Search

CN-121987950-A - Sleep-assisted deep physiotherapy cabin dynamic adjustment system and method based on artificial intelligence

CN121987950ACN 121987950 ACN121987950 ACN 121987950ACN-121987950-A

Abstract

The invention provides a sleep-assisted deep physiotherapy cabin dynamic adjustment system and method based on artificial intelligence. Belongs to the technical field of crossing of artificial intelligence and active environment regulation. The method comprises the steps of monitoring the sleep state of a user in real time through the multi-mode sensor array, obtaining multi-dimensional biomechanical parameters, constructing a user sleep biomechanical feature model based on the multi-dimensional biomechanical parameters, generating dynamic support demand assessment data, capturing biomechanical parameters such as user body pressure distribution, spine curvature and the like through the multi-mode sensor array in real time, dynamically generating a spine health-oriented suspension support force adjustment scheme through combining an artificial intelligent algorithm, effectively reducing the risk of support discomfort caused by static adjustment logic of a traditional deep physiotherapy cabin, avoiding abnormal accumulation of spine pressure caused by single physiological perception, and remarkably improving the spine health maintenance effect in sleep.

Inventors

  • HUANG ZHIQIANG
  • HUANG MINGJIE
  • HUANG MINGKAI

Assignees

  • 深圳市轻生活科技有限公司

Dates

Publication Date
20260508
Application Date
20260127

Claims (10)

