A sleep and safety fusion monitoring device for the elderly

Abstract

The application discloses a sleep and safety fusion monitoring device for the elderly, belonging to the technical field of sleep monitoring and home care. The device comprises a non-contact sensing array, a processing module and a prompt module. The sensing array comprises at least one of a millimeter wave radar unit, a piezoelectric sensing unit and a microphone unit, and is preferably combined into a three-source fusion array, for acquiring the respiratory signal, heart rate signal, body movement signal and bed leaving event of a sleeper. The processing module generates sleep state information and safety risk events accordingly, the safety risk events including bed leaving timeout, suspected falling and abnormal breathing or heart rate. The prompt module outputs the above information to the local and remote care end. The application is used for safety monitoring and sleep comfort evaluation for non-therapeutic purposes, has the advantages of no wearing, no compliance burden and controllable privacy, and is suitable for old-age care institutions and home care scenes.

Description

Technical Field

[0001] This invention relates to the field of sleep monitoring and home care technology, specifically to a non-contact sleep and safety integrated monitoring device and method for the elderly. Background Technology

[0002] Population aging has increased concerns about nighttime safety (falls, getting out of bed, abnormal breathing) and sleep quality in both nursing homes and home-based care facilities. Existing monitoring methods have shortcomings: wearable bracelets require continuous wear, leading to poor compliance among the elderly and a tendency for them to fall off at night; medical monitoring mattresses mostly use single-type sensors, are expensive, and are geared towards medical positioning; camera solutions raise privacy concerns. There is a lack of a wearable, non-intrusive product that balances safety monitoring and sleep comfort, and is positioned for non-therapeutic purposes. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a sleep and safety integrated monitoring solution for the elderly that is positioned for non-therapeutic purposes, without wearable devices or single-modal limitations, thereby improving sleep quality and comfort while reducing false alarms and protecting privacy.

[0004] To achieve the above objectives, the present invention adopts the following technical solution (subject to claim 1): A sleep and safety integrated monitoring device for the elderly includes a non-contact sensor array, a processing module, and an alert module. The non-contact sensor array consists of a millimeter-wave radar unit, a piezoelectric sensing unit, and a microphone unit, acquiring respiratory signals, heart rate signals, body movement signals, and bed-exit events. The processing module generates sleep state information and safety risk events. The alert module outputs information to local and remote monitoring terminals. This device is used for non-therapeutic safety monitoring and sleep comfort assessment. Beneficial effects

[0005] (1) The wear-free solution eliminates the burden of wearing and the risk of falling off, resulting in high compliance among the elderly population; (2) The three-source fusion reduces the false alarm rate and increases the event detection rate compared to a single piezoelectric sensor; (3) Local desensitization takes into account both privacy compliance and the needs of remote care; (4) It connects with the long-term care platform to support care scheduling without changing the medical diagnosis conclusion; (5) It connects with the long-term care insurance designated nursing institution platform to achieve precise scheduling of care resources and support large-scale B-end deployment, without changing the medical diagnosis conclusion. Based on theoretical derivation and analysis of the false alarm suppression mechanism of the three-source fusion, and referring to the comparison results of similar fusion solutions in the published literature, the three-source fusion solution is expected to reduce the false alarm rate of leaving the bed by about 45-60% and increase the detection rate of respiratory or heart rate abnormal events by about 18-30% compared to the single piezoelectric solution. The specific values ​​are subject to actual measurement, and the improvement direction is stable in terms of mechanism. Attached Figure Description

[0006] Figure 1 This is a block diagram of the device of the present invention.

[0007] Figure 2 This is a schematic diagram of the deployment scenario of the device of the present invention in a bedroom bed.

[0008] Figure 3 This is a flowchart of the signal processing and event generation process of the processing module of the present invention.

[0009] Figure 2 It also shows the area of ​​interest (ROI) for the bed and the area for caregivers / non-sleepers; Figure 3 It also displays a target classification node between signal acquisition and event interpretation, which is used to distinguish indoor targets into sleeping targets and non-sleeping targets. Detailed Implementation

[0010] The present invention will be further described below with reference to the accompanying drawings and embodiments. The following embodiments are used to explain the present invention and are not intended to limit the scope of protection.

