Sleep monitoring method, device and system based on vital sign data
By presetting the scoring range and mapping sign data, the total sleep score is calculated, which solves the problem of inconspicuous sleep quality judgment in the existing technology, and achieves a more accurate and intuitive sleep quality reflection, providing users with detailed sleep reports and improving Quality of sleep and health and safety.
Patent Information
- Application Number
- CN202410377192.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-29
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-03-29
AI Technical Summary
The existing sleep monitoring technology cannot intuitively reflect sleep quality, and the judgment of sleep quality is not specific enough.
By presetting the value range of sleep duration, number of wake up, average breathing and turnover, the sign data collected by millimeter wave radar are obtained, the sleep duration value, number of wake up, average breathing and turnover values are calculated, and the sleep duration value is mapped to The corresponding score range obtains the total sleep score, reflecting the quality of sleep in a more specific and intuitive way.
It improves the accuracy and intuitiveness of sleep quality judgments, provides users with more detailed sleep reports, helping to improve sleep quality and health and safety.
Smart Images

Figure CN118058719B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of sleep monitoring, and in particular to a sleep monitoring method, device and system based on vital sign data. Background Art
[0002] The main function of sleep monitoring is to record a person's sleep state, breathing and heartbeat, in order to improve the quality of sleep. Early sleep monitoring products mainly used contact (wearable) methods to measure the user's physical signs during sleep, such as bracelets, watches, sleep belts, sleep pads, ECG monitors, cameras and other products. The output results include EEG, ECG, EMG, breathing average, blood oxygen saturation and other vital sign parameters. The sleep state of a person is calculated and output through the above test results. Wearing wearable devices will affect the user's sleep comfort.
[0003] In this regard, in the past two years, millimeter-wave radar sleep monitoring equipment has detected human breathing, heart rate, movement amplitude and other parameters by sending and receiving electromagnetic waves to comprehensively judge and output sleep results. For example, the sleep quality assessment method, equipment, system and medium based on millimeter-wave radar disclosed in the Chinese invention patent with publication number CN115778352A (hereinafter referred to as Document 1) evaluates the sleep quality of the subject based on the three dimensions of the subject's in / out of bed status, body movement, and respiratory cycle stability during sleep at night based on the real vital signs data of the subject, which can objectively reflect the sleep quality of the subject. And the sleep monitoring device, method and medium based on millimeter-wave radar disclosed in the Chinese invention patent application with publication number CN112806975A (hereinafter referred to as Document 2) can analyze the target echo data to obtain the average heart rate and breathing of the current target human body, and then judge the sleep state.
[0004] However, there are still the following problems in Documents 1 and 2: the judgment of sleep quality is not specific enough and cannot intuitively reflect the sleep quality. Summary of the invention
[0005] The purpose of the present invention is to provide a sleep monitoring method, device and system based on physical sign data to address the deficiencies in the prior art, aiming to improve the results of sleep quality judgment and intuitively reflect sleep quality.
[0006] The present invention achieves the above-mentioned purpose through the following technical scheme: A sleep monitoring method based on vital sign data comprises the following steps:
[0007] Presetting step: presetting at least two value ranges of sleep duration, value ranges of waking up times, value ranges of breathing mean value, and value ranges of turning over times, respectively assigning matching sleep scores to the at least two value ranges of sleep duration, respectively assigning matching waking up scores to the at least two value ranges of waking up times, respectively assigning matching breathing scores to the at least two value ranges of breathing mean value, and respectively assigning matching turning over scores to the at least two value ranges of turning over times;
[0008] Detection steps: obtaining the vital sign data of the subject collected by the millimeter wave radar;
[0009] Numerical value acquisition step: calculating the sleep duration value, the number of times of getting up, the average breathing value and the number of times of turning over based on the physical sign data;
[0010] Scoring step: mapping the sleep duration value to the sleep duration value range to obtain a sleep state score value; mapping the number of getting up value to the number of getting up value range to obtain a getting up score value; mapping the breathing average value to the breathing average value range to obtain a breathing score value; mapping the turning over number value to the turning over number value range to obtain a turning over score value;
[0011] Overall evaluation steps: Add the sleep state score, waking up score, breathing score and turning over score to get the total sleep score.
[0012] As a further solution of the present invention: the sleep duration value range includes a deep sleep duration value range and a light sleep duration value range, the sleep score includes a deep sleep score and a light sleep score, the sleep duration value includes a deep sleep duration value and a light sleep duration value, and the sleep state score value includes a deep sleep score value and a light sleep score value; wherein,
[0013] The presetting step is: presetting at least two value ranges of deep sleep duration, light sleep duration, number of getting-up values, breathing mean value ranges, and number of turning-over values, respectively, assigning matching deep sleep scores to the at least two deep sleep duration value ranges, assigning matching light sleep scores to the at least two light sleep duration value ranges, assigning matching getting-up scores to the at least two number of getting-up value ranges, assigning matching breathing scores to the at least two number of turning-over value ranges, and assigning matching turning-over scores to the at least two number of turning-over value ranges;
[0014] The step of obtaining the values is as follows: based on the physical sign data, calculating the deep sleep duration value, the light sleep duration value, the number of times of getting up, the average breathing value and the number of times of turning over;
[0015] The scoring steps are as follows: mapping the deep sleep duration value to the deep sleep duration value range to obtain a deep sleep score value; mapping the light sleep duration value to the light sleep duration value range to obtain a light sleep score value; mapping the number of getting up value to the number of getting up value range to obtain a getting up score value, mapping the breathing average value to the breathing average value range to obtain a breathing score value; mapping the turning over number value to the turning over number value range to obtain a turning over score value;
[0016] The overall evaluation steps are: add the deep sleep score, light sleep score, waking up score, breathing score and turning over score to get the total sleep score.
[0017] As a further solution of the present invention: the detection step is: the vital sign data includes energy intensity and point cloud orientation data;
[0018] Between the detection step and the value acquisition step, there are also:
[0019] Bed entry assessment step: Analyze the breathing average value according to the detected energy intensity of the subject, and determine whether the preset bed entry breathing threshold is triggered. If so, determine whether the subject is within the preset detection boundary. If so, determine that the subject is in bed;
[0020] Sleep state evaluation step: determine whether the patient is in bed in the bed evaluation step. If so, perform sleep state judgment once every predetermined period. Sleep state includes awake state, light sleep state and deep sleep state, and obtain deep sleep duration value and light sleep duration value;
[0021] Bed exit assessment step: determine whether the subject has left the preset detection boundary. If so, determine whether the radar detects the energy intensity of the subject. If not, start to accumulate the time when the subject leaves the detection boundary. If the accumulated time triggers the preset bed exit time threshold, determine that the subject is in bed exit state and record the number of bed exits.
