A non-contact method, system and device for monitoring bedridden data

Abstract

This specification discloses a non-contact bed rest data monitoring method, system, and device. The monitoring method includes: acquiring acquisition signals from the bed, including dynamic force detection signals and static force detection signals of the same origin; acquiring a body movement score representing the intensity of body movement of the subject in each monitoring cycle based on the acquisition signals; for any given monitoring cycle: determining the body movement state of that monitoring cycle based on the corresponding body movement score; acquiring the respiratory rate value of that monitoring cycle based on the body movement state and the acquisition signals; and outputting bed rest data including the body movement state and respiratory rate value for that monitoring cycle. This solution significantly improves the robustness, accuracy, and continuity of bed rest data monitoring under dynamic interference by quantifying body movement assessment, dynamically fusing multi-sensor signals, and employing an adaptive decision-making mechanism.

Description

Technical Field

[0001] Several embodiments of this specification relate to the field of health monitoring technology, specifically to the optimization of monitoring accuracy and continuity in non-contact bedridden data monitoring. Background Technology

[0002] Non-contact bed rest monitoring technology primarily utilizes sensors embedded in the bed to continuously and imperceptibly acquire vital signs information such as the patient's status upon leaving the bed, body movement, and heart rate and respiratory rate. Due to its advantages of requiring no wearable devices and providing a superior user experience, this technology holds broad application prospects in fields such as smart elderly care and health monitoring.

[0003] Existing solutions typically rely on a single type of sensor (such as a vibration or piezoelectric sensor) or a simple combination of multiple sensors. For example, a typical technical solution uses an array of sensors spaced apart along the lying direction of the body, determining the presence / absence status and basic lying posture by detecting the triggering of signals at each point. For body movement detection, a simple binary judgment of "present" or "absent" is made based on whether the amplitude of the sensor signal exceeds a preset threshold. Vital signs data such as respiration and heart rate are extracted from the sensor signals only when the subject is lying still.

[0004] However, the strong noise generated by body movement can drown out weak physiological signals, leading to failure or inaccuracy in the extraction of vital signs. Existing solutions cannot maintain the continuity and accuracy of monitoring in complex and dynamically changing bedridden scenarios. Summary of the Invention

[0005] This specification provides a non-contact method, system, and device for monitoring bedridden data, which solves the problems of inaccurate measurement of vital signs and coarse assessment of body movement status in existing monitoring technologies under bedridden scenarios with complex changes in body movement.

[0006] The technical solution is as follows:

[0007] Firstly, embodiments of this specification provide a non-contact method for monitoring bedridden data, comprising the following steps:

[0008] Acquire the collected signals from the bed, including dynamic force detection signals representing vibration changes and static force detection signals representing pressure changes, which are of the same origin.

[0009] Based on the acquired signals, obtain the body movement score corresponding to each monitoring cycle, which represents the intensity of the body movement of the measured object;

[0010] For any monitoring period:

[0011] The body movement status of the monitoring period is determined based on the body movement score corresponding to the monitoring period. The body movement status includes a resting state, a slight body movement state, and a vigorous body movement state.

[0012] The respiratory rate value for that monitoring period is obtained based on the body movement status and collected signals during that monitoring period.

[0013] If the monitoring period is in a resting state, the respiratory rate value of the monitoring period is obtained based on the signal segment of the corresponding acquisition signal of the monitoring period;

[0014] If the monitoring period is during a period of slight body movement, the respiratory rate value of the monitoring period is obtained based on the signal segments of the signals collected from the monitoring period and several adjacent monitoring periods.

[0015] If the monitoring period is a period of intense physical activity, the respiratory rate value of the monitoring period is obtained based on the signal segments of the corresponding acquisition signals of several monitoring periods adjacent to the monitoring period.

[0016] Output bed rest data, including body movement status and respiratory rate, corresponding to this monitoring cycle.

[0017] As a preferred option, bed rest data monitoring methods also include:

[0018] Extracting dynamic force body motion element signals based on dynamic force detection signals;

[0019] Extracting static force body motion signals based on static force detection signals;

[0020] Signal processing is performed on both dynamic and static force body motioner signals, including central difference, sliding window root mean square calculation, and abrupt change point detection.

[0021] The process of acquiring the body movement score, which characterizes the intensity of body movement of the tested object for each monitoring cycle based on the acquired signals, includes:

[0022] Based on dynamic force-body motion signals and static force-body motion signals, the motion score representing the intensity of motion of the measured object is obtained for each monitoring cycle.

[0023] As a preferred embodiment, the step of obtaining the body motion score representing the intensity of body motion of the measured object for each monitoring cycle based on dynamic force-body motion signals and static force-body motion signals includes:

[0024] The first body motion score, which represents the intensity of body motion of the measured object, is obtained for each monitoring cycle based on the dynamic force body motion signal.

[0025] The second motion score, which characterizes the intensity of motion of the object under test, is obtained for each monitoring cycle based on the static force motion sub-signal.

[0026] For any given monitoring period, the body movement score for that monitoring period is obtained based on the first and second body movement scores corresponding to that monitoring period.

[0027] As a preferred embodiment, the step of obtaining the first body motion score, which characterizes the intensity of body motion of the measured object for each monitoring cycle based on dynamic force body motion signals, includes:

[0028] Based on the dynamic force body motion signal, a first feature set including multiple dynamic force body motion signal features is obtained for each monitoring cycle;

[0029] Based on a preset first weight set including the weight coefficients corresponding to each dynamic force signal feature and a first feature set corresponding to each monitoring cycle, the first body motion score corresponding to each monitoring cycle is obtained.

[0030] The second body motion score, which characterizes the intensity of body motion of the measured object and is obtained for each monitoring cycle based on the static force body motion signal, includes:

[0031] Based on the static force-body motion sub-signals, a second feature set, including multiple static force-body motion signal features, is obtained for each monitoring cycle;

[0032] Based on a preset second weight set including the weight coefficients corresponding to each static force motion signal feature and a second feature set corresponding to each monitoring period, the second motion score corresponding to each monitoring period is obtained.

[0033] As a preferred embodiment, the step of obtaining the body motion score representing the intensity of body motion of the measured object for each monitoring cycle based on dynamic force-body motion signals and static force-body motion signals further includes:

[0034] The maximum motion monitoring cycle of the measured object is obtained based on dynamic force-body motion signals and static force-body motion signals;

[0035] Based on the signal segment corresponding to the maximum body motion monitoring period of the dynamic force body motion signal, a first feature reference set including multiple dynamic force body motion signal features is obtained, which is used to normalize the first feature set corresponding to each monitoring period.

[0036] Based on the signal segment corresponding to the maximum body motion monitoring period of the static force body motion sub-signal, a second feature reference set including various static force body motion signal features is obtained, which is used to normalize the second feature set corresponding to each monitoring period.

[0037] As a preferred option, bed rest data monitoring methods also include:

[0038] Dynamic force breathing sub-signals are extracted based on dynamic force detection signals, and the first signal quality score corresponding to each signal segment of the dynamic force breathing sub-signal in each monitoring cycle is obtained based on the signal morphology regularity.

[0039] The static force respiration sub-signal is extracted based on the static force detection signal, and the second signal quality score corresponding to each signal segment of the static force respiration sub-signal in each monitoring cycle is obtained based on the signal morphology regularity.

