A method for evaluating the ergonomics of a mattress

Abstract

The application discloses a kind of man-machine adaptability evaluation methods of mattress, it is related to the field of ergonomics health monitoring, including, according to BMI grading standard screening tester and collecting tester's height, weight, gender, age, sleep habit and chronic pain history basic data, obtain the BMI value of tester;Tester BMI value is input into mattress hardness selection model, obtain the test mattress hardness range, in constant temperature and humidity experimental environment, arrange selected hardness bed mattress to be measured, tester is in standard supine position and standard lateral decubitus in turn lie on bed mattress to be measured.The application predicts spine stiffness demand curve in real time through LSTM neural network, realizes millisecond level dynamic support adjustment, response speed is significantly improved compared with traditional method, can prevent spinal compensatory curvature in sleep;Combined with the IRI index calculation of pressure-blood oxygen coupling, accurately warn microcirculation disorder before tissue damage, and solve local ischemia problem through zoning adjustment.

Description

Technical Field

[0001] This invention relates to the field of ergonomic health monitoring, and in particular to a method for assessing the human-machine fit of a mattress. Background Technology

[0002] Current mattress fit assessment technology is mainly based on a combination of static pressure distribution analysis and subjective questionnaire surveys. It uses a high-density pressure sensor array to collect body pressure distribution data of users in supine and side-lying positions, and combines this with sleep quality questionnaires filled out by users to generate firmness adjustment suggestions. Its core technology lies in establishing a mapping relationship between pressure peaks and comfort scores. When the pressure on the shoulders or hips exceeds 32 mmHg, it automatically triggers airbag pressure adjustment. Based on the introduction of temperature and humidity sensors, it optimizes the selection of material breathability by monitoring the temperature fluctuations (±2℃) of the mattress surface microenvironment.

[0003] Although existing technologies can achieve quantitative assessment of pressure distribution, they still have significant limitations in terms of spinal biomechanical adaptation. The most prominent problem is the lack of a real-time monitoring and feedback mechanism for dynamic changes in the physiological curvature of the spine. Existing methods usually use preset stiffness classification standards, but do not consider the real-time changes in spinal curvature caused by changes in body position during sleep. In traditional assessments, the detection rate of lumbar curvature deviations exceeding 15° is less than 40%, and the response delay is more than 30 seconds, resulting in a mismatch between mattress support stiffness and the actual needs of the spine. Existing technologies treat pressure distribution and blood oxygen monitoring as independent parameters and fail to establish a multimodal coupling model of pressure-blood oxygen-spinal curvature, which limits the accuracy of local ischemia risk warning. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for assessing the human-machine adaptability of a mattress to solve the problems of delayed spinal biomechanical response and inaccurate physiological risk warning in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for assessing the human-machine adaptability of a mattress, which includes screening test subjects according to a BMI grading standard and collecting basic data of the test subjects’ height, weight, gender, age, sleep habits and chronic pain history to obtain the test subjects’ BMI value;

[0008] Input the BMI value of the test subject into the mattress firmness selection model to obtain the mattress firmness range. Set up the mattress with the selected firmness in a constant temperature and humidity test environment. The test subject lies on the mattress in the standard supine position and the standard lateral position in turn. The IMU sensor is used to collect the measured lumbar curvature data of the test subject, and the pressure distribution data on the mattress surface is recorded by the pressure sensor array.

[0009] The measured lumbar curvature data is input into the LSTM neural network model to predict the lumbar curvature sequence. The deviation between the lumbar curvature sequence and the measured lumbar curvature data is compared and the curvature deviation data is recorded to obtain mattress support failure data. Overnight sleep monitoring is performed to obtain pressure distribution data, body surface contact surface temperature field data, body movement event data, and heart rate variability data.

[0010] By embedding mattress surface pressure distribution data into a flexible fiber optic SpO2 sensor, blood oxygen monitoring is activated to obtain changes in blood oxygen saturation in the shoulders, waist, and hips.

[0011] The ischemic risk index (IRI) for the shoulder, waist, and hip is calculated based on changes in blood oxygen saturation in these areas.