[0022] Turning over assessment step: determine whether the bed-entering state is confirmed in the bed-entering assessment step. If so, monitor the point cloud orientation data detected by the millimeter-wave radar and the movement intensity based on the energy intensity in real time, and determine whether the movement intensity is sufficient to trigger the preset turning over threshold. If so, determine whether the point cloud orientation data is sufficient to trigger the preset point cloud change threshold. If so, determine it as a turning over action and record the number of turning overs;
[0023] Physiological parameter analysis steps: Based on the analysis of the real-time energy intensity detected by the millimeter wave radar, the energy range that matches the actual breathing changes of the person is screened out, and the frequency is converted to obtain the number of breaths;
[0024] The steps of obtaining the numerical values are as follows: obtaining the deep sleep duration value and the light sleep duration value based on the sleep state evaluation step, obtaining the number of getting up value based on the bed leaving evaluation step, obtaining the average breathing value based on the physiological parameter step, and obtaining the number of turning over value based on the turning over evaluation step.
[0025] As a further solution of the present invention: in the sleep state evaluation step, the sleep state judgment includes:
[0026] Light sleep judgment: within the current scheduled time period, if the number of body movements is greater than the first threshold and the body movement index is less than the second threshold, or the number of body movements is greater than the third threshold and the body movement index is less than the fourth threshold, it is determined that the state has entered the light sleep state from the awake state; if the deep sleep duration value accumulates for a predetermined time, or the number of turning over in the current scheduled time period is greater than the fifth threshold, or the number of body movements in the current scheduled time period is less than the sixth threshold and the body movement index is less than the seventh threshold, it is determined that the state has entered the light sleep state from the deep sleep state;
[0027] Deep sleep judgment: if the light sleep duration accumulates to a predetermined time and the number of body movements in the current predetermined period is less than the eighth threshold and the body movement index is not less than the ninth threshold and the number of turning over is not greater than the tenth threshold, it is determined to be a deep sleep state;
[0028] Awake judgment: Determine whether the patient is in the bed-entering state in the bed-entering assessment step. If so, report the initial sleep state as awake; or, if the deep sleep judgment condition and the light sleep judgment condition are not met, determine the patient as awake.
[0029] As a further solution of the present invention: between the detection step and the bed entry assessment step, the following is also included:
[0030] Person evaluation step: judging whether the motion intensity is sufficient to trigger a preset person threshold through the point cloud position data detected by the millimeter wave radar and the motion intensity based on the energy intensity, and if so, judging whether the point cloud position data triggers a preset distance boundary, and if so, determining that it is a person state;
[0031] The steps between the bed admission assessment step and the value acquisition step also include:
[0032] Standing up assessment step: determine whether the patient is in bed in the bed entry assessment step. If so, monitor the point cloud center of gravity height value in the point cloud orientation data in real time, and determine whether the point cloud center of gravity height value is sufficient to trigger the preset center of gravity height threshold. If so, determine that the patient is in standing up state;
[0033] The sleep state assessment step also includes: the awake state and the light sleep state can be switched between each other, the light sleep state and the deep sleep state can be switched between each other, and the awake state and the deep sleep state will not be switched directly.
[0034] As a further solution of the present invention: the detection step further includes:
[0035] Exercise sign analysis step: divide the detected real-time energy intensity into N parts, determine whether the energy intensity only has breathing and / or heartbeat exercise intensity, if so, assign a value of 1 to the exercise sign parameter, if not, the exercise sign parameter is associated with the exercise intensity, and the assignment range is 2 to N;
[0036] The physiological parameter analysis step also includes: based on the analysis of the real-time energy intensity detected by the millimeter wave radar, screening out the energy range that matches the actual heartbeat changes of the person, and converting the frequency to obtain the number of heartbeats.
[0037] As a further solution of the present invention: after the human evaluation step, the method further comprises:
[0038] Voice activation step: determine whether the state is occupied in the occupied evaluation step, and if so, activate the voice module;
[0039] The voice activation step and the motion sign analysis step also include:
[0040] Struggle assessment step: determine whether the motion sign parameter in the motion sign analysis step is 1 to 5. If not, the motion sign parameter is associated with the exercise intensity, and determine whether the motion sign parameter triggers a preset struggle threshold. If so, determine it as a struggling action;
[0041] Inquiry wake-up step: determine whether it is a struggling action in the struggle assessment step. If so, the voice module sends a preset inquiry voice;
[0042] Standing up prompting step: judging whether the standing up state is reached in the standing up assessment step, and if so, the voice module emits a safety sound;
[0043] Early warning step: determine whether there is anyone answering the inquiry voice, if not, repeat the inquiry wake-up step;
[0044] Alarm step: Determine whether the inquiry wake-up step is repeated at least twice. If so, report the alarm information to the server.
[0045] The present invention also provides another technical solution: a sleep monitoring device based on physical sign data, comprising:
[0046] a preset module, for presetting at least two value ranges of deep sleep duration, light sleep duration, number of getting-up times, breathing mean value range, and number of turning-over times, respectively, assigning matching deep sleep scores to the at least two deep sleep duration value ranges, assigning matching light sleep scores to the at least two light sleep duration value ranges, assigning matching getting-up scores to the at least two number of getting-up value ranges, assigning matching breathing scores to the at least two number of turning-over value ranges, and assigning matching turning-over scores to the at least two number of turning-over value ranges;
[0047] The detection module is used to obtain the vital sign data of the subject collected by the millimeter wave radar;
[0048] A value acquisition module, used to calculate the deep sleep duration value, light sleep duration value, number of getting up values, breathing average value and turning over times value based on the physical sign data;
[0049] A scoring module is used to search for the deep sleep duration value range including the deep sleep duration value to obtain a deep sleep scoring value; search for the light sleep duration value range including the light sleep duration value to obtain a light sleep scoring value; search for the get-up times value range including the get-up times value to obtain a get-up scoring value, search for the breathing average value range including the breathing average value to obtain a breathing scoring value; search for the turn-over times value range including the turn-over times value to obtain a turn-over scoring value;
[0050] The overall evaluation module is used to add the deep sleep score, light sleep score, waking up score, breathing score and turning over score to obtain the total sleep score.
[0051] The present invention also provides another technical solution: a sleep monitoring system based on vital sign data, characterized in that it includes:
[0052] Millimeter wave radar monitoring module, used to transmit millimeter waves and receive reflected waves, and generate vital sign data;
[0053] A main control MCU is connected to the millimeter wave radar monitoring module for receiving and analyzing the vital sign data;
[0054] A wireless communication module is connected to the main control MCU for communication;
[0055] A server, connected to the wireless communication module for communication; and
[0056] A mobile terminal, connected to the server for communication;
[0057] Among them, the main control MCU uses any of the aforementioned sleep monitoring methods based on vital sign data to generate a total sleep score or a sleep quality evaluation, and sends it to the server through the wireless communication module, and the mobile terminal downloads and displays the total sleep score or the sleep quality evaluation from the server.