[0011] Example 1: In the non-contact sensor array (10), the millimeter-wave radar unit (11) is placed above the head of the bed to acquire breathing and heart rate micro-movements, the piezoelectric sensor unit (12) is placed on the mattress to acquire body movement and getting out of bed, and the microphone unit (13) acquires snoring and ambient sound. The processing module (20) fuses the three source signals: when the getting out of bed event continues to exceed the first threshold, a timeout event is generated; when breathing or heart rate deviates from the normal range to exceed the second threshold and continues for the duration of the third threshold, an abnormal breathing or heart rate event is generated; and the spatial distribution of body movement is combined to identify suspected falls. Among them, the first threshold (timeout) is 30–120 s, the second threshold (deviation from the normal range) is 15–30% of the baseline value of breathing or heart rate, and the third threshold (abnormal duration) is 10–30 s. The above values ​​can be calibrated by those skilled in the art according to the deployment scenario; the snoring acquired by the microphone unit (13) and the breathing micro-movements acquired by the millimeter-wave radar unit (11) corroborate each other to exclude false getting out of bed events caused by misjudgment of body movement. The privacy protection module (40) completes the processing locally, only uploading the de-identified event identifier. The prompt module (30) provides local prompts at night with low-intensity light and vibration, and notifies the caregiver via the remote message push unit. The long-term care data interface (50) sends events to the long-term care platform for scheduling reference, without changing the medical diagnosis conclusion. In the large-scale deployment for long-term care insurance designated nursing institutions, the long-term care data interface (50) sends safety risk events and care scheduling suggestions to the institution service platform or long-term care insurance service management terminal, and receives scheduling feedback from the platform to adjust the local prompt strategy; the prompt module (30) outputs through the dual channels of the institution care screen and mobile terminal message push. The prompt module (30) can also light up the guide light path to assist in safe getting up when an out-of-bed timeout event is detected, forming an integrated product form of safety monitoring and environmental assistance.

[0012] Example 2 (Nighttime Care Scenario in Institutions): When caregivers patrol or stand guard at night in elderly care institutions, they enter the room, approach the bed, or bend down to care for the patient. The processing module (20) classifies targets based on the spatial distribution of body movement signals and the continuous characteristics of vital signs: targets with continuous breathing or slight heart rate movements, located within the area of ​​interest of the bed, and whose dwell time exceeds the fourth threshold (10–60 s, which can be calibrated by those skilled in the art according to the deployment scenario) are identified as sleeping targets; caregivers who remain outside the area of ​​interest of the bed or whose dwell time is less than the fourth threshold are identified as non-sleeping targets. No bed-out-of-bed timeout events, suspected fall events, or abnormal breathing or heart rate events are generated for non-sleeping targets, thereby suppressing false alarms caused by caregiving behavior.

[0013] Example 3 (Family Accompaniment Scenario): In home-based elderly care, when family members accompany or provide room care, there is a second adult on or beside the bed. The processing module (20) distinguishes different sleepers on the bed based on the spatial distribution of body movements, and identifies family members located in the bedside care area who are not in a bed-lying position as non-sleeping targets; the snoring sound obtained by the microphone unit (13) and the respiratory micro-movements obtained by the millimeter-wave radar unit (11) corroborate each other, and only generates events for sleeper targets. The breathing or body movements of the caregiver are not involved in the determination of time out of bed or abnormal breathing or heart rate. The above identification and response are used for safety monitoring and comfort assessment, and do not constitute a disease diagnosis or treatment conclusion.

[0014] The device is applicable to minors, whose consent to the processing of their physiological information is made by their guardians, and no discriminatory labels are set for minors; for infants and young children who are sleeping, the processing module (20) identifies and prioritizes specific safety risk events such as falling out of bed, kicking off the blankets, resulting in abnormal body temperature, abnormal breathing, etc. This identification and response is used for safety monitoring and comfort assessment, and does not constitute a disease diagnosis or treatment conclusion. Data compliance and privacy protection architecture

[0015] The respiratory rate, heart rate, and body movement data collected by this system / device are physiological information. The processing follows the principles of legality, legitimacy, necessity, and minimization, and the following compliance architecture is constructed: (1) Consent mechanism: In accordance with the Personal Information Protection Law and the Data Security Law, explicit consent is obtained from the information subject or their guardian. For minors, the consent is made by the guardian and no discriminatory labels are set; (2) Local priority and data minimization: Signal processing, event extraction, and model inference are completed locally by default. The original physiological waveforms are not exported or uploaded to the cloud. Only de-identified event identifiers are uploaded, which do not contain biometric features that can identify individuals; (3) Access control and auditing: Role permissions and audit logs are established for accessing, exporting, and sharing physiological data. The retention period is limited to achieving the purpose of security monitoring and is deleted upon expiration; (4) Use limitation: The output is only used for security monitoring and comfort assessment and does not constitute a disease diagnosis or treatment conclusion. The above architecture is uniformly applicable to single-person and multi-person room scenarios.

[0016] The above description is only a partial embodiment of the present invention. All equivalent substitutions or modifications that can be obviously derived by those skilled in the art based on the technical solution of the present invention fall within the protection scope of the present invention.

Claims

1. A sleep and safety integrated monitoring device for the elderly, characterized in that, include: A non-contact sensor array (10) includes at least one of a millimeter-wave radar unit (11), a piezoelectric sensing unit (12), and a microphone unit (13), and is preferably combined into a three-source fusion array. The non-contact sensor array (10) is configured to acquire the sleeper's breathing signals, heart rate signals, body movement signals, and bed exit events. A processing module (20) is configured to generate sleep state information and safety risk events based on the breathing signals, heart rate signals, body movement signals, and bed exit events. The safety risk events include bed exit timeout events, suspected fall events, and abnormal breathing or heart rate events. The prompt module (30) is configured to output the sleep state information and the safety risk event to the local and remote care terminals; wherein, the device is used for safety monitoring and sleep comfort assessment for non-therapeutic purposes.

