Desktop health monitoring system and method based on non-visual perception

Abstract

The invention relates to the technical field of health monitoring, and particularly discloses a desktop health monitoring system and method based on non-visual perception, a data acquisition module is used for non-inductive acquisition of user state data, and a data processing and decision module is used for processing the acquired data and evaluating the health state of a user. And the interaction and feedback module is used for providing feedback and intervention for the user based on the health state evaluation result. According to the invention, non-visual sensors such as millimeter wave radar and the like are adopted, and any face and environment optical images are not obtained from the physical principle, so that the privacy concerns of users are thoroughly eliminated, and obstacles are cleared for large-scale application of the technology in the office environment. Through a multi-sensor fusion algorithm, the limitation of a single sensor is overcome, high-precision attitude reconstruction and physiological parameter extraction without visual information are realized, the technical path is unique, and the effect is reliable.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of health monitoring, in particular to a desktop health monitoring system and method based on non-vision perception. BACKGROUND

[0002] With the expansion of the office population and the increase of working hours, professional health problems such as cervical spondylosis, lumbar problems, digital eye fatigue and chronic stress are increasingly prominent. The current market solutions mainly have the following defects:

[0003] 1. Camera-based solution: using computer vision technology to analyze user posture. The biggest disadvantage is that it infringes on user privacy, causing user psychological discomfort and low acceptance, and the effect is poor in insufficient light or obstruction.

[0004] 2. Wearable sensor solution: such as posture sensors worn on the back or clothes. The problem is that users need to wear them, the experience is not good, and they are easy to forget, and cannot achieve truly unobtrusive monitoring. The function is also relatively single, usually limited to posture reminders.

[0005] 3. Single-function device: such as screen hanging lights that only monitor ambient light, or software that only reminds you to rest. These solutions are fragmented and cannot provide users with an integrated health management experience.

[0006] Therefore, there is an urgent need in the art for a technical solution that can comprehensively monitor health indicators and fundamentally eliminate the risk of privacy leakage. SUMMARY

[0007] The technical problem to be solved by the present application is to provide a desktop health monitoring system and method based on non-vision perception, which realizes accurate monitoring of human posture, visual distance and physiological parameters without collecting any optical images through non-vision sensors, fundamentally solving the privacy problem. Integrating multiple health monitoring functions into one, achieving comprehensive management of skeletal muscle, vision and stress health. Through emotional and artistic interaction design, user experience and compliance are improved.

[0008] To solve the above technical problems, the technical solution provided by the present application is: a desktop health monitoring system based on non-vision perception, comprising a data acquisition module, a data processing and decision module, and an interaction and feedback module; the data acquisition module is used for unobtrusive acquisition of user state data, the data processing and decision module is used for processing and evaluating the user's health status, and the interaction and feedback module is used for providing feedback and intervention to the user based on the health status evaluation result.

[0009] Further, the data acquisition module comprises a millimeter wave radar sensor, a time-of-flight sensor, and an ambient light sensor; the millimeter wave radar sensor is used to emit a frequency-modulated continuous wave and receive a reflected signal, generate point cloud data containing user head, neck, and shoulder position information, and extract a chest micro-motion signal caused by heartbeat and respiration through phase demodulation; the time-of-flight sensor is used to emit near-infrared light to the user direction and measure the round-trip time to obtain the visual range of the user's eyes and the device screen; the ambient light sensor is used to monitor the ambient light intensity and color temperature.

[0010] Further, the data processing and decision module adopts an embedded processor, which internally has a sensor fusion algorithm, a physiological parameter calculation unit, and a health state evaluation model; the sensor fusion algorithm is used to fuse the point cloud data of the millimeter wave radar sensor and the visual range data of the time-of-flight sensor, the physiological parameter calculation unit is used to extract physiological-related data in the user state data, and the health state evaluation model is used to judge the user's health state.

[0011] Further, the sensor fusion algorithm reconstructs a real-time skeletal key point model of the user's upper body through the fused data, and calculates posture indicators such as head and neck inclination angle, cervical spine load moment, and sitting posture symmetry; the physiological parameter calculation unit extracts heart rate and heart rate variability data from the reflected signal of the millimeter wave radar sensor.

[0012] Further, the health state evaluation model comprehensively judges the user's health state based on the posture indicators, visual range data, heart rate and heart rate variability data, and ambient light intensity and color temperature data, and the health state includes normal, mild fatigue, severe fatigue, and bad posture.