[0012] After monitoring sleep throughout the night, a visual analog scale (VAS) score was obtained. Based on body movement event data and heart rate variability data, sleep efficiency data was obtained. The mattress support failure data, risk index IRI, sleep efficiency data and VAS score were combined to generate a fit score.

[0013] As a preferred embodiment of the ergonomics assessment method for the mattress described in this invention, the method includes: screening test subjects according to BMI grading standards and collecting basic data on the test subjects' height, weight, gender, age, sleep habits, and history of chronic pain to obtain the test subjects' BMI values, comprising the following steps.

[0014] Based on the BMI classification standard, individuals with low weight, normal weight, overweight, and obesity were screened. Fasting weight upon waking was measured using a calibrated body composition analyzer, and height was measured using an ultrasonic height meter. Basic data on gender, age, demographics, sleep habits, and history of chronic pain were collected through a structured electronic questionnaire.

[0015] The BMI value is calculated based on the height and weight of the test subject.

[0016] As a preferred embodiment of the ergonomics assessment method for the mattress described in this invention, the method includes: inputting the test subject's BMI value into a mattress firmness selection model to obtain the mattress firmness range; setting up the selected firmness mattress in a constant temperature and humidity experimental environment; the test subject lying on the mattress in a standard supine position and a standard lateral position; collecting measured lumbar curvature data using an IMU sensor; and recording mattress surface pressure distribution data using a pressure sensor array. The method includes the following steps:

[0017] The BMI value is input into the mattress firmness selection model and a piecewise linear function is used to obtain a firmness value with posture marking. The parameters of the climate simulation chamber are automatically adjusted, the mattress under test is deployed, and the body posture guidance line is displayed on the ground through a projector to obtain the body posture standard signal.

[0018] A miniature IMU sensor is installed at the L3 vertebral body to obtain the measured lumbar curvature data of the test subject. Based on the measured lumbar curvature data of the test subject, a pressure sensor array scan is triggered to obtain a pressure distribution matrix. The pressure distribution matrix is ​​then encapsulated and spatially mapped and transformed to convert the pressure distribution matrix into pressure distribution data on the mattress surface.

[0019] As a preferred embodiment of the mattress ergonomics assessment method described in this invention, the method includes: inputting measured lumbar curvature data into an LSTM neural network model to predict a lumbar curvature sequence; comparing the deviation between the lumbar curvature sequence and the measured lumbar curvature data; recording the curvature deviation data; obtaining mattress support failure data; and performing overnight sleep monitoring to obtain pressure distribution data, body surface contact temperature field data, body movement event data, and heart rate variability data. This includes the following steps:

[0020] Z-score standardization was performed on the measured lumbar curvature data to obtain a normalized curvature sequence.

[0021] The normalized curvature sequence is input into the LSTM neural network model to predict the lumbar curvature sequence.

[0022] The ideal curvature is obtained based on the test subject's body mass index;

[0023] The lumbar curvature requirement coefficient is obtained based on the predicted lumbar curvature sequence and the ideal curvature.

[0024] By comparing the absolute difference between the predicted lumbar curvature sequence and the measured lumbar curvature data, curvature deviation data is obtained.

[0025] When the curvature deviation data exceeds 15 degrees for 1.5 seconds, mattress support failure data is obtained;

[0026] Pressure scanning, infrared thermal imaging, and body motion monitoring were performed based on mattress support failure data to obtain pressure distribution data, body surface contact temperature field data, body motion event data, and heart rate variability data.

[0027] As a preferred embodiment of the ergonomics assessment method for the mattress described in this invention, the method includes: embedding mattress surface pressure distribution data into a flexible fiber optic SpO2 sensor, activating blood oxygen monitoring, and obtaining changes in blood oxygen saturation in the shoulders, waist, and hips, comprising the following steps.

[0028] The pressure distribution hotspots were extracted from the pressure distribution data on the mattress surface. Gaussian filtering and connected component analysis were performed on the pressure distribution hotspots to obtain a list of the center coordinates of the hotspot areas.

[0029] By alternating red and infrared light modulation, the original photoelectric signal is obtained. Frequency domain analysis is performed on the original photoelectric signal to calculate the ratio of pulsating components, thereby obtaining the changes in blood oxygen saturation in the shoulder, waist, and buttocks.