[0040] For any monitoring period, the respiratory signal segment corresponding to that monitoring period is obtained based on the signal segment and first signal quality fraction of the dynamic force breathing sub-signal, the signal segment and second signal quality fraction of the static force breathing sub-signal;

[0041] The process of obtaining the respiratory rate value for the monitoring period based on the body movement status and acquired signals includes:

[0042] The respiratory rate value for that monitoring period is obtained based on the body movement status and the corresponding respiratory signal segment.

[0043] As a preferred embodiment, obtaining the respiratory signal segment corresponding to the monitoring cycle includes:

[0044] Obtain the respiratory signal segment and its quality score corresponding to the monitoring period;

[0045] The output of bed rest data corresponding to this monitoring cycle, including body movement status and respiratory rate values, includes:

[0046] Output bed rest data corresponding to this monitoring cycle, including body movement status, respiratory rate value, and signal segment quality fraction.

[0047] As a preferred embodiment, if the monitoring period involves slight body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the signals collected from the monitoring period and several adjacent monitoring periods, including:

[0048] If the monitoring period is during a period of slight body movement, based on the respiratory signal segments of the monitoring period and several adjacent monitoring periods, the median respiratory rate of each respiratory signal segment is obtained.

[0049] Based on the signal quality scores of the respiratory signal segments corresponding to the monitoring period and several adjacent monitoring periods, as well as the timing of the monitoring period, the intermediate respiratory rate values ​​corresponding to each respiratory signal segment are weighted to obtain the respiratory rate value of the monitoring period.

[0050] Secondly, the embodiments of this specification provide a non-contact bed rest data monitoring system, including a signal acquisition unit, a body movement score evaluation unit, a body movement state division unit, a physiological data calculation unit, and a data output unit;

[0051] The signal acquisition unit acquires the collected signals from the bed body, including dynamic force detection signals characterizing vibration changes and static force detection signals characterizing pressure changes, which are of the same origin.

[0052] The body movement score evaluation unit acquires the body movement score, which represents the intensity of the body movement of the tested object, for each monitoring cycle based on the collected signals.

[0053] The body movement state classification unit determines the body movement state of any monitoring period based on the body movement score corresponding to that monitoring period. The body movement state includes resting state, slight body movement state, and vigorous body movement state.

[0054] The physiological data calculation unit, for any given monitoring period, obtains the respiratory rate value for that monitoring period based on the body movement state and the acquired signals. If the monitoring period is a resting state, the respiratory rate value for that monitoring period is obtained based on the signal segment of the acquired signal corresponding to that monitoring period. If the monitoring period is a state of slight body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the acquired signals corresponding to that monitoring period and several adjacent monitoring periods. If the monitoring period is a state of intense body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the acquired signals corresponding to several adjacent monitoring periods.

[0055] The data output unit outputs bed rest data, including body movement status and respiratory rate value, for any given monitoring period.

[0056] Thirdly, embodiments of this specification provide a non-contact bed rest data monitoring device, including the non-contact bed rest data monitoring system described in the second aspect of the above embodiments; and

[0057] The signal acquisition device includes a dynamic force detection unit for acquiring dynamic force detection signals and a static force detection unit for acquiring static force detection signals, used to provide the monitoring system with acquisition signals from the same source;

[0058] The terminal device is used to interact with the monitoring system, including receiving bedridden data.

[0059] Fourthly, embodiments of this specification provide an electronic device, including a processor and a memory; the processor is connected to the memory; the memory is used to store executable program code; the processor reads the executable program code stored in the memory to run a program corresponding to the executable program code, so as to perform the steps described in the first aspect of the above embodiments.

[0060] Fifthly, embodiments of this specification provide a computer storage medium storing a plurality of instructions adapted for loading by a processor and executing the steps described in the first aspect of the above embodiments.

[0061] The beneficial effects of the technical solutions provided in some embodiments of this specification include at least the following:

[0062] By adaptively classifying resting, mild, and vigorous body movement states based on body movement scores, and employing differentiated respiratory rate calculation strategies for each state, the problem of interrupted or inaccurate physiological signal extraction under body movement interference is effectively solved. This improves the continuity and adaptability of monitoring, reduces data loss due to body movement noise, and thus maintains stable output in dynamic environments.

[0063] By introducing enhanced processing of the body motion sub-signals, the sensitivity and robustness of body motion event detection are improved, providing high-quality input for accurate calculation of body motion scores, and thus supporting refined state judgment.

[0064] By using dynamic force detection signals and static force detection signals to calculate body movement scores in tandem, the limitations of single-type signals are overcome, and the completeness and reliability of body movement assessment are improved. In particular, when the body movement is large and the dynamic force detection signal becomes saturated, the static force detection signal can compensate for the lack of information, ensure that the body movement judgment is not interrupted, and quantify the intensity of body movement during the saturation period.

[0065] By extracting signal features from multiple dimensions such as the time and frequency domains and assigning weights to fuse and calculate body motion scores, a fine-grained quantification of body motion intensity is achieved. Compared with the traditional "present / absent" dichotomy, this method can accurately distinguish body motion of different intensities, providing more precise input for adaptive decision-making.

[0066] Based on the feature benchmark extracted from the maximum body movement monitoring cycle, the features of each body movement signal are normalized, eliminating the impact of individual user differences on the score, improving the universality and fairness of this scheme, making body movement scores comparable among different users, and enhancing the consistency of monitoring.

[0067] By calculating signal quality scores for dynamic and static force breathing sub-signals respectively, and dynamically selecting or fusing signal sources based on the signal quality scores, complementary optimization of multiple signals is achieved. The most reliable signal source is given priority to improve the accuracy and fault tolerance of respiratory rate calculation.

[0068] By outputting the signal segment quality score as confidence level along with the respiratory rate value, the monitoring results are upgraded from a single numerical value to a binary information of numerical value and confidence level. This enhances the interpretability of the system and provides a quantitative basis for subsequent data analysis.

[0069] For mild physical activity, the respiratory rate is dynamically weighted by the signal segment quality score and the time sequence of the monitoring cycle, and the median value of the respiratory rate between the current monitoring cycle and the adjacent monitoring cycles. This achieves a balance between tracking real-time physiological changes and maintaining output stability, preventing data jumps and further improving the smoothness and reliability of the respiratory rate under mild disturbances. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0071] Figure 1 This is a flowchart illustrating a non-contact bed rest data monitoring method provided in the embodiments of this specification.

[0072] Figure 2 This is a schematic diagram of a non-contact bed rest data monitoring system provided in the embodiments of this specification.

[0073] Figure 3 This is a schematic diagram of the structure of a non-contact bed rest data monitoring device provided in the embodiments of this specification.

[0074] Figure 4 This is a schematic diagram of the structure of an electronic device provided in the embodiments of this specification. Detailed Implementation

[0075] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings.

[0076] The terms "first," "second," "third," etc., in the description, claims, and accompanying drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0077] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the function and arrangement of the described elements without departing from the scope of this specification. Various processes or components may be appropriately omitted, substituted, or added to the examples. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

[0078] In existing technologies, most methods employ a single sensor or simple fusion scheme for data acquisition to monitor bedridden patients. However, the strong noise generated when the subject moves can completely drown out weak physiological signals such as heartbeat and respiration, leading to extraction failures. Therefore, it is difficult to maintain the continuity and accuracy of monitoring in complex dynamic scenarios. This solution is therefore proposed.