[0013] Further, the interaction and feedback module comprises a display screen; the display screen is used as a digital photo frame to cycle user-defined pictures and display health state data in real time in the form of generative art, and display guiding intervention content when the health state evaluation model determines that intervention is needed.

[0014] Further, the interaction and feedback module further comprises an optional electroluminescent element or RGB light strip; the electroluminescent element or RGB light strip provides a non-invasive state prompt through color change, and adopts blue to represent concentration and orange to represent fatigue.

[0015] The application also provides a health monitoring method based on the above-mentioned system, comprising the following steps:

[0016] S1: continuously and non-invasively collecting non-visual perception data of the user through the data acquisition module;

[0017] S2: running sensor fusion algorithm and health state assessment model on the embedded processor end of the data processing and decision module, converting the collected data into health indicators with medical significance in real time;

[0018] S3: judging whether the current health state of the user exceeds the preset threshold based on the health indicators;

[0019] S4: if the preset threshold is exceeded, starting personalized and guided digital intervention through the interaction and feedback module;

[0020] S5: continuously repeating steps S1-S4 to form a closed-loop management of "monitoring-evaluation-intervention".

[0021] Compared with the prior art, the present application has the following advantages:

[0022] The present application uses non-vision sensors such as millimeter wave radars, and does not obtain any facial or environmental optical images from a physical principle, thereby completely eliminating the privacy concerns of users and clearing the obstacles for large-scale application of the technology in office environments.

[0023] The present application creatively integrates four monitoring functions of posture, sight distance, ambient light and pressure into a single device, thereby providing users with an unprecedented comprehensive health management experience and avoiding the desktop clutter and high cost caused by multiple devices.

[0024] The present application overcomes the limitations of a single sensor through a multi-sensor fusion algorithm, realizes high-precision posture reconstruction and physiological parameter extraction without vision information, and has a unique technical path and reliable effect.

[0025] The present application deeply integrates health management and emotional design, changes passive and negative "reminders" into active and positive "companionship", and greatly improves the user's willingness to use and health improvement effect. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a system block diagram of a desktop health monitoring system based on non-vision perception according to the present application.

[0027] Figure 2 is a flowchart of a desktop health monitoring method based on non-vision perception according to the present application.

[0028] Figure 3 is a flowchart of a multi-sensor fusion algorithm.

[0029] Figure 4 is a flowchart of health state assessment. DETAILED DESCRIPTION

[0030] Various exemplary embodiments of the present application will be described in detail below with reference to the drawings. Note that the relative arrangement, numerical expressions, and numerical values of components and steps set forth in these embodiments are not limiting to the scope of the present application unless specifically stated otherwise.

[0031] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the scope of the application or its application or uses.

[0032] Techniques, methods, and apparatus known to those of ordinary skill in the relevant art can not be discussed in detail herein, but should be considered as part of the specification, where appropriate.

[0033] In all examples shown and discussed herein, any specific values should be interpreted as merely illustrative, and not as a limitation. Thus, other examples of the exemplary embodiments can have different values.

[0034] A non-visual perception-based desktop health monitoring system and method will be described in further detail below with reference to the accompanying drawings.

[0035] In conjunction with the accompanying Figures 1-4 , the specific implementation process of the non-visual perception-based desktop health monitoring system and method is as follows:

[0036] The non-visual perception-based desktop health monitoring system provided by the present application includes a data acquisition module, a data processing and decision module, and an interaction and feedback module, which are connected to each other through embedded communication interfaces to form a complete "data acquisition-processing-feedback" link. The specific description is as follows:

[0037] The core function of the data acquisition module is "non-intrusive acquisition of user state data", which avoids interference with the user's daily work. It includes three types of non-visual sensors, and the functions and working principles of each sensor are as follows:

[0038] Millimeter wave radar sensor: TI's IWR6843AOP millimeter wave radar module is used to emit a frequency-modulated continuous wave and receive the reflected signal from the object, generating three-dimensional point cloud data containing the position information of the user's head, neck, and shoulder. At the same time, the phase demodulation technology is used to extract the chest micro-vibration signal caused by heartbeat and respiration, providing raw data for subsequent physiological parameter calculation;

[0039] Time-of-flight (ToF) sensor: ST's VL53L5CX time-of-flight sensor is used to emit near-infrared light (non-visible light, which does not affect the user's vision) towards the user's eye, and by measuring the round-trip time of the light signal, the real-time distance between the user's eye and the device screen is calculated, providing a basis for visual fatigue assessment;

[0040] Ambient light sensor: TSL2585 ambient light sensor from AMS is used to monitor the light intensity and color temperature of the desktop environment in real time, and determine whether the ambient light meets the visual health standard.