[0058] As a further solution of the present invention: it also includes a voice module which is communicatively connected or electrically connected to the main control MCU.
[0059] Beneficial effects of the present invention: In this scheme, through the preset step, the sleep duration value, the number of times of getting up, the average breathing value and the number of turning over values are divided into various value ranges, and different sleep score values are assigned according to different value ranges; then, through the detection step, the physical sign data of the subject collected by the millimeter wave radar is obtained, and according to the value acquisition step, the existing method / algorithm is used to calculate and analyze the physical sign data to obtain the sleep duration value, the number of times of getting up, the average breathing value and the number of turning over values; then, through the scoring step, it is compared which value range in the preset step the value obtained in the detection step falls into, and according to the score matched by the value range, the matching scoring value is output, and then the various scoring values are added to obtain the total sleep score value, and then the total sleep score value is sent to the mobile terminal APP via the server in the form of a sleep report, so that there is a more specific and intuitive reference basis for judging the sleep quality of the subject, which better reflects the sleep quality and provides reference information for improving sleep quality and health and safety. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Figure 1 It is a principle block diagram of the millimeter wave radar of the present invention.
[0061] Figure 2 The present invention is a schematic diagram of the sleep quality score report generated by the sleep monitoring method based on physical sign data and the sleep monitoring system based on physical sign data.
[0062] Figure 3 Based on Figure 2 Schematic diagram of sleep assessment using the Sleep Quality Score Report in .
[0063] Figure 4 This is a flowchart of the millimeter wave radar detecting information and analyzing human body vital signs information of the present invention.
[0064] Figure 5 It is a flow chart and waveform chart of the human evaluation step of the present invention.
[0065] Figure 6 It is a flow chart of the bed entry assessment step and the bed exit assessment step of the present invention.
[0066] Figure 7 It is a schematic diagram of the process of determining the sleep state in the sleep state evaluation step of the present invention.
[0067] Figure 8 It is a flowchart of the overall process of the present invention.
[0068] Fig. 9 This is a flowchart of the cloud message push process of the present invention.
[0069] Fig.10 It is a schematic diagram of the waveform of vital sign data detected by the millimeter wave radar of the present invention.
[0070] Fig.11 It is a point cloud orientation diagram of the gesture awakening step of the present invention.
[0071] Fig.12 It is an overall block diagram of the sleep monitoring system based on vital sign data of the present invention.
[0072] Fig.13 This is a power-on flow chart of the sleep monitoring system based on vital sign data of the present invention.
[0073] Fig.14 It is an overall flow chart of the sleep monitoring system based on vital sign data of the present invention.
[0074] Fig.15 It is a working flow chart of the voice module of the present invention. DETAILED DESCRIPTION
[0075] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings.
[0076] like Figure 1 to Figure 15 In an embodiment of the present invention, a sleep monitoring method based on vital sign data is provided, including a preset step, a detection step, a value acquisition step, a scoring step and an overall evaluation step. Among them:
[0077] Presetting step: presetting at least two value ranges of sleep duration, value ranges of waking up times, value ranges of breathing mean value, and value ranges of turning over times, respectively assigning matching sleep scores to the at least two value ranges of sleep duration, respectively assigning matching waking up scores to the at least two value ranges of waking up times, respectively assigning matching breathing scores to the at least two value ranges of breathing mean value, and respectively assigning matching turning over scores to the at least two value ranges of turning over times;
[0078] Detection steps: obtaining the vital sign data of the subject collected by the millimeter wave radar;
[0079] Numerical value acquisition step: based on the physical sign data, the sleep duration value, the number of times of getting up, the average breathing value and the number of times of turning over are calculated; in detail, based on the physical sign data, the sleep start time, the bed leaving time and the sleep end time are also calculated;
[0080] Scoring step: mapping the sleep duration value to the sleep duration value range to obtain a sleep state score value; mapping the number of getting up value to the number of getting up value range to obtain a getting up score value; mapping the breathing average value to the breathing average value range to obtain a breathing score value; mapping the turning over number value to the turning over number value range to obtain a turning over score value;
[0081] Overall evaluation step: Add the sleep state score, wake-up score, breathing score, and turning over score to get the total sleep score. Figure 2 The current sleep model will flexibly allocate the proportion of each score according to the different sleep stages of the person, adjust the weight, and finally output the total score.
[0082] In this solution, through the preset step, the sleep duration value, the number of times of getting up, the average breathing value and the number of turning over times are divided into various value ranges, and different sleep score values are assigned according to different value ranges; then, through the detection step, the physical sign data of the subject collected by the millimeter wave radar is obtained, and according to the value acquisition step, the existing method / algorithm is used to calculate and analyze the physical sign data to obtain the sleep duration value, the number of times of getting up, the average breathing value and the number of turning over times; then, through the scoring step, it is compared to which value range in the preset step the value obtained in the detection step falls into, and according to the score matched by the value range, the matching scoring value is output, and then the various scoring values are added to obtain the total sleep score value, and then the total sleep score value is sent to the mobile terminal APP via the server in the form of a sleep report, so as to have a more specific and intuitive reference basis for judging the sleep quality of the subject, better reflect the sleep quality, and provide reference information for improving sleep quality and health and safety.
[0083] Specifically, Figure 1 and 4 As shown in the figure, the millimeter wave radar transmits electromagnetic waves of fixed frequency through its own antenna, and then synchronously receives the reflection of electromagnetic waves. Through the changes in the frequency and angle of the reflected signal, the information of the movement amplitude and the azimuth information of the moving object are collected. Among them, the movement amplitude of the millimeter wave radar is divided into three levels: the signal strength of the moving object, the signal strength of the breathing movement, and the movement intensity of the human heartbeat. The azimuth information is divided into three directional dimensions of distance information: X (the left and right distance information between the radar horizontal and the ground), Y (the distance information of the moving object from the radar), and Z (the height information of the object vertical to the ground). Through the superposition of the above different dimensional information, different vital signs of personnel are analyzed and judged: based on the large movement intensity and the changes in the azimuth information of the point cloud, the movement status of the personnel (active & still), the turning over, getting up and even struggling of the personnel are analyzed; based on the breathing movement intensity, the breathing depth and number of breaths are analyzed; through the heartbeat movement intensity, the heartbeat frequency is analyzed. Combined with the different vital signs of the personnel analyzed by the above millimeter-wave radar, a series of judgments on sleep are made: when to get in / out of bed, when to start to sleep, judgment of awakeness, light sleep or deep sleep during sleep, judgment of waking up during sleep, integration of the entire sleep information to output a sleep report (such as Figure 2 ), judge and output evaluation of sleep quality (such as Figure 3) to achieve basic care for the daily sleep health of personnel, and to link lighting and sleep-aiding music according to real-time changes in sleep status to help improve the sleep quality of personnel.