[0079] Reference Figure 1 As shown, Figure 1 This is a flowchart illustrating a method for monitoring bedridden data according to an embodiment of this specification, which may include at least the following steps:

[0080] Step 102: Acquire the acquisition signals on the bed, including dynamic force detection signals representing vibration changes and static force detection signals representing pressure changes, which are of the same origin.

[0081] Step 104: Based on the acquired signals, obtain the body movement score corresponding to each monitoring cycle, which represents the intensity of the body movement of the object under test;

[0082] For any monitoring period:

[0083] Step 106: Determine the body movement status of the monitoring period based on the body movement score corresponding to the monitoring period. The body movement status includes resting state, slight body movement state, and vigorous body movement state.

[0084] Step 108: Obtain the respiratory rate value for the monitoring period based on the body movement status and collected signals during the monitoring period;

[0085] If the monitoring period is in a resting state, the respiratory rate value of the monitoring period is obtained based on the signal segment of the corresponding acquisition signal of the monitoring period;

[0086] If the monitoring period is during a period of slight body movement, the respiratory rate value of the monitoring period is obtained based on the signal segments of the signals collected from the monitoring period and several adjacent monitoring periods.

[0087] If the monitoring period is a period of intense physical activity, the respiratory rate value of the monitoring period is obtained based on the signal segments of the corresponding acquisition signals of several monitoring periods adjacent to the monitoring period.

[0088] Step 110: Output the bed rest data corresponding to this monitoring cycle, including body movement status and respiratory rate values.

[0089] In the field of bedridden data monitoring, "non-contact" refers to the ability of monitoring devices to collect data without direct contact or attachment to the skin. This involves indirectly sensing physiological activities and states through sensors. This method avoids the discomfort and constraint of wearing devices, making it suitable for long-term, non-invasive monitoring in bedridden settings. Body movement refers to the activities that occur during sleep or rest, encompassing everything from subtle rises and falls caused by breathing and minor movements of the hands and feet to large-scale changes in position such as turning over, sitting up, and even getting in and out of bed. Body movement is not only an important behavioral indicator that needs to be monitored, but also a major source of noise interference when accurately extracting vital signs such as heart rate and respiratory rate. Therefore, detecting and classifying body movement is crucial for improving the accuracy and continuity of monitoring data.

[0090] This embodiment simultaneously acquires dynamic and static force detection signals from the bed, and these signals originate from the same source. A health monitoring sensor integrating two non-contact sensing units, such as a piezoelectric sensor and a strain gauge, can be used to simultaneously acquire these co-originating dynamic and static force detection signals. The piezoelectric sensor is extremely sensitive to dynamic micro-vibrations and excels at capturing high-frequency vibration signals generated by heartbeat, respiration, and body movement; the strain gauge is used to measure static or slowly changing pressure changes, stably sensing body weight, center of gravity shifts, and periodic pressure fluctuations caused by respiration. The two complement each other in terms of signal characteristics: the piezoelectric sensor has high sensitivity but is prone to saturation when the measured object undergoes significant body movement; the strain gauge has a large range and good stability but is not sensitive to high-frequency details. In this embodiment, the static force detection signal is used to supplement the dynamic force detection signal analysis, thus enabling effective identification of body movement events even when the dynamic force detection signal saturates and fails.

[0091] Interpretively, wavelet decomposition can be used to extract the corresponding body motion and respiratory components from the dynamic force detection signals and static force detection signals respectively, achieving signal separation. Prior to this, the dynamic and static force detection signals can be used together to determine whether the subject is in bed or out of bed. The in / out of bed determination thresholds for both signals are adaptively determined based on the subject's weight and historical data of the collected signals. When the subject is determined to be in bed, the steps of this scheme are executed.

[0092] The dynamic force detection signal and the static force detection signal are divided into monitoring cycles, for example, by dividing them into 10-second monitoring cycles according to time sequence. For real-time bedridden data output scenarios, the dynamic and static force detection signals are divided into 10-second intervals, corresponding to one monitoring cycle. For each monitoring cycle, the signal fluctuation characteristics are analyzed based on the signal segment corresponding to the body movement component of the acquired signal, and a body movement score is calculated to represent the intensity of the body movement of the tested subject. Existing methods for detecting body movement often use a simple binary judgment of presence or absence, which cannot quantify the degree and type of body movement, and lacks mechanisms for compensating for and optimizing physiological parameter output under body movement interference.

[0093] Furthermore, since different degrees of body movement interfere with the dynamic force detection signal to varying degrees, this scheme uses any monitoring period as the target monitoring period and determines the body movement state of the subject at that time based on the body movement score. For example, two state judgment thresholds can be set: when the body movement score is greater than both state judgment thresholds, it corresponds to a vigorous body movement state, such as large-scale turning over, sitting up, getting in and out of bed, etc.; when the body movement score is less than both state judgment thresholds, it corresponds to a resting state, such as slight fluctuations caused by breathing; when the body movement score is between the two state judgment thresholds, it is a slight body movement state, such as slight movements of the hands and feet.

[0094] Therefore, different respiratory rate calculation strategies are adopted for different body movement states. For the resting state, the respiratory rate value is directly calculated using the signal segment of the respiratory component of the acquired signal corresponding to the target monitoring period. For the state of vigorous body movement, the dynamic force detection signal is saturated and ineffective due to severe interference. Therefore, the calculation of the respiratory rate value using the signal segment of the respiratory component of the acquired signal corresponding to the target monitoring period is abandoned. Instead, the signal segment of the respiratory component corresponding to other adjacent monitoring periods is selected to estimate the result of the target monitoring period. For example, two monitoring periods before and after the target period are selected (if it is real-time bed rest data output, then the four monitoring periods adjacent to the target monitoring period are selected). The respiratory rate can be directly calculated for the signal segment corresponding to each of the four monitoring periods, and the arithmetic mean is taken as the respiratory rate value of the target monitoring period. For the state of slight body movement, although the dynamic force detection signal is affected to some extent, it is not completely ineffective. Therefore, the result of the target monitoring period is calculated by combining the signal of the target monitoring period and the signal segments of the respiratory component corresponding to other adjacent monitoring periods.

[0095] During periods when the signal quality of dynamic force detection signals is disturbed, relying on nearby reliable data can significantly improve the stability and continuity of physiological parameter outputs such as respiratory rate, avoiding monitoring interruptions caused by the loss of single-point data. Ultimately, bedridden data, including body movement status and respiratory rate values, will be output for storage and analysis.

[0096] This solution addresses the technical problems of inaccurate and easily interrupted vital sign measurements, as well as the coarse assessment of body movement status in bedridden scenarios where existing monitoring technologies are used to monitor dynamic changes in body movement.

[0097] In one embodiment of this specification, the bed rest data monitoring method further includes:

[0098] Extracting dynamic force body motion element signals based on dynamic force detection signals;

[0099] Extracting static force body motion signals based on static force detection signals;

[0100] Signal processing is performed on both dynamic and static force body motioner signals, including central difference, sliding window root mean square calculation, and abrupt change point detection.

[0101] Based on the acquired signals, a body movement score representing the intensity of body movement of the tested object is obtained for each monitoring cycle, including:

[0102] Based on dynamic force-body motion signals and static force-body motion signals, the motion score representing the intensity of motion of the measured object is obtained for each monitoring cycle.