[0041] The data processing and decision module is the core control unit of the system, which uses i.MXRT1170 crossover MCU from NXP as the embedded processor, with high-performance data processing capability and low-power consumption characteristics. It has built-in sensor fusion algorithm, physiological parameter calculation unit and health status evaluation model, with the following specific functions:

[0042] Sensor fusion algorithm: The point cloud data of millimeter wave radar and the range data of ToF sensor are fused and processed, and the real-time skeletal key point model of the user's upper body (including head, cervical spine, shoulder, trunk, etc.) is reconstructed through the "clustering-tracking-modeling" process. Based on the biomechanical principle, the posture index is calculated; among them, the head and neck inclination angle reflects the inclination angle of the head and trunk, the cervical spine load moment is estimated by the formula "moment = head mass x gravity acceleration x force arm x sin(inclination angle)" (the head mass is defaulted to 5 kg, and the force arm is the distance from the head center of gravity to the cervical spine), and the sitting symmetry is judged by the height difference between the left and right shoulders.

[0043] Physiological parameter calculation unit: heart rate and heart rate variability (HRV) data are extracted from the chest micro-motion signals collected by the millimeter wave radar, and the time domain index of HRV can be used as the core basis for evaluating the user's cognitive stress and fatigue level;

[0044] Health status evaluation model: The posture index (head and neck inclination angle, cervical spine load moment), range data, HRV data and ambient light data are integrated to judge the user's health status using the "threshold judgment + weighted scoring" mechanism, including: normal (all indicators are within the safety threshold), mild fatigue (HRV index is slightly low or range is occasionally less than 50 cm), severe fatigue (HRV index is consistently low and cervical spine load > 2.5 N.m), and bad posture (head and neck inclination angle > 30° for more than 10 seconds).

[0045] The interaction and feedback module is used to provide intuitive and friendly health feedback and intervention guidance to the user, avoiding the harshness of traditional "popup reminders", including the following components:

[0046] Display screen: 4-inch IPS LCD display screen is used, with three main functions: first, as a digital photo frame, rotating user-defined uploaded pictures; second, real-time visualization of health status data through generative art form; third, display guiding content when intervention is needed, such as 90-second breathing training animation and cervical spine stretching guide diagram;

[0047] Optional electroluminescent element or RGB light strip: integrated at the edge of the device, providing non-invasive state prompts through color changes - blue represents that the user is currently in a "focus state" (normal health indicators), orange represents "mild fatigue", and red represents "immediate intervention is needed" (such as excessive cervical load), avoiding the user from frequently checking the screen to perceive their own health status.

[0048] Based on the above system, the application further provides a desktop health monitoring system and method based on non-visual perception, comprising the following steps:

[0049] S1: Data acquisition starts: after the system is powered on, the millimeter wave radar sensor, ToF sensor and ambient light sensor of the data acquisition module start simultaneously, continuously and unobtrusively collecting point cloud data, sight distance data, ambient light data and chest cavity micro-motion signals of the user;

[0050] S2: Data processing and index conversion: the embedded processor runs the sensor fusion algorithm to convert the point cloud data and sight distance data into posture indicators such as head and neck inclination angle and cervical load moment; at the same time, the physiological parameter calculation unit extracts heart rate and HRV data; the health status evaluation model integrates the posture indicators, physiological parameters and ambient light data to generate a health score with medical significance;

[0051] S3: Health status judgment: the system compares the health score and each single indicator with the preset threshold value, and if any of the following conditions is met, it is determined that "intervention is needed": ① cervical load > 2.5 N.m and lasts for 30 seconds; ② sight distance < 50 cm and lasts for 15 seconds; ③ HRV indicator (RMSSD) < 20 ms for 20 seconds; ④ ambient light intensity < 200 lux or > 800 lux;

[0052] S4: Personalized intervention starts: the interaction and feedback module starts the corresponding function according to the intervention requirement: if the cervical load is excessive, the display screen smoothly transitions from the digital photo frame mode to the "cervical stretching guide" animation, and displays the text prompt "cervical pressure detected, suggest following the animation to stretch for 30 seconds"; if visual fatigue (sight distance is too close or ambient light is abnormal), the RGB light strip turns orange and the display screen displays "suggest adjusting the screen brightness to 400 lux and maintaining a sight distance of 60 cm"; if the cognitive pressure is too high (HRV is low), the display screen plays a 90-second breathing training animation with a rhythm prompt sound;

[0053] S5: Closed-loop management: after the intervention is over, the system returns to step S1 to continue monitoring, and if the user's health indicators return to normal, the interaction module returns to normal.

