, and performs day time activity state classification based on the target existence information and the target physiological information in the day time period, and finally generates a full time period in bed state monitoring report based on the sleep state staging result and the day time activity state classification result. The full time period in bed state monitoring method proposed in the application first uses a non-contact radar monitoring technology, does not need to wear any equipment, improves the use comfort and monitoring continuity, and is suitable for long-term health monitoring; only outputs analysis results, does not store original radio frequency signals or image data, and protects user privacy; secondly, can cover the full time period, synchronously supports night sleep staging and day time activity classification, automatically switches day and night modes without manual operation, effectively makes up for the deficiency of the prior art in day time monitoring of long-term bedridden people, and takes into account the use requirements of long-term bedridden and non-bedridden people; in addition, the scheme can also generate detailed night and day reports, as well as sleep scores, vitality indexes and other data, which not only helps users to intuitively master their own health status, but also provides strong data support for medical personnel and health management institutions to develop scientific health management schemes, and improves the value of health management.

[0113] It should be understood that the size of the serial number of each step in the embodiment does not mean the order of execution of the steps, and the execution order of each step should be determined according to its function and inherent logic, and should not constitute the only limitation on the implementation process of the embodiment of the application.

[0114] In summary, the detailed process of the full time period in bed state monitoring method according to the embodiment of the application can be referred to Figure 4 , specifically as follows:

[0115] Step 401, performing signal preprocessing on the collected radar echo signal to obtain target existence information and target physiological information;

[0116] Step 402, performing minute-level feature extraction on the target existence information and the target physiological information;

[0117] Step 403, identifying the sleep-in time and the wake-up time of the day based on the minute-level feature extraction result through a preset determination condition;

[0118] Step 404, obtaining the sleep-in time and the wake-up time of the previous day, and adaptively dividing time periods in combination with the sleep-in time and the wake-up time of the day to obtain the day time period and the night time period of the day;

[0119] Step 405, obtaining first sleep feature information based on the target existence information and the target physiological information in the night period, and dividing the night period into a wake period and a sleep period based on the first sleep feature information and a first determination rule;

[0120] Step 406, obtaining second sleep feature information based on the target existence information and the target physiological information in the sleep period, and dividing the sleep period into a light sleep period, a deep sleep period and a REM period based on the second sleep feature information and a second determination rule;

[0121] Step 407, obtaining third sleep feature information based on the target existence information and the target physiological information in a preset sliding time window, and dividing the sleep period into a normal sleep state and a dozing state based on the third sleep feature information and a third determination rule;

[0122] Step 408, quantifying a body movement index in the daytime period to obtain a body movement amplitude level, and dividing a daytime activity state into an unoccupied state, a resting state, a little activity state and a frequent activity state based on the target existence information, the body movement amplitude level and a fourth determination rule;

[0123] Step 409, calculating a night sleep score based on the sleep state staging result and the target physiological information, and calculating a daytime vitality index based on the daytime activity state classification result and the target physiological information;

[0124] Step 410, generating a whole-period-in-bed state monitoring report based on the night sleep score and the daytime vitality index.

[0125] The more detailed processes of each of steps 401-410 can be found in the foregoing descriptions of the related parts, which will not be repeated here.

[0126] Please refer to Figure 5 , Figure 5 The radar-based whole-period-in-bed state monitoring system provided by the embodiments of the present application. The system can be used to implement the whole-period-in-bed state monitoring method related by the embodiments of the present application. The radar-based whole-period-in-bed state monitoring system 500 mainly includes:

[0127] The signal processing module 501 is configured to perform signal preprocessing on the collected radar echo signal, and obtain target existence information and target physiological information; wherein the target physiological information at least includes a body movement index, a breathing frequency and a heartbeat frequency;

[0128] The feature extraction module 502 is configured to perform feature extraction on the target existence information and the target physiological information, and adaptively divide a daytime period and a night period based on the feature extraction result;

[0129] The night sleep staging module 503 is configured to perform multi-modal sleep state staging based on the target existence information and the target physiological information in the night period, and calculate a night sleep score according to the sleep state staging result.