[0103] Illustratively, this embodiment uses wavelet decomposition to extract the corresponding body motion components (body motion sub-signals) from the dynamic force detection signal and the static force detection signal, respectively. By introducing a body motion enhancement algorithm, the extracted body motion sub-signals are subjected to central difference, sliding window root mean square calculation and abrupt change point detection to improve the sensitivity and robustness of body motion event detection.

[0104] Interpretively, center difference is a mathematical method for calculating the instantaneous rate of change of a signal. For each point in the signal, the value of the preceding point is subtracted from the value of the following point; the result represents the instantaneous rate of change at that point. Center difference can amplify weak, slow body movements, making them easier to detect.

[0105] The sliding window root mean square (RMS) calculation is a metric for measuring the average energy of a signal within a time window. Sliding window calculation involves processing data sequentially for each small time interval (e.g., 2 seconds per window) and calculating the RMS value of the signal within that window. The original signal or the differentiated signal may contain random high-frequency noise (such as interference from electronic devices). These isolated noise points can be averaged out, thus smoothing the signal and reducing false alarms due to random noise. Furthermore, it directly reflects the energy intensity of bodily movements within that time period; the greater the intensity, the higher the RMS value, providing a stable and quantitative basis for subsequently judging the intensity of bodily movements.

[0106] Abrupt change detection is an algorithm used to accurately locate the moments when signals undergo sudden changes (i.e., abrupt changes). By analyzing the root mean square sequence, it identifies points where values ​​rise or fall sharply, thus pinpointing the start and end points of a dynamic event. It can also be used to distinguish between consecutive events, improving the temporal accuracy and event resolution capabilities of the detection.

[0107] This embodiment can analyze the intensity of the rate of change (such as velocity and acceleration) of bodily sub-signals by first performing central difference calculation followed by sliding window root mean square calculation, which can be used to detect fluctuations in the rate of change or vibrational energy. Alternatively, it can first perform sliding window root mean square calculation followed by central difference calculation to analyze the trend of changes in the energy or amplitude of bodily sub-signals, which can be used to monitor the rate of energy change. Finally, a mutation point detection method is used to identify each independent bodily motion event in the energy sequence, improving accuracy. This provides high-quality and highly reliable input data for subsequent calculation of bodily motion fractions and determination of bodily motion states.

[0108] In one embodiment of this specification, a body motion score characterizing the intensity of body motion of the measured object is obtained for each monitoring cycle based on dynamic force-body motion signals and static force-body motion signals, including:

[0109] The first body motion score, which represents the intensity of body motion of the measured object, is obtained for each monitoring cycle based on the dynamic force body motion signal.

[0110] The second motion score, which characterizes the intensity of motion of the object under test, is obtained for each monitoring cycle based on the static force motion sub-signal.

[0111] For any given monitoring period, the body movement score for that monitoring period is obtained based on the first and second body movement scores corresponding to that monitoring period.

[0112] To illustrate, static force detection signals have a lower frequency band and do not include high-frequency components such as heartbeats; the surges in energy and frequency are closely related to body movement. Dynamic force detection signals, on the other hand, have a wider frequency domain and stronger time-domain sensitivity. Therefore, the two signals can complement each other for more refined body movement perception.

[0113] Interpretively, the body motion fractions of the separated dynamic and static force-body motion signals are calculated separately to obtain the first and second body motion fractions for each monitoring period. For each monitoring period, a comprehensive evaluation is performed based on the corresponding first and second body motion fractions to determine the final body motion fraction.

[0114] For example, the first and second body motion scores are weighted and fused to obtain the final body motion score. Since piezoelectric sensors and pressure strain gauges each have their advantages and disadvantages—piezoelectric sensors are sensitive but prone to saturation, while pressure strain gauges are stable but lack fine detail—a larger weighting coefficient is assigned to the first body motion score, and a smaller weighting coefficient is assigned to the second body motion score. The weighting coefficients are determined after verification using extensive experimental data to ensure that the final body motion score most accurately reflects the intensity of body motion.

[0115] It enables the static force detection signal to compensate for the missing information when the dynamic force detection signal is saturated, thus accurately and continuously reflecting the body motion state of the measured object.

[0116] In addition, when the dynamic force detection signal is in a saturated state, the first body motion fraction reaches its maximum value. At this point, as the intensity of the body motion further increases, the weighted fusion result (or other fusion methods) will gradually become inaccurate. Therefore, the amplitude saturation period of the dynamic force detection signal can be actively identified. During the amplitude saturation period of the dynamic force detection signal, only the static force body motion sub-signal is used to determine the amplitude of the body motion change of the measured object, thus avoiding inaccuracies.

[0117] Specifically, synchronously acquired dynamic and static force detection signals are obtained. The 3-second sliding window root mean square (RMS) of the dynamic force motion sub-signal is calculated, and the central difference of this RMS sequence is performed. Threshold judgments are applied to both the sliding window RMS and the central difference values ​​to determine the amplitude saturation period of the dynamic force detection signal. Within the amplitude saturation period, the central difference of the static force motion sub-signal is calculated, and its absolute value is taken. The peak value and duration of the high value of this absolute difference are positively correlated with the amplitude of the body movement change. Summing the absolute values ​​of the differences within a certain time window reflects the differences in the body movement changes of the measured object under full amplitude of the dynamic force detection signal. For example, the peak value of the absolute difference corresponding to the getting in and out of bed event is significantly greater than that of the turning over event in bed, and the sum of the absolute values ​​of the differences for large body movement events is significantly greater than that for small body movement events. The sliding window root mean square (RMS) of the dynamic force motion sub-signal is calculated to reflect the magnitude of the signal energy; the RMS value will remain high during amplitude saturation. Calculating the central difference of the RMS sequence is equivalent to calculating the rate of change of the RMS value. At the start and end of amplitude saturation, the RMS value undergoes a sharp jump, causing a spike in the difference value. Thresholds are set for both the RMS value itself and its difference value. Only when both exceed the thresholds is the time point determined to be within the amplitude saturation period, accurately marking the start and end times of saturation failure of the dynamic force detection signal to avoid misjudgment.

[0118] In one embodiment of this specification, obtaining a first body motion score representing the intensity of body motion of the measured object for each monitoring cycle based on dynamic force body motion signals includes:

[0119] Based on the dynamic force body motion signal, a first feature set including multiple dynamic force body motion signal features is obtained for each monitoring cycle;

[0120] Based on a preset first weight set including the weight coefficients corresponding to each dynamic force signal feature and a first feature set corresponding to each monitoring cycle, the first body motion score corresponding to each monitoring cycle is obtained.

[0121] The second body motion score, representing the intensity of body motion of the measured object, is obtained for each monitoring cycle based on the static force body motion signal, including:

[0122] Based on the static force-body motion sub-signals, a second feature set, including multiple static force-body motion signal features, is obtained for each monitoring cycle;

[0123] Based on a preset second weight set including the weight coefficients corresponding to each static force motion signal feature and a second feature set corresponding to each monitoring period, the second motion score corresponding to each monitoring period is obtained.

[0124] For example, features are extracted from the time domain and frequency domain respectively for dynamic force body motion signals and static force body motion signals, and different weight sets are assigned to dynamic force body motion signals and static force body motion signals respectively to calculate the first motion score and the second motion score.