[0054] All raw sensor data in the application are processed locally on the device, and are not uploaded to the cloud or third-party server, only the health indicators and interaction instructions are transmitted within the system, ensuring privacy and security.

[0055] The above describes the present application and its embodiments, which are not limited, and the drawings only show one of the embodiments of the present application, and the actual structure is not limited thereto. In general, if a person skilled in the art is inspired thereby, without departing from the purpose of the present application, without creative design, similar structure and embodiments of the technical solution are not creative, and should belong to the protection scope of the present application.

Claims

1. A desktop health monitoring system based on non-visual perception, characterized in that: It includes a data acquisition module, a data processing and decision-making module, and an interaction and feedback module; the data acquisition module is used to collect user status data without being noticed, the data processing and decision-making module is used to process the collected data and assess the user's health status, and the interaction and feedback module is used to provide feedback and intervention to the user based on the health status assessment results.

2. The desktop health monitoring system based on non-visual perception according to claim 1, characterized in that: The data acquisition module includes a millimeter-wave radar sensor, a time-of-flight sensor, and an ambient light sensor. The millimeter-wave radar sensor is used to transmit frequency-modulated continuous waves and receive reflected signals to generate point cloud data containing the position information of the user's head, neck, and shoulders, and to extract the chest cavity micro-motion signals caused by heartbeat and breathing through phase demodulation. The time-of-flight sensor is used to emit near-infrared light towards the user and measure the round-trip time to obtain the viewing distance between the user's eyes and the device screen. The ambient light sensor is used to monitor the ambient light intensity and color temperature.

3. The desktop health monitoring system based on non-visual perception according to claim 2, characterized in that: The data processing and decision-making module employs an embedded processor, which incorporates a sensor fusion algorithm, a physiological parameter calculation unit, and a health status assessment model. The sensor fusion algorithm is used to fuse point cloud data from a millimeter-wave radar sensor and line-of-sight data from a time-of-flight sensor. The physiological parameter calculation unit is used to extract physiologically relevant data from the user's status data. The health status assessment model is used to determine the user's health status.

4. The desktop health monitoring system based on non-visual perception according to claim 3, characterized in that: The sensor fusion algorithm reconstructs a real-time skeletal key point model of the user's upper body using the fused data, and calculates posture indices such as head and neck tilt angle, cervical spine load torque, and sitting posture symmetry. The physiological parameter calculation unit extracts heart rate and heart rate variability data from the reflected signals of the millimeter-wave radar sensor.

5. A desktop health monitoring system based on non-visual perception according to claim 4, characterized in that: The health status assessment model comprehensively judges the user's health status based on posture indicators, visual distance data, heart rate and heart rate variability data, and ambient light intensity and color temperature data. The health status includes normal, mild fatigue, severe fatigue, and poor posture.

6. A desktop health monitoring system based on non-visual perception according to claim 5, characterized in that: The interaction and feedback module includes a display screen; the display screen is used as a digital photo frame to slide user-defined images, displaying health status data in real time in a generative art form, and displaying guiding intervention content when the health status assessment model determines that intervention is needed.

7. A desktop health monitoring system based on non-visual perception according to claim 6, characterized in that: The interaction and feedback module also includes optional electroluminescent elements or RGB light strips; the electroluminescent elements or RGB light strips provide non-intrusive status cues through color changes, with blue representing focus and orange representing fatigue.

8. A health monitoring method based on the system according to any one of claims 1-7, characterized in that, Includes the following steps: S1: Continuously and seamlessly collects non-visual perception data from users through the data acquisition module; S2: On the embedded processor side of the data processing and decision-making module, the sensor fusion algorithm and health status assessment model are run to convert the collected data into medically significant health indicators in real time. S3: Determine whether the user's current health status exceeds a preset threshold based on health indicators; S4: If the preset threshold is exceeded, a personalized, guided digital intervention will be initiated through the interaction and feedback module; S5: Continuously repeat steps S1-S4 to form a closed-loop management of "monitoring-evaluation-intervention".

Images