[0084] In one embodiment, a sleep monitoring method based on vital sign data includes a presetting step, a detection step, a value acquisition step, a scoring step, and an overall evaluation step.
[0085] Preset steps: Figure 2 As shown, at least 2 deep sleep duration value ranges, light sleep duration value ranges, waking-up frequency value ranges, breathing average value ranges and turning-over frequency value ranges are preset, and corresponding deep sleep scores are assigned to the at least 2 deep sleep duration value ranges, corresponding light sleep scores are assigned to the at least 2 light sleep duration value ranges, corresponding waking-up scores are assigned to the at least 2 waking-up frequency value ranges, corresponding breathing scores are assigned to the at least 2 breathing average value ranges, and corresponding turning-over scores are assigned to the at least 2 turning-over frequency value ranges.
[0086] Detection steps: Obtain the vital sign data of the subject collected by the millimeter wave radar.
[0087] Numerical value acquisition step: based on the vital sign data, calculate the deep sleep duration value, light sleep duration value, number of times of getting up, average breathing value and number of times of turning over.
[0088] The scoring steps are: mapping the deep sleep duration value to the deep sleep duration value range to obtain a deep sleep score value; mapping the light sleep duration value to the light sleep duration value range to obtain a light sleep score value; mapping the number of getting up value to the number of getting up value range to obtain a getting up score value, mapping the breathing average value to the breathing average value range to obtain a breathing score value; mapping the turning over number value to the turning over number value range to obtain a turning over score value.
[0089] Overall review steps: Figure 2 As shown, the deep sleep score, light sleep score, waking up score, breathing score and turning over score are added together to obtain the total sleep score.
[0090] In one embodiment, the presetting step further includes: presetting at least two total sleep duration value ranges and assigning corresponding total sleep duration scores to each. The value acquisition step further includes: summing the deep sleep duration value and the light sleep duration value to obtain a total sleep duration value. The scoring step further includes: mapping the total sleep duration value to the total sleep duration value range to obtain a total sleep score value. The overall evaluation step is: Figure 2As shown in FIG. 1 , the total sleep score, deep sleep score, light sleep score, waking up score, breathing score, and turning over score are added together to obtain the total sleep score. Specifically, by introducing the total sleep duration value range, the total sleep duration value, and the total sleep score value, the total sleep score can more accurately evaluate and reflect the quality of sleep monitoring.
[0091] In another embodiment, Figure 4 As shown, the detection steps are: the vital sign data includes energy intensity and point cloud orientation data, the energy intensity includes large movement intensity, breathing movement intensity and heartbeat movement intensity; the movement intensity of the object being measured is obtained according to the frequency change of the reflected wave of the millimeter wave radar.
[0092] The detection step and the value acquisition step also include: bed entry assessment step, sleep state assessment step, bed exit assessment step, turning over assessment step, physiological parameter analysis step and value acquisition step. Among them:
[0093] Bed entry assessment steps: 1) Detect the angle and distance changes of the subject's movement, reach the predetermined range, and preliminarily determine whether someone is approaching the device; 2) Analyze the breathing average value based on the detected energy intensity of the subject, and determine whether the preset bed entry breathing threshold is triggered. If so, determine whether the subject is within the preset detection boundary. If so, determine that the subject is in bed. For example, Figure 6 and Fig.10 As shown in the figure, according to the installation method, the radar sets the detection boundary for entering the bed (just covering a bed position); when the person lies on the bed, the millimeter-wave radar detects a relatively stable breathing movement intensity, and then compares the distance to confirm that the breathing intensity is within a reasonable range for entering the bed, and the radar reports the bed entering status. Among them, the calculation of the average breathing value can be analyzed by the millimeter-wave radar based on the energy intensity detected in real time, and the real-time detected breathing movement intensity is screened to confirm the energy value or energy range that matches the actual breathing intensity change of the person, and the frequency conversion is obtained to obtain the number of breaths. According to this algorithm, the breathing accuracy will gradually improve with the accumulation of time.
[0094] Sleep state evaluation step: determine whether the patient is in bed in the bed evaluation step. If so, perform sleep state judgment once every predetermined period. Sleep states include awake state, light sleep state and deep sleep state, and obtain deep sleep duration value and light sleep duration value. For example, Figure 7 As shown, the radar has internal restrictions. Only when the user enters the bed state will the sleep state be comprehensively judged every 10 minutes. The sleep state change sequence is limited to: (There is no direct switching between wakefulness and deep sleep states).
[0095] Out-of-bed assessment steps: determine whether the subject has left the preset detection boundary. If so, determine whether the radar detects the energy intensity of the subject. If not, start accumulating the time when the subject leaves the detection boundary. If the accumulated time triggers the preset out-of-bed duration threshold, it is determined as out-of-bed status and the number of out-of-bed times is recorded. For example, when a person leaves the set detection boundary and the radar cannot detect any energy of large body movements, breathing and heartbeat intensity, it starts to accumulate time. If the condition of no one for 30 seconds is met, the out-of-bed status is output.
[0096] Turning over assessment step: Determine whether the bed-entering state is confirmed in the bed-entering assessment step. If so, monitor the point cloud orientation data detected by the millimeter-wave radar and the movement intensity based on the energy intensity in real time, and determine whether the movement intensity is sufficient to trigger the preset turning over threshold. If so, determine whether the point cloud orientation data is sufficient to trigger the preset point cloud change threshold. If so, determine it as a turning over action and record the number of turning over. If the position change of the vital sign data and the height change of the point cloud are detected, it is counted as turning over. For example, after the radar determines that the person has entered the bed, it will detect the movement intensity and point cloud distribution changes of the person lying on the bed in real time. If it is determined that the movement momentum reaches the threshold set for turning over and the point cloud change amplitude is large enough to exceed the set value, the number of turning over is recorded. Report the number of turning over every 10 minutes, and summarize the total number of times at the end of sleep. If you continue to turn over, report an abnormal sleep alarm.
[0097] Physiological parameter analysis steps: The millimeter wave radar transmits signals at a frequency of 2Hz to 30Hz, and based on the analysis of the real-time energy intensity detected by the millimeter wave radar, screens out the energy range that matches the breathing changes of real people, and converts the frequency to get the number of breaths. For example, the millimeter wave radar analyzes the energy intensity detected in real time, screens the real-time detected breathing intensity, confirms that the energy range that matches the intensity changes of real people's breathing is converted to the number of breaths; according to this algorithm, the breathing accuracy will gradually improve with time accumulation. The millimeter wave radar calculates the amplitude of breathing by detecting the movement amplitude of 0.5-10mm, and measures the average value per unit time. The current average value is 10 seconds. If the breathing amplitude becomes smaller, there will be a apnea phenomenon in the record.