[0125] Specifically, dynamic force-body signal characteristics include full-amplitude ratio, root mean square (RMS), variance, kurtosis, waveform factor, impulse factor, margin factor, spectral centroid, spectral roll-off point, spectral centroid standard deviation, average energy, energy standard deviation, total energy, and spectral entropy. Static force-body signal characteristics include variance, waveform factor, peak factor, impulse factor, margin factor, MMS frequency, spectral bandwidth, spectral centroid standard deviation, and energy standard deviation.

[0126] Establish a fine-grained body movement assessment system, and based on this, realize the adaptive switching and fusion of physiological parameter calculation strategies to optimize the reliability of physiological parameter output.

[0127] In one embodiment of this specification, a body motion score representing the intensity of body motion of the measured object is obtained for each monitoring cycle based on dynamic force-body motion signals and static force-body motion signals. This is further illustrated by:

[0128] The maximum motion monitoring cycle of the measured object is obtained based on dynamic force-body motion signals and static force-body motion signals;

[0129] Based on the signal segment corresponding to the maximum body motion monitoring period of the dynamic force body motion signal, a first feature reference set including multiple dynamic force body motion signal features is obtained, which is used to normalize the first feature set corresponding to each monitoring period.

[0130] Based on the signal segment corresponding to the maximum body motion monitoring period of the static force body motion sub-signal, a second feature reference set including various static force body motion signal features is obtained, which is used to normalize the second feature set corresponding to each monitoring period.

[0131] Specifically, based on a preset first weight set including weight coefficients corresponding to each dynamic force-body motion signal feature and a first feature set corresponding to each monitoring period, a first body motion score corresponding to each monitoring period is obtained, including:

[0132] The first feature set corresponding to each monitoring cycle is normalized based on the first feature benchmark set.

[0133] Based on a preset first weight set including the weight coefficients corresponding to each dynamic force signal feature and a normalized first feature set corresponding to each monitoring cycle, the first body motion score corresponding to each monitoring cycle is obtained.

[0134] Based on a preset second weight set including weight coefficients corresponding to each static force motion signal feature and a second feature set corresponding to each monitoring period, the second motion score corresponding to each monitoring period is obtained, including:

[0135] The second feature set corresponding to each monitoring cycle is normalized based on the second feature benchmark set.

[0136] Based on a preset second weight set including the weight coefficients corresponding to each static force motion signal feature and a normalized second feature set corresponding to each monitoring cycle, the second motion score corresponding to each monitoring cycle is obtained.

[0137] To illustrate, since different subjects have different weights, their baseline levels of body movement signals vary significantly. Using the same standard to calculate body movement scores would lead to inaccurate judgments. Therefore, an adaptive, personalized scoring process must be established for each subject. First, the signal segment representing the subject's maximum body movement intensity is identified from their historical data. The most typical example is the event of getting in and out of bed, where both dynamic and static force-body movement signals will continuously saturate and undergo dramatic and sustained changes. Then, based on the signal segments of the dynamic and static force-body movement signals corresponding to the maximum body movement monitoring period, various dynamic and static force-body movement signal features are extracted, representing the maximum degree of body movement the subject can produce in that monitoring scenario. Using the values ​​of the dynamic and static force-body movement signal features corresponding to the maximum body movement monitoring period as the maximum baseline value, the signal segments of the dynamic and static force-body movement signals corresponding to each monitoring period are normalized. For example: Normalized feature value = Original feature value / Maximum baseline value.

[0138] Furthermore, a segment of signals confirmed to be in a resting position can be extracted from the historical data of the tested object. Similarly, the values ​​of its dynamic force-body signal characteristics and static force-body signal characteristics are calculated as resting baseline values, which, together with the maximum baseline value, form the dynamic performance range of each characteristic. For example: Normalized characteristic value = (original characteristic value - resting baseline value) / (maximum baseline value - resting baseline value).

[0139] Interpretive, normalized values ​​transform the eigenvalues ​​of each feature into dimensionless values ​​between 0 and 1, eliminating numerical scaling differences and making the weights assigned to different features more equitable and reasonable. Furthermore, it ensures fair and accurate assessments regardless of the subject's weight or size, within the range of their physical activity.

[0140] In one embodiment of this specification, the bed rest data monitoring method further includes:

[0141] Dynamic force breathing sub-signals are extracted based on dynamic force detection signals, and the first signal quality score corresponding to each signal segment of the dynamic force breathing sub-signal in each monitoring cycle is obtained based on the signal morphology regularity.

[0142] The static force respiration sub-signal is extracted based on the static force detection signal, and the second signal quality score corresponding to each signal segment of the static force respiration sub-signal in each monitoring cycle is obtained based on the signal morphology regularity.

[0143] For any monitoring period, the respiratory signal segment corresponding to that monitoring period is obtained based on the signal segment and first signal quality fraction of the dynamic force breathing sub-signal, the signal segment and second signal quality fraction of the static force breathing sub-signal;

[0144] The respiratory rate value for that monitoring period is obtained based on the body movement status and acquired signals, including:

[0145] The respiratory rate value for that monitoring period is obtained based on the body movement status and the corresponding respiratory signal segment.

[0146] For example, indicators corresponding to the signal morphology regularity may include peak-to-peak value, J-wave rising slope, difference and standard deviation between half-width at half-maximum and template, sum of intra-class distances based on the first three points, standard deviation of RR intervals, overall DTW distance, heart rate abnormality limiting parameters, sliding standard deviation, root mean square, peak-to-peak ratio, and overall frequency band energy proportion. Weights are assigned to each indicator based on their positive or negative correlation with the signal morphology regularity, and the signal quality score is obtained by weighted summation of all indicators.

[0147] Interpretive, calculated signal quality scores are used to quantify and compare the merits of two signal sources, providing a basis for subsequent signal source processing decisions. Each monitoring cycle corresponds to two signal segments, one from dynamic force breathing sub-signal and the other from static force breathing sub-signal. Signal processing is performed based on the signal quality scores of the two signal segments to obtain the respiratory signal segments corresponding to each monitoring cycle.

[0148] For example, signal processing can be performed based on the signal quality scores of the two signal segments. The ratio of the first signal quality score to the second signal quality score can be used as the basis for signal processing, and the ratio of the signal quality scores can be compared with a preset quality evaluation threshold. When the ratio is less than both quality evaluation thresholds, the signal segment of the static force breathing sub-signal corresponding to the monitoring period is used as the breathing signal segment of that monitoring period, meaning that the breathing rate calculation of this breathing signal segment depends entirely on the pressure strain gauge. When the ratio is greater than both quality evaluation thresholds, the signal segment of the dynamic force breathing sub-signal corresponding to the monitoring period is used as the breathing signal segment of that monitoring period, meaning that the breathing rate calculation of this breathing signal segment depends entirely on the piezoelectric sensor. When the ratio is between the two quality evaluation thresholds, the signal segments of the static force breathing sub-signal and the dynamic force breathing sub-signal corresponding to the monitoring period are used together as the breathing signal segment of that monitoring period. When calculating the breathing rate of this breathing signal segment, the weighted fusion value of the breathing rate calculated by the two signal sources and their respective corresponding signal quality scores is used as the calculation result.

[0149] This embodiment evaluates the reliability of the signal source by signal quality score and dynamically selects the optimal signal source based on reliability. Thus, in the scenario of monitoring bedridden data with complex body movement changes, it can always automatically select the most reliable signal source and maximize the reliability of physiological parameter output.