[0030] As a preferred embodiment of the ergonomics assessment method for the mattress described in this invention, the ischemia risk index (IRI) for the shoulder, waist, and hip is calculated based on changes in blood oxygen saturation in the shoulder, waist, and hip, including the following steps:

[0031] Zero-phase filtering is applied to blood oxygen saturation to obtain the filtered blood oxygen signal. The moving average of the initial filtered blood oxygen signal is taken to obtain the blood oxygen saturation baseline. The dynamic time warping algorithm is used to align the pressure and blood oxygen signals, and the ischemic risk index (IRI) of the shoulder, waist and hip is calculated.

[0032] As a preferred embodiment of the human-machine adaptability assessment method for the mattress described in this invention, the method includes the following steps: obtaining a visual analog scale score after overnight sleep monitoring, and obtaining sleep efficiency data based on body movement event data and heart rate variability data.

[0033] After monitoring sleep throughout the night, the pain level and comfort scores at night were marked on the visual analog scale to obtain the visual analog scale score.

[0034] Sleep efficiency data is obtained by calculating the body movement event data and heart rate variability data.

[0035] As a preferred embodiment of the mattress ergonomics assessment method of the present invention, the method involves combining mattress support failure data, risk index IRI, sleep efficiency data, and visual analog scale scores to generate an ergonomics score, including the following steps:

[0036] Mattress support failure data, risk index IRI, sleep efficiency data, and visual analog scale scores were normalized to obtain dimensionless standardized values, and a weighted method was used to generate a fit score.

[0037] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the mattress human-machine fit assessment method as described in the first aspect of the present invention.

[0038] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the mattress human-machine fit assessment method as described in the first aspect of the present invention.

[0039] The beneficial effects of this invention are as follows: by predicting the spinal stiffness demand curve in real time through LSTM neural network, millisecond-level dynamic support adjustment is achieved, which significantly improves the response speed compared with traditional methods and can prevent compensatory curvature of the spine during sleep; combined with the calculation of IRI index of pressure-blood oxygen coupling, it can accurately warn of microcirculation disorders before tissue damage and solve local ischemia problems through zonal adjustment. Attached Figure Description

[0040] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 A flowchart for the method of assessing the human-machine fit of a mattress.

[0042] Figure 2 This is a diagram illustrating mattress support failure data.

[0043] Figure 3 A schematic diagram of pressure distribution recorded by a pressure sensing array.

[0044] Figure 4 This is a flowchart for calculating BMI. Detailed Implementation

[0045] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0046] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0047] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0048] ReferenceFigures 1-4 As one embodiment of the present invention, this embodiment provides a method for evaluating the ergonomics of a mattress, comprising the following steps:

[0049] S1. Select test subjects according to the BMI classification standard and collect basic data such as height, weight, gender, age, sleep habits and chronic pain history of the test subjects to obtain the test subjects' BMI value.

[0050] S1.1. Based on the BMI classification standard, individuals with low weight, normal weight, overweight, and obesity were screened. Fasting weight upon waking was measured using a calibrated body composition analyzer, and height was measured using an ultrasonic height meter. Basic data on gender, age, demographic data, sleep habits, and history of chronic pain were collected through a structured electronic questionnaire.

[0051] Furthermore, based on the BMI classification standard, four groups were selected: underweight (BMI<18.5), normal weight (18.5≤BMI<24), overweight (24≤BMI<28), and obese (BMI≥28). Fasting weight upon waking was measured using a metrology-certified RGZ-160 electronic body composition analyzer. Before measurement, participants were required to remove shoes and heavy clothing. The recording accuracy was 0.1 kg. Height was measured using an HC-1000 ultrasonic height meter. Participants maintained an upright posture with their heels, sacrum, and scapula in contact with the measuring rod. The recording accuracy was 0.1 cm. Demographic data, including gender, age, occupation, and education level, were collected using a GCP-compliant electronic questionnaire. Sleep habits were assessed using the PSQI scale, and the location and severity of chronic pain were recorded using the VAS score.