  1. 1. The artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjustment method is characterized by comprising the following steps of: S1, monitoring the sleep state of a user in real time through a multi-mode sensor array, obtaining multi-dimensional biomechanical parameters, constructing a user sleep biomechanical feature model based on the multi-dimensional biomechanical parameters, and generating dynamic support demand evaluation data; S2, according to dynamic support demand evaluation data, combining a pre-established spinal health biomechanical index system, generating a suspension support force adjustment target value of the deep physiotherapy cabin cushion through an artificial intelligent algorithm, and simultaneously constructing a water dynamics physical mapping model, and converting the suspension support force target value into a flow field pressure distribution control instruction in the deep physiotherapy cabin cushion; S3, according to a flow field pressure distribution control instruction, implementing the regional pressure regulation of the depth physiotherapy cabin cushion by adopting an intelligent pump valve cooperative control system, and correcting the regulation parameters through a real-time feedback mechanism to generate a dynamic suspension supporting force field; S4, constructing a spine health intervention effect evaluation model based on the verified dynamic suspension support force field data and combining the user sleep stage recognition result; s5, carrying out weighted health benefit analysis according to the comprehensive evaluation report to generate a dynamic regulation efficiency index, triggering the self-adaptive learning mechanism to optimize the regulation strategy when the efficiency index is lower than a preset threshold, outputting backbone health daemon data, and synchronizing to the user terminal and the medical monitoring system.
  2. 2. The artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjustment method according to claim 1, wherein the step S1 comprises: s11, performing starting-up self-check and parameter calibration on the multi-mode sensor array, outputting a reference signal through a standard pressure module and a physiological signal simulator, and generating a sensor calibration coefficient set; s12, starting a sensor array based on a calibration coefficient set, and performing millisecond-level high-frequency monitoring on a sleeping state of a user to obtain body pressure distribution original data, spine curvature profile data, micro turn-over time sequence data, body movement amplitude data and heart rate respiratory wave original data, so as to form a multidimensional biomechanical original parameter matrix; S13, carrying out data preprocessing on the original parameter matrix, filtering out environmental noise interference by adopting a wavelet transformation algorithm, and supplementing missing data points by adopting an interpolation algorithm to generate a standardized biomechanical parameter set; s14, extracting feature vectors in the standardized parameter set, and constructing a sleep biomechanical feature model of the user; And S15, inputting the feature model into a gradient lifting decision tree algorithm for iterative operation, and outputting dynamic support demand evaluation data.
  3. 3. The artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjustment method according to claim 1, wherein the step S2 comprises: s21, calling basic health file data of a user, performing personalized adaptation on a pre-established spinal health biomechanics index system, and correcting an intervertebral disc pressure threshold value and a muscle tension optimization interval; S22, performing characteristic association on the dynamic support demand evaluation data and the adapted index system, inputting an LSTM neural network model for training operation, and generating a deep physiotherapy cabin cushion partition suspension supporting force adjustment target value matrix; s23, carrying out safety boundary verification based on a target value matrix, judging whether the supporting force value is between the compression limit of the intervertebral disc and the bearing range of the mattress, and generating a target value set passing verification; S24, constructing a water body dynamics physical mapping model based on the Pascal principle, and defining a conversion relation between water flow pressure and supporting force and a flow field diffusion coefficient; S25, inputting the checked target value set into a physical mapping model, converting the target value set into opening instructions of all subareas flow control valves in the deep physiotherapy cabin cushion and pressure output parameters of the intelligent water pump, and generating a flow field pressure distribution control instruction set.
  4. 4. The artificial intelligence-based sleep-assisted deep therapy capsule dynamic adjustment method according to claim 3, wherein S24 comprises: Extracting core physical parameters of the Pascal principle, obtaining actual transmission error values through laboratory calibration, and generating a model basic physical parameter set; The method comprises the steps of obtaining partition structure data of a depth physiotherapy cabin pad, constructing a partition flow field calculation frame by combining basic physical parameters, and defining space boundary conditions of flow field analysis; developing a pressure-supporting force association experiment based on a computing frame, acquiring corresponding actual supporting force data by applying gradient pressure on each partition, and fitting a function relationship between the two by adopting a least square method to generate a water flow pressure-supporting force conversion relationship table; Analyzing the influence of liquid viscosity and the diameters of the partition communication holes on the water flow diffusion speed, simulating flow field distribution under different pressures by using CFD simulation, calculating the flow field diffusion coefficients of each partition, and generating a flow field diffusion coefficient matrix; And integrating the partitioned flow field calculation frame, the pressure-supporting force conversion relation table and the flow field diffusion coefficient matrix, constructing a complete water dynamics physical mapping model, correcting model errors through 3 groups of verification experiments, and generating a final water dynamics physical mapping model based on the Pascal principle.
  5. 5. The artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjustment method according to claim 1, wherein the step S3 comprises: S31, initializing an intelligent pump valve cooperative control system, and setting a water pump start-stop threshold and a valve response delay parameter for ensuring that the system is in a state to be regulated; S32, driving a partition hydraulic control valve to adjust the opening according to a flow field pressure distribution control instruction set, and dynamically adjusting the water supply pressure through water pump pressure difference feedback to adjust the partition pressure of the deep physiotherapy cabin cushion; S33, synchronously starting the pressure sensor and the optical contour sensor, and collecting the regulated body pressure distribution data and the spine curvature data in real time to generate an adjustment effect real-time feedback matrix; s34, performing deviation calculation on the feedback matrix and the supporting force target value, correcting the pressure parameter of the water pump and the valve opening instruction by adopting a PID control algorithm, and optimizing the adjustment precision; And S35, repeating the adjusting and correcting processes until the deviation is smaller than 5%, generating dynamic suspension supporting force field data, and quantitatively evaluating the coincidence degree of the actual supporting effect and the target value.
  6. 6. The artificial intelligence based sleep assist depth physiotherapy cabin dynamic adjustment method according to claim 5, wherein S34 comprises: Extracting actual supporting force values of all the subareas from the real-time feedback matrix of the adjusting effect, extracting target supporting force of the corresponding subareas from the supporting force target value, aligning data according to the subarea numbers, and generating a deviation calculation basic data set; Calculating the supporting force deviation rate of each partition by adopting a relative deviation formula, and simultaneously, counting auxiliary deviation parameters of the water pump to generate a multidimensional deviation matrix; Initializing PID control parameters according to the characteristics of the deep physiotherapy cabin cushion system, setting a proportional coefficient Kp=2.5, an integral time Ti=10s and a differential time Td=2s, and generating a PID parameter configuration set; Inputting the multidimensional deviation matrix into a PID control algorithm, calculating an instant correction amount through a proportional term, eliminating static deviation through an integral term, inhibiting overshoot through a differential term, and comprehensively outputting a water pump pressure correction value and a valve opening correction amplitude to generate a preliminary correction instruction; And checking whether the preliminary correction instruction exceeds the safety range of the system, removing the overrun correction value, and recalculating to generate a final water pump pressure correction parameter and a valve opening correction instruction.
  7. 7. The artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjustment method according to claim 1, wherein S4 comprises: S41, acquiring sleep brain electrical signals of a user through a non-invasive brain wave sensor, and combining body movement characteristic data acquired by a multi-mode sensor array to form a sleep physiological characteristic fusion matrix; S42, performing classification operation on the fusion matrix by adopting a convolutional neural network, identifying different sleep stages, and generating a sleep stage time sequence distribution result; S43, carrying out space-time alignment on the verified dynamic suspension support force field data and the sleep stage time sequence distribution result, and extracting intervertebral disc stress data and muscle relaxation parameters of each sleep stage; S44, constructing a spinal health intervention effect evaluation model, and inputting aligned data to calculate an intervertebral disc pressure relief index, a muscle tension optimization coefficient and a sleep quality improvement degree; S45, integrating various evaluation indexes and sleep stage analysis results to generate a comprehensive evaluation report.
  8. 8. The artificial intelligence based sleep assist depth physiotherapy cabin dynamic adjustment method according to claim 7, wherein S43 comprises: Extracting the time stamp of the acquisition time stamp of the dynamic suspension support force field data after verification and the time slice of the time sequence distribution result of the sleep stage, uniformly converting the time slice into a UTC standard time format, and generating a double data set with consistent time reference; Based on a unified time reference, establishing an association mapping rule of the supporting force field data and the sleep stage, matching the supporting force field data to the corresponding sleep stage, and generating a space-time association data set; Invoking a spinal biomechanics labeling model, determining key monitoring segments of the intervertebral disc, extracting real-time pressure values of each segment in different sleep stages from a space-time associated data set, and generating an intervertebral disc stress time sequence data table; Combining the muscle tremor frequency data acquired by the multi-mode sensor with uniformity parameters of the supporting force field, calculating the muscle relaxation through a fuzzy comprehensive evaluation method, and generating a muscle relaxation parameter set of each sleep stage; and classifying according to sleep stages, integrating the intervertebral disc stress data table and the muscle relaxation parameters of the corresponding stages, and generating a special spinal physiological characteristic parameter set of each sleep stage.
  9. 9. The artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjustment method according to claim 1, wherein the step S5 comprises: S51, determining health benefit weight based on an analytic hierarchy process, wherein the intervertebral disc pressure relieving proportion is 40%, the muscle tension optimizing proportion is 30%, the sleep quality improving proportion is 30%, and performing weighting operation on the comprehensive evaluation report to generate a dynamic regulation efficiency index; s52, comparing the efficiency index with a preset threshold, and triggering an adaptive learning mechanism if the efficiency index is lower than the threshold, and calling historical sleep data of a user and adjusting and recording parameters of a neural network model; s53, generating backbone health daemon data based on the optimized model; S54, synchronizing the daemon data to the mobile terminal of the user through the encryption transmission protocol, visually inquiring the sleep data, and uploading the daemon data to the medical monitoring system according to the medical data interaction standard.
  10. 10. Sleep assistance degree of depth physiotherapy cabin dynamic adjustment system based on artificial intelligence includes: One or more processors; A memory for storing one or more programs, Wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1 to 9.