[0150] In one embodiment of this specification, due to the uncertain nature of the tested object and various complex body motion states, the dynamic force breathing sub-signal extracted from the dynamic force detection signal by a set of signal filtering combinations exhibits significant differences in performance across different specific scenarios. Therefore, a multi-path parallel extraction and selection strategy is adopted for the dynamic force breathing sub-signal. The dynamic force detection signal is extracted using multiple wavelet filtering combinations (different wavelet bases and frequency bands: 0.121 ~ 1.97Hz and 0.243 ~ 0.983Hz). The waveform of the dynamic force breathing sub-signal extracted by wavelet filtering may still contain some minor spikes or random fluctuations. Further smoothing filtering using moving averages can be applied to make the signal waveform curve smoother, facilitating subsequent accurate calculations. A first signal quality score is calculated for each of the extracted filtered signals based on the regularity of the signal morphology. The filtered signal with the highest score and its first signal quality score are used to represent the dynamic force detection signal for comparison. For the static force detection signal, wavelet transform is used to remove low-frequency drift (0 ~ 0.1953Hz) and body motion interference to obtain the static force breathing sub-signal. Low-frequency drift is typically caused by temperature changes, the characteristics of the acquisition device itself, or slow peristalsis of the body in bed. It manifests as a very slow rise or fall in the signal baseline, severely distorting the static force respiratory sub-signal. Body movements such as turning over generate sudden signals with frequencies overlapping with the respiratory portion but much stronger energy, completely drowning out the static force respiratory sub-signal. Therefore, both must be removed. Wavelet transform has powerful multi-resolution analysis capabilities, decomposing the signal into different frequency sub-bands and identifying and removing signal components with frequencies below 0.1953Hz. It can also locate points of drastic change, effectively suppressing the influence of body movement by discarding or weakening these interfering components during signal reconstruction. Ultimately, the originally very weak and easily masked respiratory pressure fluctuations are extracted. Similarly, the signal can be further purified by moving average smoothing to ensure a more regular respiratory waveform for subsequent analysis.

[0151] In addition, since high-frequency heartbeat signals cannot be effectively captured from static force detection signals, heart rate analysis in this scheme relies entirely on dynamic force detection signals. To address the susceptibility of static force detection signals to interference, a multi-channel parallel, merit-based strategy is also employed. First, the same static force detection signal is simultaneously processed using Empirical Mode Decomposition (EMD) and three different wavelet basis filtering methods (sym8, db6, and coif5) to generate four candidate heartbeat signals. Next, a signal quality score is calculated for each channel based on the regularity of its signal morphology. Finally, the signal quality scores of these four channels are compared, and the channel with the highest score is selected to calculate and output the heart rate value. This significantly improves the success rate and accuracy of heart rate monitoring under conditions of physical movement and other interference.

[0152] In one embodiment of this specification, obtaining the respiratory signal segment corresponding to the monitoring period includes:

[0153] Obtain the respiratory signal segment and its quality score corresponding to the monitoring period;

[0154] Output the bed rest data corresponding to this monitoring cycle, including body movement status and respiratory rate values, including:

[0155] Output bed rest data corresponding to this monitoring cycle, including body movement status, respiratory rate value, and signal segment quality fraction.

[0156] Explanatoryly, based on the first signal quality score, second signal quality score, signal segment of dynamic force breathing sub-signal, and signal segment of static force breathing sub-signal corresponding to the monitoring period, the respiratory signal segment corresponding to the monitoring period is obtained, and the signal segment quality score of the respiratory signal segment corresponding to the monitoring period is also obtained. For example, when the aforementioned quality score ratios are all less than two quality evaluation thresholds, the second signal quality score is used as the signal segment quality score; when the ratios are all greater than two quality evaluation thresholds, the first signal quality score is used as the signal segment quality score; when the ratios are between the two quality evaluation thresholds, the average of the first and second signal quality scores is used as the signal segment quality score.

[0157] The interpretable output of bedridden data includes signal segment quality scores. These scores serve as the signal confidence level for the corresponding monitoring period of the respiratory signal. Signal confidence directly reflects the quality of the original signal used to calculate the respiratory rate value; a higher score indicates less interference and higher waveform regularity within that monitoring period, resulting in more reliable calculations. This provides a reliability metric for subsequent data interpretation and application, upgrading monitoring results from single numerical values ​​to a binary system of value and confidence. It effectively avoids misjudgments caused by misinterpreting unreliable data, enhancing the interpretability and practicality of the entire monitoring scheme.

[0158] In one embodiment of this specification, if the monitoring period is during a period of slight body movement, the respiratory rate value of the monitoring period is obtained based on the signal segments of the signals corresponding to the monitoring period and several adjacent monitoring periods, including:

[0159] If the monitoring period is during a period of slight body movement, based on the respiratory signal segments of the monitoring period and several adjacent monitoring periods, the median respiratory rate of each respiratory signal segment is obtained.

[0160] Based on the signal quality scores of the respiratory signal segments corresponding to the monitoring period and several adjacent monitoring periods, as well as the timing of the monitoring period, the intermediate respiratory rate values ​​corresponding to each respiratory signal segment are weighted to obtain the respiratory rate value of the monitoring period.

[0161] Interpretively, by combining the signal quality score of the respiratory signal segment corresponding to the monitoring cycle with the temporal sequence of the monitoring cycle, the reliability of the respiratory rate value calculated under slight body movement is further improved. Monitoring cycles with higher signal quality scores for respiratory signal segments are considered more reliable, and vice versa. Therefore, the respiratory rate is first calculated separately for the respiratory signal segments corresponding to the target monitoring cycle and adjacent target monitoring cycles, serving as an intermediate value for calculating the respiratory rate value of the target monitoring cycle; hence, this is called the intermediate respiratory rate value. Then, the overall weight of each intermediate respiratory rate value is determined based on the signal quality score of the respiratory signal segment and the temporal sequence of each monitoring cycle. The higher the signal quality score of the respiratory signal segment, the greater the weight will be assigned to the intermediate respiratory rate value of its corresponding monitoring cycle. Simultaneously, long-term and short-term weighting is applied according to the temporal sequence. Long-term refers to monitoring cycles farther from the target monitoring cycle, and short-term refers to monitoring cycles closer to the target monitoring cycle. The intermediate respiratory rate value corresponding to the monitoring cycle closer to the target monitoring cycle is assigned a greater weight. Finally, a weighted fusion is performed to obtain the respiratory rate value of the target monitoring cycle.

[0162] For example, the respiratory rate value of the target monitoring period is calculated by selecting the target monitoring period and the two preceding adjacent monitoring periods, as detailed in Table 1.

[0163] Table 1:

[0164]

[0165] The reliability-based dynamic weighted fusion mechanism in this embodiment ensures a relative balance between the conflicting goals of tracking real-time physiological changes and maintaining stable and continuous output under slight body movement, significantly improving the reliability and robustness of the overall monitoring results.

[0166] It is easy to understand that, for periods of intense physical activity, the same method can be used to calculate the respiratory rate value for the target monitoring period using only historical monitoring cycles.