[0052] S1.2 Calculate the BMI value based on the height and weight of the test subject.

[0053] ;

[0054] in, BMI Body Mass Index (BMI) G For the height of the test subject, H The test subject's weight.

[0055] S2. Input the test subject's BMI value into the mattress firmness selection model to obtain the mattress firmness range. Set up the selected firmness mattress in a constant temperature and humidity test environment. The test subject lies on the mattress in the standard supine position and the standard lateral position in turn. The IMU sensor is used to collect the test subject's measured lumbar curvature data, and the pressure distribution data on the mattress surface is recorded through the pressure sensor array.

[0056] S2.1 Input the BMI value into the mattress firmness selection model and use a piecewise linear function to obtain a firmness value with posture marking. Automatically adjust the parameters of the climate simulation chamber, deploy the mattress to be tested, and display the body position guidance line on the ground through a projector to obtain the body position compliance signal.

[0057] Furthermore, the BMI value is input into a mattress firmness selection model based on the T / CESS21-2024 standard. This model uses a piecewise linear function to calculate and output a firmness value with posture markers (e.g., supine HM3.5 / sideways MM5.0). Based on the output firmness value, the climate simulation chamber parameters (temperature 23±0.5℃, humidity 50±3%RH) are automatically adjusted. A robotic arm selects and positions the mattress with the corresponding firmness (horizontal error <0.5°). A DLP projector projects standard body position guide lines (supine head position) onto the mattress surface. With the mattress positioned 15±0.3cm from the head of the bed and legs flexed at 130°±2° in a side-lying position, a position compliance signal is generated when the pressure sensor array detects that the body pressure distribution meets the preset contact area threshold (>95%). The spatial coordinates of the position guide line are aligned with the detection area of ​​the pressure sensor array at the millimeter level (error <1mm) through a pre-calibrated transformation matrix, ensuring that the guide position is consistent with the actual pressure monitoring area. During mattress deployment, environmental parameter fluctuations are monitored in real time (e.g., PID control compensation is triggered when the temperature change is >0.3℃) until the test environment stability standard is reached.

[0058] S2.2. Install a miniature IMU sensor at the L3 vertebral body to obtain the measured lumbar curvature data of the test subject. Based on the measured lumbar curvature data of the test subject, trigger the pressure sensor array scan to obtain the pressure distribution matrix. Then, encapsulate and spatially map the pressure distribution matrix to convert it into pressure distribution data on the mattress surface.

[0059] Specifically, the expression is,

[0060] ;

[0061] in, The lumbar curvature data was measured for the test subjects. The vertical coordinate of the first lumbar vertebra. The vertical coordinate of the fifth lumbar vertebra. for to The vertical coordinates of the vertebral body. The horizontal coordinate of the fifth lumbar vertebra. The horizontal coordinate of the first lumbar vertebra. for to The coordinates of the vertebral body in the horizontal direction.

[0062] It should be noted that an Xsens DOT miniature IMU sensor is attached to the center point of the spinous process of the L3 lumbar vertebra. The IMU sensor axis is parallel to the sagittal plane of the spine. The real-time lumbar curvature data is transmitted via Bluetooth 5.2. When the IMU sensor detects a stable body position signal (angle fluctuation <1° / s for 5 seconds), it triggers the Tekscan 5315 pressure sensor array to start a high-density scan (resolution 1.27cm / point) to obtain the original pressure matrix. Three-point calibration compensation (zero point / linear / temperature compensation) is performed on the original pressure matrix to generate a calibrated pressure matrix. Then, missing points are filled by bicubic spline interpolation to form a complete 1024-point pressure distribution matrix. The pressure distribution matrix is ​​converted into mattress surface pressure distribution data conforming to the ISO 2439 standard format using preset spatial mapping transformation parameters (calibration error <1mm). The data includes the absolute pressure value (mmHg) and local coordinates (relative to the center point of the mattress) of each measurement point. The pressure distribution data is encapsulated in JSON format and includes metadata fields such as acquisition timestamp, sensor ID, and ambient temperature and humidity.