[0098] The steps of obtaining the numerical values are as follows: obtaining the deep sleep duration value and the light sleep duration value based on the sleep state evaluation step, obtaining the number of getting up value based on the bed leaving evaluation step, obtaining the average breathing value based on the physiological parameter step, and obtaining the number of turning over value based on the turning over evaluation step.
[0099] In yet another embodiment, Figure 7 and Figure 8 As shown, in the sleep state assessment step, the sleep state judgment includes: light sleep judgment, deep sleep judgment and awake judgment. Among them:
[0100] Light sleep judgment: within the current scheduled time period, if the number of body movements is greater than the first threshold and the body movement index is less than the second threshold, or the number of body movements is greater than the third threshold and the body movement index is less than the fourth threshold, it is determined that the state has entered a light sleep state from the awake state; if the deep sleep duration value accumulates for a predetermined time, or the number of turning over in the current scheduled time period is greater than the fifth threshold, or the number of body movements in the current scheduled time period is less than the sixth threshold and the body movement index is less than the seventh threshold, it is determined that the state has entered a light sleep state from the deep sleep state. For example, Figure 7 As shown, when awake and entering light sleep, the previous scheduled period is 10 minutes, the proportion of body movements greater than 8.0 in the 10 minutes is less than 4.5%, or the proportion of body movements greater than 40 in the 10 minutes is less than 0.5%. When entering light sleep from deep sleep, the deep sleep duration is continuously accumulated for 30 minutes, or the number of turns over in the 10 minutes is ≥4 times, or the proportion of body movements less than 3.5 in the 10 minutes is less than 86.7%.
[0101] Deep sleep judgment: If the light sleep duration accumulates for a predetermined time and the number of body movements in the current predetermined period is less than the eighth threshold, the body movement index is not less than the ninth threshold, and the number of turning over is not greater than the tenth threshold, it is determined to be a deep sleep state. For example, when light sleep enters deep sleep, the light sleep duration accumulates for 50 minutes in a row, the body movement within the 10 minutes is less than 3.5 and the proportion is ≥86.7%, and the number of turning over is ≤4 times.
[0102] Awake judgment: Determine whether the bed-entering state is in the bed-entering state in the bed-entering assessment step. If so, report the initial sleep state as awake; or, if the deep sleep judgment condition and the light sleep judgment condition are not met, determine it as awake. For example, the initial sleep state of bed-entering is awake by default. According to the above light sleep and deep sleep state judgment, if the light sleep and deep sleep state conditions are not met, report the awake state.
[0103] Another embodiment, such as Figure 5 and Figure 8 , between the detection step and the bed entry assessment step, there is also a occupancy assessment step. Between the bed entry assessment step and the value acquisition step, there is also a standing assessment step. Among them:
[0104] Person evaluation step: through the point cloud position data detected by the millimeter wave radar and the motion intensity based on the energy intensity, determine whether the motion intensity is sufficient to trigger the preset threshold of person presence. If so, determine whether the point cloud position data triggers the preset distance boundary. If so, determine that the person presence state is determined. Specifically, determine whether the motion intensity is sufficient to trigger the preset threshold of person presence. If so, determine whether the point cloud position data triggers the preset distance boundary. For example, Figure 5As shown in the figure, the millimeter-wave radar detects a relatively strong movement entering, accompanied by gradually approaching point cloud position information, until the set distance boundary condition is triggered, the radar reports the presence status. After starting to record the presence of people, the presence / absence status is reported every hour. If it occurs for a certain period of time (2-3 hours) continuously, the device reports an abnormal status.
[0105] Standing up assessment step: Determine whether the patient is in bed in the bed assessment step. If so, monitor the point cloud gravity center height value of the point cloud orientation data detected by the millimeter wave radar in real time, and determine whether the point cloud gravity center height value is sufficient to trigger the preset gravity center height threshold. If so, determine that the patient is in standing up state. The standing up state includes half standing up action and full standing up action.
[0106] The sleep state assessment step also includes: the awake state and the light sleep state can be switched between each other, the light sleep state and the deep sleep state can be switched between each other, and the awake state and the deep sleep state will not be switched directly.
[0107] In another embodiment, the detection step further includes: a motion sign analysis step. The motion sign analysis step further includes: an ambulation assessment step. In which:
[0108] Motion sign analysis step: Divide the detected real-time energy intensity into N parts, determine whether the energy intensity is only breathing and / or heartbeat motion intensity, if so, assign a motion sign parameter value of 1, if not, the motion sign parameter is associated with the motion intensity, and the assignment range is 2 to N, preferably, N is 100. For example, the millimeter wave radar divides the energy intensity detected in real time into 100 parts. When the energy intensity is detected to be only breathing / heartbeat intensity, it is considered to be in a static state, and the motion sign parameter is 1; when the detected energy intensity has a large motion intensity, the motion sign parameter is associated with the motion intensity, and the range value is 2 to 100. The greater the motion, the higher the motion sign parameter. For example, in the motion sign parameter, the deep sleep value is 1-3, quietly fiddling with the phone is 5 to 10, turning over is 10 to 30, coughing, struggling is 30 to 70, and walking is 70 to 100. Continuously monitor for 5 minutes, and determine how many times you move in the 5 minutes, and make a sleep score and abnormal alarm based on the number of records.
[0109] Walking assessment step: determine whether the motion sign parameter in the motion sign analysis step is 1. If not, the motion sign parameter is associated with the exercise intensity, and determine whether the motion sign parameter triggers the preset walking threshold. If so, it is determined to be a walking action and the main light is turned on. For example, in the motion sign parameter, the walking action is 70-100. If the walking action is detected, the main light is turned on. If there is no walking, the main light is adjusted, thereby further improving the linkage between the sign action and the lighting.
[0110] The physiological parameter analysis step also includes: according to the analysis of the real-time energy intensity detected by the millimeter-wave radar, the energy range that matches the real person's heartbeat changes is screened out, and the frequency is converted to obtain the number of heartbeats. For example, the millimeter-wave radar analyzes the energy intensity detected in real time, screens the real-time detected heartbeat intensity, confirms the energy range that matches the regular intensity change of the real person's heartbeat, and converts the frequency to obtain the number of heartbeats; according to the algorithm, the heartbeat accuracy will gradually improve as time accumulates.