[0130] The daytime activity classification module 504 is configured to perform daytime activity state classification based on the target existence information and the target physiological information in the daytime period, and calculate a daytime vitality index according to the daytime activity state classification result.

[0131] The report generation module 505 is configured to generate a whole-period-in-bed state monitoring report based on the sleep state staging result and the daytime activity state classification result.

[0132] In some embodiments of the present embodiment, when the signal processing module 501 performs the function of performing signal preprocessing on the collected radar echo signal to obtain the target existence information and the target physiological information, the signal processing module 501 is configured to: perform preprocessing on the collected original radar echo signal to obtain a first discrete echo signal; perform fast Fourier transform on the first discrete echo signal in a fast time dimension to obtain a second discrete echo signal; perform distance dimension clutter suppression and coherent accumulation on the second discrete echo signal to obtain a third discrete echo signal; perform target detection based on the third discrete echo signal to obtain the target existence information; and obtain the target physiological information based on the second discrete echo signal and the target existence information.

[0133] Further, in some embodiments of the present embodiment, when the signal processing module 501 performs the function of performing target detection based on the third discrete echo signal, the signal processing module 501 is specifically configured to: obtain a Doppler power spectrum of the third discrete echo signal; perform peak value extraction on the Doppler power spectrum based on a constant false alarm rate algorithm, and perform target detection on the extracted peak value points to obtain effective target points.

[0134] Further, in some embodiments of the present embodiment, when the feature extraction module 502 performs the function of performing feature extraction on the target existence information and the target physiological information, and adaptively dividing the daytime period and the night period based on the feature extraction result, the feature extraction module 502 is configured to: perform minute-level feature extraction on the target existence information and the target physiological information; identify the sleep-in time and the wake-up time of the day based on the minute-level feature extraction result through a preset judgment condition; obtain the sleep-in time and the wake-up time of the previous day, and combine the sleep-in time and the wake-up time of the day to adaptively divide the period to obtain the daytime period and the night period of the day.

[0135] In some embodiments of the present embodiment, the nighttime sleep staging module 503, when performing the function of performing multi-modal sleep state staging based on the target presence information and the target physiological information in the nighttime period, is configured to: obtain first sleep feature information based on the target presence information and the target physiological information in the nighttime period; wherein the first sleep feature information comprises body movement frequency, feature window length, body movement index mean value in a feature extraction window, medium frequency, and high frequency body movement threshold; divide the nighttime period into wake period and sleep period based on the first sleep feature information and a first determination rule; obtain second sleep feature information based on the target presence information and the target physiological information in the sleep period; wherein the second sleep feature information comprises basal respiration rate, basal body movement index, basal respiration rate variance, observation window information, decision threshold, observation window body movement index, observation window respiration rate variance, and observation window respiration frequency; divide the sleep period into light sleep period, deep sleep period, and REM period based on the second sleep feature information and a second determination rule.

[0136] In some embodiments of the present embodiment, the nighttime sleep staging module 503 in the whole-period-in-bed state monitoring system 500 is further configured to: obtain third sleep feature information based on the target presence information and the target physiological information in a preset sliding time window; wherein the third sleep feature information comprises variation trend coefficients of body movement index, respiration frequency, and heartbeat frequency in the preset sliding time window, and length of the preset sliding time window; divide the sleep period into normal sleep state and dozing state based on the third sleep feature information and a third determination rule.

[0137] In some embodiments of the present embodiment, the daytime activity classification module 504, when performing the function of classifying daytime activity state based on the target presence information and the target physiological information in the daytime period, is configured to: quantify body movement index in the daytime period to obtain body movement amplitude level; divide the daytime activity state into no one state, rest state, little activity state, and frequent activity state based on the target presence information, the body movement amplitude level, and a fourth determination rule.