[0063] S3. Input the measured lumbar curvature data into the LSTM neural network model to predict the lumbar curvature sequence, compare the deviation between the lumbar curvature sequence and the measured lumbar curvature data, record the curvature deviation data, obtain the mattress support failure data, and perform overnight sleep monitoring to obtain pressure distribution data, body surface contact surface temperature field data, body movement event data, and heart rate variability data.

[0064] S3.1. The measured lumbar curvature data are Z-score standardized to obtain a normalized curvature sequence.

[0065] Furthermore, the measured lumbar curvature data acquired by the Xsens DOT miniature IMU sensor were subjected to Z-score normalization. The normalized sequence was then filtered by a fourth-order Butterworth low-pass filter to eliminate high-frequency noise, resulting in a normalized curvature sequence.

[0066] S3.2 Input the normalized curvature sequence into the LSTM neural network model to predict the lumbar curvature sequence.

[0067] Specifically, the expression is,

[0068] ;

[0069] in, To predict lumbar curvature sequences, It is a five-dimensional feature set.

[0070] It should be noted that the lumbar curvature sequence, which has been normalized by Z-score, is input into the pre-trained LSTM neural network model (128 hidden nodes, dropout rate 0.2). The input of the LSTM neural network model is a five-dimensional feature set containing the current curvature, the rate of change of curvature (obtained by first-order difference of measured lumbar curvature data collected by IMU sensor), BMI value, sleeping posture code (supine = 0 / lateral = 1) and the current mattress firmness, to predict the lumbar curvature sequence.

[0071] The LSTM neural network model is trained using the backpropagation algorithm. The loss function is the mean square error between the predicted curvature and the actual curvature. The training data comes from time-series data of 500 spinal biomechanical experiments. The prediction results are restored to the actual angle values ​​through denormalization.

[0072] S3.3. Obtain the ideal curvature based on the tester's body mass index.

[0073] Specifically, the expression is,

[0074] ;

[0075] in, For ideal curvature, This is the body position correction factor. This is the age correction factor.

[0076] S3.4. Based on the predicted lumbar curvature sequence and the ideal curvature, the lumbar curvature requirement coefficient is obtained.

[0077] Specifically, the expression is,

[0078] ,

[0079] in, This is the stiffness requirement coefficient for the lumbar spine. For the maximum permissible deviation, For the first i Predicted lumbar curvature values ​​at various time points i Indexed by time point.

[0080] S3.5. Compare the absolute difference between the predicted lumbar curvature sequence and the measured lumbar curvature data to obtain curvature deviation data.

[0081] Specifically, the expression is,

[0082] ;

[0083] in, This is the curvature deviation data.

[0084] S3.6 When the curvature deviation data exceeds 15 degrees for 1.5 seconds, the mattress support failure data is obtained.

[0085] Furthermore, the predicted lumbar curvature is compared with the measured lumbar curvature in real time. When the absolute deviation exceeds the 15-degree threshold for more than 1.5 seconds, it is marked as a mattress support failure event. The event record includes the start timestamp, duration, maximum deviation value and body position (supine / side-lying) at the time of occurrence.

[0086] S3.7. Based on mattress support failure data, perform pressure scanning, infrared thermal imaging, and body motion monitoring to obtain pressure distribution data, body surface contact surface temperature field data, body motion event data, and heart rate variability data.

[0087] Furthermore, when a mattress support failure event is triggered, the Tekscan 5315 pressure sensor array switches to a 100Hz high-frequency scanning mode to acquire a 1024-point pressure distribution matrix. Simultaneously, the FLIR A655sc infrared thermal imager acquires the temperature field of the body surface at a sampling rate of 5fps, the Delsys Trigno accelerometer records body movement event data at a frequency of 200Hz, and the Empatica E4 wristband monitors heart rate variability at a sampling rate of 64Hz. The pressure distribution matrix is ​​converted into standard pressure values ​​after three-point calibration compensation, the temperature field data is calibrated by blackbody radiation, and the body movement event data is filtered by a 0.1-10Hz bandpass filter to extract SVM features. Time synchronization is achieved through the IEEE 1588v2 protocol, and the data is packaged into a time-series dataset according to the ISO 19833 standard format.