[0111] In another embodiment, the step of assessing the person further includes: a voice activation step. The step of voice activation and the step of analyzing the movement signs further include: a step of struggling assessment, a step of awakening by inquiry, a step of getting up prompting, a step of early warning, and a step of alarming. Among them:
[0112] Voice activation step: Determine whether the user status is in the user evaluation step. If yes, activate the voice module. For example, the millimeter wave radar sends a user signal. The device starts working. The device activates the voice function. This product is an offline voice trigger. When the device makes the set sound: turn on the light / turn off the light / someone is coming / help, the millimeter wave radar will trigger the remote APP push and make a remote emergency alarm. After receiving the voice trigger, it will reply with voice: the light is on / the light is off / the guardian has been notified.
[0113] Struggle assessment step: determine whether the motion sign parameter in the motion sign analysis step is 1 to 5. If not, the motion sign parameter is associated with the exercise intensity, and determine whether the motion sign parameter triggers a preset struggle threshold. If so, determine it as a struggling action.
[0114] Inquiry wake-up step: determine whether it is a struggling action in the struggle assessment step. If so, the voice module sends a preset inquiry voice.
[0115] Getting up prompt step: Determine whether the person is in the getting up state in the getting up assessment step. If so, the voice module emits a safety sound. For example, when a person gets up at night, the clock and auxiliary lighting will automatically light up according to the getting up action, and a safety sound will be prompted. When the person walks, the lighting will be provided, and the light will be automatically turned off after getting into bed.
[0116] Early warning step: determine whether there is anyone answering the inquiry voice, if not, repeat the inquiry wake-up step.
[0117] Alarm step: Determine whether the inquiry wake-up step is repeated at least twice. If so, report the alarm information to the server.
[0118] A sleep monitoring device based on vital sign data includes: a preset module, a detection module, a value acquisition module, a scoring module and an overall evaluation module. Among them:
[0119] a preset module, for presetting at least two value ranges of deep sleep duration, light sleep duration, number of getting-up times, breathing mean value range, and number of turning-over times, respectively, assigning matching deep sleep scores to the at least two deep sleep duration value ranges, assigning matching light sleep scores to the at least two light sleep duration value ranges, assigning matching getting-up scores to the at least two number of getting-up value ranges, assigning matching breathing scores to the at least two number of turning-over value ranges, and assigning matching turning-over scores to the at least two number of turning-over value ranges;
[0120] The detection module is used to obtain the vital sign data of the subject collected by the millimeter wave radar;
[0121] A value acquisition module, used to calculate the deep sleep duration value, light sleep duration value, number of getting up values, breathing average value and turning over times value based on the physical sign data;
[0122] A scoring module is used to search for the deep sleep duration value range including the deep sleep duration value to obtain a deep sleep scoring value; search for the light sleep duration value range including the light sleep duration value to obtain a light sleep scoring value; search for the get-up times value range including the get-up times value to obtain a get-up scoring value, search for the breathing average value range including the breathing average value to obtain a breathing scoring value; search for the turn-over times value range including the turn-over times value to obtain a turn-over scoring value;
[0123] The overall evaluation module is used to add the deep sleep score, light sleep score, waking up score, breathing score and turning over score to obtain the total sleep score.
[0124] A sleep monitoring system based on vital sign data, such as Figure 12 to Figure 15 As shown, it includes: meter-wave radar monitoring module, main control MCU, wireless communication module, server and mobile terminal. Among them:
[0125] Meter-wave radar monitoring module, used to transmit millimeter waves and receive reflected waves, and generate vital sign data and location information;
[0126] A main control MCU is connected to the millimeter wave radar monitoring module for receiving and parsing the vital sign data to control other functional modules, specifically, voice, lighting, SOS modules and other sensor modules;
[0127] A wireless communication module is connected to the main control MCU for communication, and includes a wireless module;
[0128] A server, connected to the wireless communication module for communication;
[0129] A mobile terminal, connected to the server for communication;
[0130] The main control MCU generates a total sleep score or sleep quality evaluation using any of the aforementioned sleep monitoring methods based on vital sign data, and sends it to the server through the wireless communication module, and the mobile terminal downloads and displays the total sleep score or sleep quality evaluation from the server. When installed, it is within 3.5 meters above the head of the bed or the top of the room, or in the horizontal direction of the bed surface (bedside table, table), facing the body part.
[0131] Another embodiment, such as Figure 12 to Figure 15 As shown, the sleep monitoring system based on vital sign data also includes a voice module that is communicatively connected or electrically connected to the main control MCU.
[0132] A lighting control method based on vital sign data, such as Figure 12 to Figure 15 , including the following steps: a standing up monitoring step, a half-standing up evaluation step, a full-standing up evaluation step and a first light control step. Among them:
[0133] Stand-up monitoring steps: obtaining the point cloud center of gravity height value of each point in the point cloud azimuth data detected by the millimeter wave radar;
[0134] Half-rise evaluation step: determining whether the number of points whose point cloud gravity center height values trigger a preset gravity center height threshold within a first set time exceeds a preset half-rise threshold, and if so, determining it as a half-rise action;
[0135] Full-stand evaluation step: determining whether the number of points whose point cloud gravity center height values trigger a preset gravity center height threshold within a second set time exceeds a preset full-stand threshold, and if so, determining it as a full-stand action;
[0136] The first light control step: if the person being tested is half-standing up, the light is controlled to work in the first state; if the person being tested is fully standing up, the light is controlled to work in the second state. Specifically, the first stage of standing up is a half-standing up action (possible standing up), and the second stage of standing up is a full standing up action (confirmed standing up). The first state and the second state can be set to different brightness or different lighting effects, such as the brightness of the light for the half-standing up action is lower than the brightness of the light for the full standing up action.
[0137] For example, Fig.10 As shown, the radar has internal restrictions and will only start to judge whether to get up at night after entering the bed state; in the bed state, the radar judges the current point cloud center of gravity height value in real time, and sets a center of gravity height threshold through the previous model acquisition. In the actual judgment process, if ≥3 points exceeding the threshold appear within 4 seconds, the first stage of getting up (possible getting up) will be reported. If the cumulative total number of points exceeding the threshold is ≥5 within the next 5 seconds, the second stage of getting up (confirmed getting up) will be reported. Figure 4 and Figure 8As shown, the radar vital sign action is linked with the lighting, and the linkage effect is achieved through the standing up judgment (possible standing up / confirmed standing up) in the radar function. When a person sits up, the radar judges and reports the possibility of standing up, and the night light brightness is linked to 30%, giving the person who stands up a brightness adaptation process; when the person stands up completely, the radar judges and reports the confirmation of standing up, and the night light brightness is linked to 100%, realizing a complete night lighting function.