[0138] In some embodiments of the present embodiment, the report generation module 505, when performing the function of generating whole-period-in-bed state monitoring report based on the sleep state staging result and the daytime activity state classification result, is configured to: calculate nighttime sleep score based on the sleep state staging result and the target physiological information, and calculate daytime activity index based on the daytime activity state classification result and the target physiological information; generate whole-period-in-bed state monitoring report based on the nighttime sleep score and the daytime activity index.

[0139] In detail, the modules in the radar-based all-time bed state monitoring system 500 provided by the embodiments of the present application use the same technical means as the all-time bed state monitoring method in the above Figure 1 , and can produce the same technical effects, which will not be described here again.

[0140] Please refer to Figure 6 , Figure 6 for the module block diagram of the radar provided by the embodiments of the present application.

[0141] As shown in Figure 6 , the embodiments of the present application further provide a radar, which can be used to implement the all-time bed state monitoring method in the foregoing embodiments, and the radar comprises a memory 601 and at least one processor 602; the memory 601 is configured to store at least one program, and when the at least one program is executed by the at least one processor 602, the at least one processor 602 executes the all-time bed state monitoring method provided by the embodiments of the present application.

[0142] Please refer to Figure 7 , Figure 7 for the module block diagram of the computer readable storage medium provided by the embodiments of the present application.

[0143] As shown in Figure 7 , the embodiments of the present application further provide a computer readable storage medium 700, and the computer readable storage medium 700 stores executable instructions 710, and the executable instructions 710 are executed to execute the all-time bed state monitoring method provided by the embodiments of the present application.

[0144] The steps of the methods or algorithms described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can be located in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0145] In the embodiments described above, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded on a computer and executed, all or part of the processes or functions described in the present application are produced. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available media can be magnetic media (such as floppy disk, hard disk, magnetic tape), optical media (such as DVD), or semiconductor media (such as Solid State Disk) and the like.

[0146] It should be noted that each of the embodiments in the present application is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For product type embodiments, since they are similar to method type embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method type embodiment.

[0147] It should also be noted that in the present application, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or equipment including the element.

[0148] The above description of disclosed embodiments allows a skilled person to implement or use the present content. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined can be applied to other embodiments without departing from the spirit or scope of the content. Thus, the content is not to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed.

Claims

1. A radar-based method for monitoring bed status throughout the entire time period, characterized in that, include: The collected radar echo signals are preprocessed to obtain target presence information and target physiological information; wherein, the target physiological information includes at least body movement index, respiratory rate and heart rate. Feature extraction is performed on the target's existence information and physiological information, and daytime and nighttime periods are adaptively divided based on the feature extraction results; Multimodal sleep state staging is performed based on the target's presence information and physiological information during the nighttime period; Classify daytime activity status based on the target's presence information and physiological information during the daytime period; A full-time bedtime monitoring report is generated based on the sleep state staging results and the daytime activity state classification results.

2. The method for monitoring bedside status throughout the entire time period according to claim 1, characterized in that, The process of preprocessing the acquired radar echo signals to obtain target presence information and target physiological information includes: The acquired raw radar echo signal is preprocessed to obtain the first discrete echo signal; A second discrete echo signal is obtained by performing a fast Fourier transform on the first discrete echo signal in the fast time dimension. Based on the second discrete echo signal, range clutter suppression and coherent accumulation are performed to obtain the third discrete echo signal; Target detection is performed based on the third discrete echo signal to obtain target presence information; Target physiological information is obtained based on the second discrete echo signal and the target presence information.

3. The method for monitoring bedside status throughout the entire time period according to claim 2, characterized in that, The target detection based on the third discrete echo signal includes: Obtain the Doppler power spectrum of the third discrete echo signal; The peak values ​​of the Doppler power spectrum are extracted based on the constant false alarm rate algorithm, and the extracted peak values ​​are used for target detection to obtain effective target points.