[0088] S4. Embed the mattress surface pressure distribution data into a flexible fiber optic SpO2 sensor to activate blood oxygen monitoring and obtain changes in blood oxygen saturation in the shoulders, waist, and hips.

[0089] S4.1 Extract the pressure distribution hotspots from the mattress surface pressure distribution data, perform Gaussian filtering and connected component analysis on the pressure distribution hotspots, and obtain a list of the center coordinates of the hotspot areas.

[0090] Furthermore, for the mattress surface pressure distribution data collected by the Tekscan 5315 pressure sensor array, the coordinates of all three consecutive frames with pressure values ​​exceeding 32 mmHg were marked. Two-dimensional convolution filtering with a Gaussian kernel was used to eliminate isolated noise points. Then, closed regions were identified through 8-neighborhood connected component analysis. The coordinate transformation was performed using a preset affine transformation matrix to map the sensor array coordinates to the mattress physical coordinate system. For each hot spot region, three features were recorded: maximum pressure value, area, and pressure gradient, resulting in a list of center coordinates of the hot spot regions.

[0091] S4.2. By alternating modulation of red and infrared light, the original photoelectric signal is obtained. Frequency domain analysis is performed on the original photoelectric signal to calculate the ratio of pulsating components, thereby obtaining the changes in blood oxygen saturation in the shoulder, waist, and buttocks.

[0092] Specifically, the expression is,

[0093] ,

[0094] in, Blood oxygen saturation For red light exchange components, The red light DC component, For infrared light AC component, This is the DC component of infrared light.

[0095] S5. Calculate the ischemic risk index (IRI) for the shoulder, waist, and hip based on changes in blood oxygen saturation in these areas.

[0096] S5.1 Perform zero-phase filtering on blood oxygen saturation to obtain the filtered blood oxygen signal. Take the moving average of the initial filtered blood oxygen signal to obtain the blood oxygen saturation baseline. Use dynamic time warping algorithm to align pressure and blood oxygen signals, and calculate the ischemic risk index (IRI) for the shoulder, waist, and hip.

[0097] Specifically, the expression is,

[0098] ;

[0099] in, This is an ischemia risk index. T To assess the time window, P This is the real-time pressure value. For the safety pressure threshold, This refers to the blood oxygen level. This refers to the amount of decrease in blood oxygen saturation. As pressure weight, The weighting of blood oxygen saturation.

[0100] It should be noted that the blood oxygen saturation signal collected by the blood oxygen sensor is zero-phase filtered using a bidirectional IIR Butterworth filter to obtain the blood oxygen saturation baseline. The moving average of the blood oxygen saturation baseline over the previous 10 minutes is then used to calculate the ischemic risk index (IRI) for the shoulder, waist, and hip.

[0101] S6. After overnight sleep monitoring, a visual analog scale score is obtained, and sleep efficiency data is obtained based on body movement event data and heart rate variability data.

[0102] S6.1 After overnight sleep monitoring, mark the nighttime pain level and comfort score on the visual analog scale to obtain the visual analog scale score.

[0103] Furthermore, after completing the overnight sleep monitoring, the test subjects used a calibrated electronic stylus to mark the level of nighttime pain (0=no pain, 10=severe pain) and comfort rating (0=extreme discomfort, 10=extreme comfort) on a 10 cm visual analog scale. The scale's built-in laser locator ensured that the marking position was accurate to ±0.5 mm. After marking, a high-definition camera (0.1 mm resolution) automatically captured the position of the marked point on the scale, and the physical distance was converted into a digital score of 0-10 through image processing algorithms to generate a comprehensive visual analog scale score.

[0104] S6.2 Calculate sleep efficiency data by analyzing body movement event data and heart rate variability data.

[0105] Specifically, the expression is,

[0106] ;

[0107] Among them, SE represents sleep efficiency data. TIB Total bed rest time, TST Total sleep time.

[0108] S7. Combine mattress support failure data, risk index IRI, sleep efficiency data, and visual analog scale scores to generate a fit score.

[0109] S7.1, mattress support failure data, risk index IRI, sleep efficiency data and visual analog scale scores were normalized to obtain dimensionless standardized values, and a weighted method was used to generate a fit score.