[0138] In the light control method based on vital sign data, the half-standing up assessment step and the full-standing assessment step are used to correspondingly identify the half-standing up action (possible standing up) and the full-standing up action (confirmed standing up) of the subject. The half-standing up action matches the first state of the lighting lamp, and the full-standing action matches the second state of the lighting lamp to achieve a complete night lighting function. The first state and the second state can be set to different brightness or different lighting effects, so that the lighting lamp is linked to the user's vital sign action and added to the sleep monitoring equipment to achieve diverse functions and improve the user experience. When set to different brightness, it can adaptively adjust the lighting, save energy and protect the environment, and intelligently adjust the brightness instead of keeping it full light or full dark all the time. At the same time, it also provides a comfortable lighting environment to avoid damage to the human eye caused by sudden strong light. There is no need to operate the switch laboriously, and there is no need to grope in the dark, thereby reducing the risk of accidents such as falls.
[0139] In one embodiment, Figure 4 and Figure 8 As shown, the full-standing assessment step is: determine whether the half-standing assessment step is a half-standing action, and if so, determine whether the number of points whose point cloud center of gravity height value triggers the preset center of gravity height threshold within the second set time exceeds the preset full-standing threshold, and if so, determine it as a full-standing action. Specifically, the half-standing assessment step is performed first and then the full-standing assessment step is performed. When a person sits up, the radar determines and reports that he may stand up, and the brightness of the lighting is 30%, giving the person who stands up at night a brightness adaptation process; when the person stands up completely, the radar determines and reports that he confirms that he stands up, and the brightness of the lighting is 100%, realizing a complete night lighting function.
[0140] In addition, after confirming that they have gotten up, the voice module can be controlled to emit a sound, such as a buzzing sound, to remind the elderly that they have gotten up. Because most elderly people feel dizzy when getting up and may fall, making a sound when getting up can serve as a good reminder.
[0141] In yet another embodiment, Figure 4 and Fig.11 As shown, the light control method based on vital sign data also includes the following steps:
[0142] Gesture wake-up step: The millimeter-wave radar detects the point cloud position data and the angle and speed of the electromagnetic wave signal reflection in real time, and obtains the energy intensity and speed value reflected by the moving object through the changes in the frequency and angle of the electromagnetic wave. It determines whether the point cloud position data triggers the preset gesture threshold. If so, it determines whether the speed value triggers the preset gesture speed threshold. If so, it controls the lighting of the lamp. Among them, the gesture threshold includes a gesture direction threshold or a gesture position threshold, such as waving left, waving right, waving continuously, double-clicking and clapping, etc. Energy intensity or motion intensity can also be added as an auxiliary judgment condition for triggering the gesture wake-up step, and the motion intensity includes the intensity of large movements. Preferably, the clock light is controlled to light up. For example, such as Fig.11 The millimeter-wave radar detects the distribution changes and speed values of the point cloud in real time. If the speed value reaches the set threshold and the point cloud change amplitude is large enough to exceed the gesture recognition setting value, the gesture command is reported. For example, when a person waves at the device, the device will automatically light up the clock to indicate the time; or when a person waves at the device, the device will flash the light, thereby improving the linkage between the action and the lighting effect.
[0143] like Fig.12 As shown in the figure, the millimeter wave radar monitoring module is connected to the server through the wireless communication module, and the relevant subscription information is displayed on the mobile phone App or the background. Fig.13 As shown in the figure, the power-on process is triggered by the physical button until the device is normally connected to the server, and the device's LED light status and data are reported normally. Fig.14 As shown in the figure, the millimeter-wave radar detects the presence and sleep of the human body. The radar sends the detected information status to the controller MCU (i.e. the main control MCU) through the serial port connection method. The MCU analyzes the radar detection attribute status separately. When the person is in the bed sleeping state, the MCU combines whether there are abnormal breathing and heartbeat during the sleep process, whether there are abnormal struggles during the sleep process, and controls the voice system interaction to issue a safety inquiry. If there is a state of getting up, the MCU will also control the lighting to light up. At the same time, the wireless module reports these states to the cloud server, and the mobile App can subscribe to messages. Fig.15 As shown, the sleep monitoring system supports offline voice function and can actively or passively interact with voice. When a person is in danger, he or she can directly call for help through the intercom, and the person can reply according to the voice inquiry. When a person is detected to have abnormal breathing or struggling during sleep, the voice system will actively ask the person whether he or she needs help and report the alarm information to the service center.
[0144] Specifically, the main control MCU will filter and process some of the information according to the human presence information, sleep status information, and voice intercom information reported by the millimeter wave radar, and then transmit it to the cloud through the MQTT protocol. When a person approaches the device at a certain distance, the device light will turn on, or the light will gradually brighten when someone gets up; when a person is away from the device at a certain distance, the device light will turn off, or someone will leave the bed, and the person's departure can also be recorded. When a person approaches the bed, the breathing and heartbeat counts begin to assess whether it is in a stable sleep state, and the deep and shallow sleep states are assessed at intervals of 5-10 minutes. At the same time, the sleep duration will also be recorded. After a certain sleep duration, a sleep quality report will be output. While sleeping, the local algorithm will determine the turning over and struggling state based on the body movement characteristics of the person, and issue an alarm in time. The cloud application mainly relies on the data status. After filtering the data, the wireless module transmits the status of the local application to the cloud background, and then displays the data to the user through the app. At the same time, the user can also control the device and set functions through the management background and app. Wireless communication follows the standard protocols MQTT, zigbee3.0, and GATT, and transmits data to the background. After the background stores and records, it makes real-time linkage according to the importance. The main alarm status has one-click call, sleep struggle, voice call, etc., which can be customized through some cloud services to realize app message push, SMS push, or phone notification, etc., and the background can also associate the guardian. The mobile terminal App can set the function of the device, and can also check the real-time alarm information and historical records, or set the sleep time period according to the user's sleep habits, and compare the sleep quality reports of different cycles on the page.