[0167] Additionally, for periods of intense physical activity, a pre-trained Long Short-Term Memory (LSTM) model can be used to predict the respiratory rate for the target monitoring period. Specifically, the static force-motion signal features and dynamic force-motion signal features corresponding to each of multiple historical monitoring periods, as well as the signal morphology regularity features of the respiratory sub-signals corresponding to static and dynamic forces, are all input into the LSTM model for prediction, outputting the respiratory rate value for the target monitoring period. Simultaneously, a prediction flag is added to the output bedridden data to indicate that the respiratory rate value at this moment is the predicted value under intense physical activity disturbance.

[0168] Furthermore, based on the duration of the intense body movement state, it can be further divided into short-term intense body movement state and long-term intense body movement state. In the short-term intense body movement state, the corresponding bed rest data for that monitoring cycle, including body movement status (intense body movement marker), respiratory rate value, predictive marker, and signal segment quality score, is output. In the long-term intense body movement state (lasting at least two monitoring cycles), a continuous intense body movement warning is output, indicating that the subject may be in an abnormal state, and the output of respiratory rate values ​​is suspended.

[0169] It should be noted that, based on this scheme, the heart rate value of the tested subject can be calculated using the same logic as the respiratory rate value, but only based on the dynamic force detection signal, so as to achieve accurate detection of heart rate under body movement interference.

[0170] This invention, by constructing a three-layer intelligent architecture that integrates quantitative body motion assessment, real-time quality evaluation, and adaptive decision fusion, elevates dual signal sources from simple information redundancy backup to a collaborative and complementary intelligent sensing system, effectively solving the key problem of body motion interference that has long plagued the field of non-contact monitoring.

[0171] Working principle:

[0172] Based on the amplitude threshold of the dynamic force detection signal, the amplitude threshold of the static force detection signal, and the dynamic force detection signal and the static force detection signal, it is determined whether the object under test is in bed or out of bed.

[0173] In bed condition, wavelet decomposition method was used to extract body motion component and respiratory component from dynamic force detection signal and static force detection signal respectively;

[0174] The extracted bodily signals are subjected to central difference, sliding window root mean square calculation, and abrupt change point detection.

[0175] The body movement signal features of the body movement sub-signal are extracted from the time domain, frequency domain, and time frequency domain, and different weight sets are assigned to them respectively to calculate the body movement score and distinguish between resting, slight body movement and violent body movement states.

[0176] Calculate the signal quality score of the respiratory sub-signal;

[0177] Select the optimal filtering channel based on the signal quality score;

[0178] The signal source for calculating the final respiratory rate value is selected based on the ratio of the first signal quality fraction to the second signal quality fraction.

[0179] The respiratory rate value is calculated using the corresponding strategy based on the specific body movement state;

[0180] The output includes bedridden data including body movement status, respiratory rate, and signal segment quality fraction.

[0181] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0182] Please refer to the following. Figure 2 , Figure 2 A schematic diagram of a non-contact bed rest data monitoring system provided in an embodiment of this specification is shown.

[0183] The monitoring system 200 includes a signal acquisition unit 201, a body movement score evaluation unit 202, a body movement state classification unit 203, a physiological data calculation unit 204, and a data output unit 205;

[0184] The signal acquisition unit 201 acquires the collected signals from the bed, including dynamic force detection signals characterizing vibration changes and static force detection signals characterizing pressure changes, which are of the same origin.

[0185] The body movement score evaluation unit 202 acquires the body movement score, which represents the intensity of body movement of the tested object, for each monitoring cycle based on the collected signals.

[0186] The body movement state classification unit 203 determines the body movement state of any monitoring period based on the body movement score corresponding to that monitoring period. The body movement state includes resting state, slight body movement state and vigorous body movement state.

[0187] The physiological data calculation unit 204, for any given monitoring period, obtains the respiratory rate value for that monitoring period based on the body movement state and the acquired signals. If the monitoring period is a resting state, the respiratory rate value for that monitoring period is obtained based on the signal segment of the acquired signal corresponding to that monitoring period. If the monitoring period is a state of slight body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the acquired signals corresponding to that monitoring period and several adjacent monitoring periods. If the monitoring period is a state of intense body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the acquired signals corresponding to several adjacent monitoring periods.

[0188] The data output unit 205 outputs bed rest data, including body movement status and respiratory rate values, for any given monitoring period.

[0189] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the bed rest data monitoring system embodiment is basically similar to the bed rest data monitoring method embodiment, so the description is relatively simple; relevant parts can be referred to the description of the bed rest data monitoring method embodiment.

[0190] Please refer to the following. Figure 3 , Figure 3 A schematic diagram of a non-contact bed rest data monitoring device provided in an embodiment of this specification is shown.

[0191] The monitoring device 300 includes the monitoring system 200 described in the previous embodiment; and

[0192] The signal acquisition device 301 includes a dynamic force detection unit 3011 for acquiring dynamic force detection signals and a static force detection unit 3012 for acquiring static force detection signals, and is used to provide the monitoring system 200 with acquisition signals from the same source.

[0193] Terminal device 302 is used to interact with monitoring system 200, including receiving bedridden data.

[0194] Please see Figure 4 The diagram shown is a structural schematic of an electronic device provided in an embodiment of this specification.

[0195] like Figure 4 As shown, the electronic device 400 may include at least one processor 401, at least one network interface 404, a user interface 403, a memory 405, and at least one communication bus 402.

[0196] The communication bus 402 can be used to realize the connection and communication of the above components.

[0197] The user interface 403 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.

[0198] Among them, network interface 404 may include, but is not limited to, Bluetooth module, NFC module, Wi-Fi module, etc.

[0199] The processor 401 may include one or more processing cores. The processor 401 connects to various parts within the electronic device 400 using various interfaces and lines. It executes various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 405, and by calling data stored in the memory 405. Optionally, the processor 401 may be implemented using at least one hardware form of DSP, FPGA, or PLC. The processor 401 may integrate one or more of the following: CPU, GPU, and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content to be displayed on the screen; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 401 and may be implemented as a separate chip.

[0200] The memory 405 may include RAM or ROM. Optionally, the memory 405 may include a non-transitory computer-readable medium. The memory 405 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 405 may also be at least one storage device located remotely from the aforementioned processor 401. As a computer storage medium, the memory 405 may include an operating system, a network communication module, a user interface module, and a bedridden data monitoring application. The processor 401 may be used to call the bedridden data monitoring application stored in the memory 405 and execute the steps of the bedridden data monitoring method mentioned in the foregoing embodiments.

[0201] This specification also provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform one or more steps in the above-described embodiments of the bedridden data monitoring method. If the constituent modules of the above-described electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0202] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).

[0203] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks. Unless otherwise specified, the technical features of this embodiment and its implementation can be combined arbitrarily.

[0204] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Any modifications and improvements made by those skilled in the art to the technical solutions of this specification without departing from the spirit of this specification should fall within the protection scope defined by the claims of this specification.