[0110] Specifically, the expression is,

[0111] ,

[0112] in, Y For fit rating, To standardize support failure frequency, To standardize sleep efficiency, To standardize the ischemia risk index. For standardized visual analog scale scoring.

[0113] This embodiment also provides a computer device applicable to the mattress ergonomics assessment method, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the mattress ergonomics assessment method proposed in the above embodiment.

[0114] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0115] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the mattress human-machine adaptability assessment method as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0116] In summary, this invention achieves millisecond-level dynamic support adjustment by using an LSTM neural network to predict the spinal stiffness demand curve in real time, which significantly improves the response speed compared to traditional methods and can prevent compensatory curvature of the spine during sleep. Combined with the calculation of the pressure-blood oxygen coupling IRI index, it can accurately warn of microcirculatory disorders before tissue damage and solve local ischemia problems through zonal adjustment.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for evaluating the ergonomics of a mattress, characterized in that: include, Test subjects were selected according to the BMI classification standard, and their basic data such as height, weight, gender, age, sleep habits and chronic pain history were collected to obtain their BMI value. Input the BMI value of the test subject into the mattress firmness selection model to obtain the mattress firmness range. Set up the mattress with the selected firmness in a constant temperature and humidity test environment. The test subject lies on the mattress in the standard supine position and the standard lateral position in turn. The IMU sensor is used to collect the measured lumbar curvature data of the test subject, and the pressure distribution data on the mattress surface is recorded by the pressure sensor array. The measured lumbar curvature data, curvature change rate, test subject BMI value, sleeping posture code, and current mattress firmness are combined into a five-dimensional feature group. After Z-score standardization, the data is input into an LSTM neural network model to predict the lumbar curvature sequence. The deviation between the predicted lumbar curvature sequence and the measured lumbar curvature data is compared and the curvature deviation data is recorded to obtain mattress support failure data. Overnight sleep monitoring is also performed to obtain event synchronous pressure distribution data, body surface contact surface temperature field data, body movement event data, and heart rate variability data. By embedding mattress surface pressure distribution data into a flexible fiber optic SpO2 sensor, blood oxygen monitoring is activated to obtain changes in blood oxygen saturation in the shoulders, waist, and hips. The ischemic risk index (IRI) for the shoulder, waist, and hip is calculated based on changes in blood oxygen saturation in these areas. After monitoring sleep throughout the night, a visual analog scale (VAS) score was obtained. Based on body movement event data and heart rate variability data, sleep efficiency data was obtained. The mattress support failure data, risk index IRI, sleep efficiency data and VAS score were combined to generate a fit score.

2. The method for assessing the ergonomics of a mattress as described in claim 1, characterized in that: Participants were screened according to the BMI classification standard, and their basic data, including height, weight, gender, age, sleep habits, and history of chronic pain, were collected to obtain their BMI values. This process included the following steps: Based on the BMI classification standard, individuals with low weight, normal weight, overweight, and obesity were screened. Fasting weight upon waking was measured using a calibrated body composition analyzer, and height was measured using an ultrasonic height meter. Basic data on gender, age, demographics, sleep habits, and history of chronic pain were collected through a structured electronic questionnaire. The BMI value is calculated based on the height and weight of the test subject.

3. The method for evaluating the ergonomics of a mattress as described in claim 2, characterized in that: The test subject's BMI value is input into the mattress firmness selection model to obtain the mattress firmness range. The selected firmness mattress is then placed in a constant temperature and humidity experimental environment. The test subject lies on the mattress in a standard supine position and a standard side-lying position, respectively. IMU sensors are used to collect measured lumbar curvature data, and a pressure sensor array is used to record the pressure distribution data on the mattress surface. This process includes the following steps: The BMI value is input into the mattress firmness selection model and a piecewise linear function is used to obtain a firmness value with posture marking. The parameters of the climate simulation chamber are automatically adjusted, the mattress under test is deployed, and the body posture guidance line is displayed on the ground through a projector to obtain the body posture compliance signal. A miniature IMU sensor is installed at the L3 vertebral body to obtain the measured lumbar curvature data of the test subject. Based on the measured lumbar curvature data of the test subject, a pressure sensor array scan is triggered to obtain a pressure distribution matrix. The pressure distribution matrix is ​​then encapsulated and spatially mapped and transformed to convert the pressure distribution matrix into pressure distribution data on the mattress surface.