[0145] Finally, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A sleep monitoring method based on vital sign data, characterized in that: The steps include: Presetting step: presetting at least two value ranges for deep sleep duration, light sleep duration, number of times of getting up, average breathing rate, and number of times of turning over, and assigning corresponding scores to different value ranges; Detection step: obtaining the subject's vital sign data collected by the millimeter wave radar, wherein the vital sign data includes energy intensity and point cloud orientation data; Bed entry assessment step: Analyze the average breathing value based on the detected energy intensity of the subject and determine whether the preset bed entry breathing threshold is triggered. If so, determine whether the subject is within the preset detection boundary. If so, determine that the subject is in bed; Sleep state assessment step: determining whether the patient is in bed in the bed assessment step. If so, performing a sleep state assessment every predetermined period to obtain a deep sleep duration value and a light sleep duration value; Numerical value acquisition step: calculating the deep sleep duration value, light sleep duration value, number of times of getting up value, average breathing value and number of times of turning over value based on the physical sign data; Scoring step: mapping the deep sleep duration value to the deep sleep duration value range to obtain a deep sleep score value; mapping the light sleep duration value to the light sleep duration value range to obtain a light sleep score value; Mapping the number of times of getting up to the number of times of getting up value range to obtain a getting up score value, and mapping the average breathing value to the average breathing value range to obtain a breathing score value; Mapping the turning number value to the turning number value range to obtain a turning score value; Overall evaluation step: Add up the deep sleep score, light sleep score, waking up score, breathing score, and turning over score to get the total sleep score; Wherein, in the sleep state evaluation step, the sleep state judgment includes: Light sleep judgment: If the number of body movements in the current predetermined time period is greater than the first threshold and the body movement index is less than the second threshold, or the number of body movements is greater than the third threshold and the body movement index is less than the fourth threshold, it is determined that the person has entered a light sleep state from the awake state; if the deep sleep duration value accumulates for a predetermined time, or the number of tossing and turning in the current predetermined time period is greater than the fifth threshold, or the number of body movements in the current predetermined time period is less than the sixth threshold and the body movement index is less than the seventh threshold, it is determined that the person has entered a light sleep state from the deep sleep state; Deep sleep judgment: if the light sleep duration accumulates for a predetermined time and the number of body movements in the current predetermined period is less than the eighth threshold, the body movement index is not less than the ninth threshold, and the number of tossing and turning is not greater than the tenth threshold, the state is determined to be deep sleep; Awake judgment: Determine whether the patient is in the bed-entering state in the bed-entering assessment step. If so, report the initial sleep state as awake; or, if the deep sleep judgment condition and light sleep judgment condition are not met, determine the patient as awake.
2. The sleep monitoring method based on vital sign data according to claim 1, characterized in that: The sleep state assessment step also includes: Bed exit assessment step: determine whether the subject has left the preset detection boundary. If so, determine whether the radar detects the subject's energy intensity. If not, start accumulating the time the subject has left the detection boundary. If the accumulated time triggers the preset bed exit time threshold, determine that the subject has left the bed and record the number of bed exits. Turnover assessment step: Determine whether the patient is in bed in the bed entry assessment step. If so, monitor the point cloud orientation data detected by the millimeter-wave radar and the motion intensity based on the energy intensity in real time, and determine whether the motion intensity is sufficient to trigger a preset turning threshold. If so, determine whether the point cloud orientation data is sufficient to trigger a preset point cloud change threshold. If so, determine that it is a turning action and record the number of turns. Physiological parameter analysis steps: Based on the analysis of the real-time energy intensity detected by the millimeter-wave radar, the energy range that matches the actual breathing changes of the person is screened out, and the frequency is converted to the number of breaths; The steps for obtaining the numerical values are: obtaining the deep sleep duration value and the light sleep duration value based on the sleep state evaluation step, obtaining the number of getting up value based on the bed leaving evaluation step, obtaining the average breathing value based on the physiological parameter step, and obtaining the number of turning over value based on the turning over evaluation step.
3. The sleep monitoring method based on vital sign data according to claim 2, characterized in that: Between the detection step and the bedside assessment step also includes: A presence assessment step: using the millimeter-wave radar's point cloud orientation data and the energy-intensity-based motion intensity to determine whether the motion intensity is sufficient to trigger a preset presence threshold; if so, determining whether the point cloud orientation data triggers a preset distance boundary; if so, determining a presence state; Between the bed admission assessment step and the value acquisition step also includes: Standing-up assessment step: Determine whether the patient is in bed in the bed entry assessment step. If so, monitor the point cloud center of gravity height value in the point cloud orientation data in real time and determine whether the point cloud center of gravity height value is sufficient to trigger the preset center of gravity height threshold. If so, determine that the patient is in standing-up state. The sleep state assessment step also includes: switching between the awake state and the light sleep state, switching between the light sleep state and the deep sleep state, and not directly switching between the awake state and the deep sleep state.
4. The sleep monitoring method based on vital sign data according to claim 2 or 3, characterized in that: The detection step is followed by: Exercise sign analysis step: Divide the detected real-time energy intensity into N parts, and determine whether the energy intensity only includes breathing and / or heartbeat exercise intensity. If so, the exercise sign parameter is assigned a value of 1. If not, the exercise sign parameter is associated with the exercise intensity and the assignment range is 2 to N. The physiological parameter analysis step also includes: based on the analysis of the real-time energy intensity detected by the millimeter wave radar, screening out the energy range that is consistent with the actual heartbeat changes of the person, and converting the frequency to obtain the heartbeat number.
5. The sleep monitoring method based on vital sign data according to claim 4 as cited in claim 3, characterized in that: The evaluation step is followed by: Voice activation step: determine whether the state is occupied in the occupied evaluation step, and if so, activate the voice module; The voice activation step and the motion sign analysis step also include: Struggle assessment step: determining whether the motion sign parameter in the motion sign analysis step is between 1 and 5. If not, the motion sign parameter is associated with the exercise intensity, and determining whether the motion sign parameter triggers a preset struggle threshold. If so, determining that it is a struggling action; Inquiry awakening step: determine whether the struggle action is determined to be a struggle action in the struggle assessment step. If so, the voice module issues a preset inquiry voice; Stand-up prompting step: determining whether the state is the stand-up state in the stand-up assessment step. If so, the voice module emits a safety sound; Early warning step: determine whether there is anyone answering the inquiry voice, if not, repeat the inquiry wake-up step; Alarm step: Determine whether the inquiry wake-up step is repeated at least twice. If so, report the alarm information to the server.
6. A sleep monitoring system based on vital sign data, characterized in that: include: Millimeter-wave radar monitoring module, used to transmit millimeter waves, receive reflected waves, and generate vital sign data; as well as A main control MCU is connected to the millimeter wave radar monitoring module for receiving and analyzing the vital sign data; The main control MCU generates a total sleep score or a sleep quality evaluation using the sleep monitoring method based on vital sign data described in any one of claims 1 to 5.
7. The sleep monitoring system based on vital sign data according to claim 6, characterized in that: It also includes a voice module that is communicatively connected or electrically connected to the main control MCU.
Citation Information
Patent Citations
Sleep monitoring device and method based on millimeter wave radar and medium
CN112806975A
Sleep quality evaluation method, device and system based on millimeter wave radar and medium
CN115778352A
Night lamp control method and device based on radar, radar and storage medium
CN115968080A
Sleep quality determination method and device, equipment and medium
CN117045201A
Monitoring system for communication between smart ward and millimeter wave sign radar
CN220651405U