4. The method for monitoring bedside status throughout the entire time period according to claim 1, characterized in that, The step of extracting features from the target's existence information and physiological information, and adaptively dividing daytime and nighttime periods based on the feature extraction results, includes: Minute-level feature extraction is performed on the target's existence information and its physiological information; Based on minute-level feature extraction results, the time of falling asleep and waking up on the same day are identified using preset judgment conditions; The system obtains the time when the user fell asleep and the time when the user woke up the previous day, and then uses these two times to adaptively divide the time period into daytime and nighttime periods for the current day.

5. The method for monitoring bedside status throughout the entire time period according to claim 1, characterized in that, The multimodal sleep state staging based on the target's presence information and physiological information during the nighttime period includes: First sleep feature information is obtained based on the target presence information and the target physiological information during the nighttime period; wherein, the first sleep feature information includes body movement frequency, feature window length, mean body movement index within the feature extraction window, and mid-frequency and high-frequency body movement thresholds; Based on the first sleep characteristic information and the first determination rule, the nighttime period is divided into a waking period and a sleep period; Second sleep characteristic information is obtained based on the target presence information and target physiological information during the sleep period; wherein, the second sleep characteristic information includes baseline respiratory rate, baseline body movement index, baseline respiratory rate variance, observation window information, decision threshold, body movement index within the observation window, respiratory rate variance within the observation window, and respiratory frequency within the observation window; Based on the second sleep characteristic information and the second determination rule, the sleep period is divided into light sleep, deep sleep and REM sleep.

6. The method for monitoring bedside status throughout the entire time according to claim 5, characterized in that, The method further includes: The third sleep characteristic information is obtained based on the target presence information and the target physiological information within a preset sliding time window; wherein, the third sleep characteristic information includes the trend coefficients of the changes in body movement index, respiratory rate, and heart rate within the preset sliding time window, as well as the length of the preset sliding time window; Based on the third sleep characteristic information and the third determination rule, the sleep period is divided into normal sleep state and drowsy state.

7. The method for monitoring bedside status throughout the entire time period according to claim 1, characterized in that, The classification of daytime activity status based on the target's presence information and physiological information during the daytime period includes: The body movement index during the daytime period is quantified to obtain the body movement amplitude level; Based on the target presence information, the level of body movement amplitude, and the fourth determination rule, the daytime activity state is divided into unmanned state, resting state, minimal activity state, and frequent activity state.

8. The method for monitoring bedside status throughout the entire time period according to claim 1, characterized in that, The process of generating a full-time bedtime status monitoring report based on the sleep state staging results and the daytime activity state classification results includes: Nighttime sleep score is calculated based on sleep state stage results and the target physiological information, and daytime vitality index is calculated based on daytime activity state classification results and the target physiological information; A full-time bedtime monitoring report is generated based on the nighttime sleep score and the daytime vitality index.

9. A radar-based system for monitoring the status of patients in bed at all times, characterized in that, include: The signal processing module is used to preprocess the acquired radar echo signals to obtain target presence information and target physiological information; wherein, the target physiological information includes at least body movement index, respiratory rate and heart rate; The feature extraction module is used to extract features from the target's existence information and physiological information, and adaptively divide daytime and nighttime periods based on the feature extraction results; The nighttime sleep staging module is used to perform multimodal sleep state staging based on the target's presence information and physiological information during the nighttime period, and to calculate the nighttime sleep score based on the sleep state staging results. The daytime activity classification module is used to classify daytime activity status based on the target's presence information and physiological information during the daytime period, and to calculate the daytime vitality index based on the daytime activity status classification results. The report generation module is used to generate a full-time bedtime status monitoring report based on the sleep state staging results and the daytime activity state classification results.

10. A radar, characterized in that, Includes memory and processor, of which: The processor is used to execute computer programs stored in the memory; When the processor executes the computer program, it implements the steps of the full-time bed condition monitoring method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the full-time bed condition monitoring method according to any one of claims 1 to 8.

Images