4. The method for evaluating the ergonomics of a mattress as described in claim 3, characterized in that: The measured lumbar curvature data, curvature change rate, test subject's BMI value, sleeping posture code, and current mattress firmness were combined into a five-dimensional feature set. After Z-score standardization, the data was input into an LSTM neural network model to predict the lumbar curvature sequence. The deviation between the predicted lumbar curvature sequence and the measured lumbar curvature data was compared and recorded to obtain mattress support failure data. Overnight sleep monitoring was also performed to obtain event-synchronized pressure distribution data, body surface contact temperature field data, body movement event data, and heart rate variability data. This process includes the following steps. Based on the Z-score-standardized measured lumbar curvature data, curvature change rate, test subject BMI value, sleeping posture code, and current mattress firmness, a five-dimensional feature group was formed to obtain a normalized feature sequence; The normalized curvature sequence is input into the LSTM neural network model to predict the lumbar curvature sequence. The ideal curvature is obtained based on the test subject's body mass index; The lumbar curvature requirement coefficient is obtained based on the predicted lumbar curvature sequence and the ideal curvature. By comparing the absolute difference between the predicted lumbar curvature sequence and the measured lumbar curvature data, curvature deviation data is obtained. When the curvature deviation data exceeds 15 degrees for 1.5 seconds, mattress support failure data is obtained; Pressure scanning, infrared thermal imaging, and body motion monitoring were performed based on mattress support failure data to obtain event-synchronized pressure distribution data, body surface contact temperature field data, body motion event data, and heart rate variability data.

5. The method for evaluating the ergonomics of a mattress as described in claim 4, characterized in that: The pressure distribution data on the mattress surface is embedded into a flexible fiber optic SpO2 sensor to activate blood oxygen monitoring, obtaining changes in blood oxygen saturation in the shoulders, lower back, and hips. This includes the following steps: The pressure distribution hotspots were extracted from the pressure distribution data on the mattress surface. Gaussian filtering and connected component analysis were performed on the pressure distribution hotspots to obtain a list of the center coordinates of the hotspot areas. By alternating red and infrared light modulation, the original photoelectric signal is obtained. Frequency domain analysis is performed on the original photoelectric signal to calculate the ratio of pulsating components, thereby obtaining the changes in blood oxygen saturation in the shoulder, waist, and buttocks.

6. The method for evaluating the ergonomics of a mattress as described in claim 5, characterized in that: The ischemia risk index (IRI) for the shoulder, lower back, and hip was calculated based on changes in blood oxygen saturation in these areas. Includes the following steps, Zero-phase filtering is applied to blood oxygen saturation to obtain the filtered blood oxygen signal. The moving average of the initial filtered blood oxygen signal is taken to obtain the blood oxygen saturation baseline. The dynamic time warping algorithm is used to align the pressure and blood oxygen signals, and the ischemic risk index (IRI) of the shoulder, waist and hip is calculated.

7. The method for evaluating the ergonomics of a mattress as described in claim 6, characterized in that: After overnight sleep monitoring, a visual analog scale score was obtained. Based on body movement event data and heart rate variability data, sleep efficiency data was obtained, including the following steps. After monitoring sleep throughout the night, the pain level and comfort scores at night were marked on the visual analog scale to obtain the visual analog scale score. Sleep efficiency data is obtained by calculating the body movement event data and heart rate variability data.

8. The method for evaluating the ergonomics of a mattress as described in claim 7, characterized in that: By combining mattress support failure data, the IRI (Intensity Risk Index), sleep efficiency data, and visual analog scale (VAS) scores, a fit score is generated. Includes the following steps, Mattress support failure data, risk index IRI, sleep efficiency data, and visual analog scale scores were normalized to obtain dimensionless standardized values, and a weighted method was used to generate a fit score.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the human-computer adaptation assessment method for the mattress according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the human-machine fit assessment method for the mattress according to any one of claims 1 to 8.

Images