2. A sleep and safety integrated monitoring method, applied to a device including a non-contact sensor array (10), a processing module (20), and an alerting module (30), characterized in that, include: The non-contact sensor array (10) acquires the sleeper's breathing signals, heart rate signals, body movement signals, and bed exit events. The processing module (20) generates sleep state information and safety risk events based on the breathing signal, heart rate signal, body movement signal and bed exit event. The safety risk events include bed exit timeout event, suspected fall event and breathing or heart rate abnormality event. The processing module (20) verifies the snoring signal obtained by the microphone unit (13) and the breathing micro-movement signal obtained by the millimeter wave radar unit (11) to exclude false bed exit events caused by misjudgment of body movement. The sleep status information and the safety risk event are output by the prompting module (30) to the local and remote care terminals; wherein, the method is used for safety monitoring and sleep comfort assessment for non-therapeutic purposes, and the safety risk event is only used to issue a prompting mark to the care terminal and does not constitute a medical diagnosis or treatment conclusion.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it performs the following steps: acquiring the sleeper's breathing signals, heart rate signals, body movement signals and bed exit events through a non-contact sensor array (10); and generating sleep state information and safety risk events by the processing module (20) based on the breathing signals, heart rate signals, body movement signals and bed exit events, wherein the safety risk events include bed exit timeout events, suspected fall events and abnormal breathing or heart rate events. The snoring signal obtained by the processing module (20) based on the microphone unit (13) and the respiratory micro-movement signal obtained by the millimeter-wave radar unit (11) are mutually verified to eliminate false bed-leaking events caused by misjudgment of body movement; the sleep status information and the safety risk event are output to the local and remote care terminals by the prompting module (30); wherein, the computer program is used for safety monitoring and sleep comfort assessment for non-therapeutic purposes, and the safety risk event is only used to issue prompts and does not constitute a medical diagnosis or treatment conclusion.

4. The use of the sleep and safety integration monitoring device for the elderly population as described in claim 1, characterized in that, The device is deployed in the rooms of elderly care facilities or bedrooms of home-based elderly care facilities for non-therapeutic safety monitoring and sleep comfort assessment of sleepers. The safety monitoring includes prompts for exceeding the time allowed to leave the bed, suspected falls, and abnormal breathing or heart rate, and the prompts do not constitute a medical diagnosis or treatment conclusion.

5. The sleep and safety integrated monitoring device according to claim 1, characterized in that, The processing module (20) is configured to generate an out-of-bed timeout event when an out-of-bed event is detected and the duration exceeds a first threshold, and to generate an abnormal breathing or heart rate event when the respiratory signal or heart rate signal deviates from the normal range by more than a second threshold and lasts for a third threshold duration; the prompting module (30) includes a local audio-visual prompting unit and a remote message push unit, configured to output to the remote care terminal through the dual channels of the institutional care screen and mobile terminal message push, and to prompt at night with low-intensity light and vibration to avoid disturbing sleep.

6. The sleep and safety integrated monitoring device according to claim 1, characterized in that, It also includes a privacy protection module (40) and a long-term care data interface (50); the privacy protection module (40) is configured to complete signal processing locally on the device, and only upload de-identified event identifiers to the remote care terminal without uploading the original physiological waveforms; the long-term care data interface (50) is configured to send the safety risk events and care scheduling suggestions to the long-term care insurance designated nursing institution service platform or the long-term care insurance service management terminal, and receive scheduling feedback from the platform to adjust the local prompt strategy for care scheduling reference rather than medical diagnosis.

7. The sleep and safety integrated monitoring device according to claim 1, characterized in that, The processing module (20) is configured to verify the snoring signal obtained by the microphone unit (13) and the respiratory micro-movement signal obtained by the millimeter-wave radar unit (11) to exclude false bed-leaking events caused by misjudgment of body movement; it also includes a safety assistance execution module, which is configured to output a safety assistance response according to a safety risk event, including lighting up the guide light path and using low-intensity light and vibration to prompt when a bed-leaking timeout event is detected, for the purpose of assisting safety response rather than adjusting sleep comfort; wherein, the output of the safety assistance execution module does not constitute a disease diagnosis or treatment conclusion.

8. The sleep and safety integrated monitoring device according to claim 1, characterized in that, The device is configured to be suitable for single beds and double beds. The processing module (20) is configured to distinguish different sleepers on the same bed based on the spatial distribution of body movement signals. Based on the spatial distribution of the body movement signals and the continuous characteristics of vital signs, the target in the room is divided into sleeper targets and non-sleeper targets. The non-sleeper targets include caregivers or temporary entrants. The safety risk event is generated only for the sleeper targets. Furthermore, the processing module (20) is configured to determine targets that stay outside the bed's area of ​​interest or whose stay duration is less than a fourth threshold as non-sleep targets based on a preset bed area of ​​interest and a stay duration threshold, and not generate bed-out-of-bed-time events, suspected fall events, or abnormal breathing or heart rate events for the non-sleep targets.

Images