Claims

1. A non-contact method for monitoring bedridden data, characterized in that, Includes the following steps: Acquire the collected signals from the bed, including dynamic force detection signals representing vibration changes and static force detection signals representing pressure changes, which are of the same origin. Extracting dynamic force body motion element signals based on dynamic force detection signals; Extracting static force body motion signals based on static force detection signals; Based on dynamic force-body motion signals and static force-body motion signals, obtain the motion score representing the intensity of motion of the measured object in each monitoring cycle; For any monitoring period: The body movement status of the monitoring period is determined based on the body movement score corresponding to the monitoring period. The body movement status includes a resting state, a slight body movement state, and a vigorous body movement state. The respiratory rate value for that monitoring period is obtained based on the body movement status and collected signals during that monitoring period. If the monitoring period is in a resting state, the respiratory rate value of the monitoring period is obtained based on the signal segment of the corresponding acquisition signal of the monitoring period; If the monitoring period is during a period of slight body movement, the respiratory rate value of the monitoring period is obtained based on the signal segments of the signals collected from the monitoring period and several adjacent monitoring periods. If the monitoring period is a period of intense physical activity, the respiratory rate value of the monitoring period is obtained based on the signal segments of the corresponding acquisition signals of several monitoring periods adjacent to the monitoring period. Output bed rest data, including body movement status and respiratory rate, corresponding to this monitoring cycle.

2. The non-contact bed rest data monitoring method according to claim 1, characterized in that, The process of obtaining the body motion score, which characterizes the intensity of body motion of the measured object, for each monitoring cycle based on dynamic and static force-body motion signals includes: Signal processing is performed on both dynamic and static force body motioner signals, including central difference, sliding window root mean square calculation, and abrupt change point detection. Based on the signal processing results, the motion score representing the intensity of the body movement of the object under test is obtained for each monitoring cycle.

3. The non-contact bed rest data monitoring method according to claim 2, characterized in that, The process of obtaining the body motion score, which characterizes the intensity of body motion of the measured object, for each monitoring cycle based on dynamic and static force-body motion signals includes: The first body motion score, which represents the intensity of body motion of the measured object, is obtained for each monitoring cycle based on the dynamic force body motion signal. The second motion score, which characterizes the intensity of motion of the object under test, is obtained for each monitoring cycle based on the static force motion sub-signal. For any given monitoring period, the body movement score for that monitoring period is obtained based on the first and second body movement scores corresponding to that monitoring period.

4. The non-contact bed rest data monitoring method according to claim 3, characterized in that, The method of obtaining the first body motion score, which characterizes the intensity of body motion of the measured object for each monitoring cycle based on dynamic force body motion signals, includes: Based on the dynamic force body motion signal, a first feature set including multiple dynamic force body motion signal features is obtained for each monitoring cycle; Based on a preset first weight set including the weight coefficients corresponding to each dynamic force signal feature and a first feature set corresponding to each monitoring cycle, the first body motion score corresponding to each monitoring cycle is obtained. The second body motion score, which characterizes the intensity of body motion of the measured object and is obtained for each monitoring cycle based on the static force body motion signal, includes: Based on the static force-body motion sub-signals, a second feature set, including multiple static force-body motion signal features, is obtained for each monitoring cycle; Based on a preset second weight set including the weight coefficients corresponding to each static force motion signal feature and a second feature set corresponding to each monitoring period, the second motion score corresponding to each monitoring period is obtained.

5. The non-contact bed rest data monitoring method according to claim 4, characterized in that, The process of obtaining the body motion score, which characterizes the intensity of body motion of the measured object, for each monitoring cycle based on dynamic and static force-body motion signals, previously included: The maximum motion monitoring cycle of the measured object is obtained based on dynamic force-body motion signals and static force-body motion signals; Based on the signal segment corresponding to the maximum body motion monitoring period of the dynamic force body motion signal, a first feature reference set including multiple dynamic force body motion signal features is obtained, which is used to normalize the first feature set corresponding to each monitoring period. Based on the signal segment corresponding to the maximum body motion monitoring period of the static force body motion sub-signal, a second feature reference set including various static force body motion signal features is obtained, which is used to normalize the second feature set corresponding to each monitoring period.

6. The non-contact bed rest data monitoring method according to claim 1, characterized in that, Also includes: Dynamic force breathing sub-signals are extracted based on dynamic force detection signals, and the first signal quality score corresponding to each signal segment of the dynamic force breathing sub-signal in each monitoring cycle is obtained based on the signal morphology regularity. The static force respiration sub-signal is extracted based on the static force detection signal, and the second signal quality score corresponding to each signal segment of the static force respiration sub-signal in each monitoring cycle is obtained based on the signal morphology regularity. For any monitoring period, the respiratory signal segment corresponding to that monitoring period is obtained based on the signal segment and first signal quality fraction of the dynamic force breathing sub-signal, the signal segment and second signal quality fraction of the static force breathing sub-signal; The process of obtaining the respiratory rate value for the monitoring period based on the body movement status and acquired signals includes: The respiratory rate value for that monitoring period is obtained based on the body movement status and the corresponding respiratory signal segment.

7. The non-contact bed rest data monitoring method according to claim 6, characterized in that, The acquisition of the respiratory signal segment corresponding to the monitoring cycle includes: Obtain the respiratory signal segment and its quality score corresponding to the monitoring period; The output of bed rest data corresponding to this monitoring cycle, including body movement status and respiratory rate values, includes: Output bed rest data corresponding to this monitoring cycle, including body movement status, respiratory rate value, and signal segment quality fraction.

8. The non-contact bed rest data monitoring method according to claim 7, characterized in that, If the monitoring period involves slight body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the signals collected from the monitoring period and several adjacent monitoring periods, including: If the monitoring period is during a period of slight body movement, based on the respiratory signal segments of the monitoring period and several adjacent monitoring periods, the median respiratory rate of each respiratory signal segment is obtained. Based on the signal quality scores of the respiratory signal segments corresponding to the monitoring period and several adjacent monitoring periods, as well as the timing of the monitoring period, the intermediate respiratory rate values ​​corresponding to the multiple respiratory signal segments are weighted to obtain the respiratory rate value of the monitoring period.

9. A non-contact bed rest data monitoring system, characterized in that, It includes a signal acquisition unit, a body movement score evaluation unit, a body movement state classification unit, a physiological data calculation unit, and a data output unit; The signal acquisition unit acquires the collected signals from the bed body, including dynamic force detection signals characterizing vibration changes and static force detection signals characterizing pressure changes, which are of the same origin. The body motion score evaluation unit extracts dynamic force body motion sub-signals based on dynamic force detection signals; Extracting static force body motion signals based on static force detection signals; Based on dynamic force-body motion signals and static force-body motion signals, obtain the motion score representing the intensity of motion of the measured object in each monitoring cycle; The body movement state classification unit determines the body movement state of any monitoring period based on the body movement score corresponding to that monitoring period. The body movement state includes resting state, slight body movement state, and vigorous body movement state. The physiological data calculation unit, for any given monitoring period, obtains the respiratory rate value for that monitoring period based on the body movement state and the acquired signals. If the monitoring period is a resting state, the respiratory rate value for that monitoring period is obtained based on the signal segment of the acquired signal corresponding to that monitoring period. If the monitoring period is a state of slight body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the acquired signals corresponding to that monitoring period and several adjacent monitoring periods. If the monitoring period is a state of intense body movement, the respiratory rate value for that monitoring period is obtained based on the signal segments of the acquired signals corresponding to several adjacent monitoring periods. The data output unit outputs bed rest data, including body movement status and respiratory rate value, for any given monitoring period.

10. A non-contact bedridden data monitoring device, characterized in that, Including a non-contact bed rest data monitoring system as described in claim 9; and The signal acquisition device includes a dynamic force detection unit for acquiring dynamic force detection signals and a static force detection unit for acquiring static force detection signals, used to provide the monitoring system with acquisition signals from the same source; The terminal device is used to interact with the monitoring system, including receiving bedridden data.

Images