Description

Sleep-assisted deep physiotherapy cabin dynamic adjustment system and method based on artificial intelligence Technical Field The invention provides a sleep-assisted deep physiotherapy cabin dynamic adjustment system and method based on artificial intelligence, and belongs to the technical field of intersection of artificial intelligence and active environment regulation. Background The prior high-end depth physiotherapy cabin or pneumatic mattress with intelligent regulation function has obvious defects in technical realization. The adjusting logic is mostly based on a preset program or a simple timer, and cannot be dynamically adjusted according to the real-time sleep stage and the physical state of the user, so that unnecessary interference is often caused in sensitive time periods such as a deep sleep stage. In the aspect of physiological perception, only single parameters such as heart rate, respiration and the like are relied on, key biomechanical indexes such as body pressure distribution, spine curvature, micro turning frequency and the like are ignored, and the real supporting requirement of a user is difficult to accurately evaluate. On the control targets, lack of quantitative indicators directly related to spinal health, such as intervertebral disc pressure, muscle tension, etc., results in lack of medical basis for the regulation effect. In addition, the prior art does not establish a physical mapping model of the contact pressure between the flow field in the deep physiotherapy cabin cushion and the human body, the adjusting process is unpredictable, and the effect is unstable. More importantly, the smart deep physiotherapy pod products on the market are still "ambulatory beds" in nature, and do not achieve a paradigm shift from furniture to spinal health intervention equipment. Disclosure of Invention The invention provides a sleep-assisted deep physiotherapy cabin dynamic adjustment system and method based on artificial intelligence, which are used for solving the problems mentioned in the background art: the invention provides an artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjustment method, which comprises the following steps: S1, monitoring the sleep state of a user in real time through a multi-mode sensor array, obtaining multi-dimensional biomechanical parameters, constructing a user sleep biomechanical feature model based on the multi-dimensional biomechanical parameters, and generating dynamic support demand evaluation data; S2, according to dynamic support demand evaluation data, combining a pre-established spinal health biomechanical index system, generating a suspension support force adjustment target value of the deep physiotherapy cabin cushion through an artificial intelligent algorithm, and simultaneously constructing a water dynamics physical mapping model, and converting the suspension support force target value into a flow field pressure distribution control instruction in the deep physiotherapy cabin cushion; S3, according to a flow field pressure distribution control instruction, implementing the regional pressure regulation of the depth physiotherapy cabin cushion by adopting an intelligent pump valve cooperative control system, and correcting the regulation parameters through a real-time feedback mechanism to generate a dynamic suspension supporting force field; S4, constructing a spine health intervention effect evaluation model based on the verified dynamic suspension support force field data and combining the user sleep stage recognition result; s5, carrying out weighted health benefit analysis according to the comprehensive evaluation report to generate a dynamic regulation efficiency index, triggering the self-adaptive learning mechanism to optimize the regulation strategy when the efficiency index is lower than a preset threshold, outputting backbone health daemon data, and synchronizing to the user terminal and the medical monitoring system. The invention provides an artificial intelligence-based sleep-assisted deep physiotherapy cabin dynamic adjusting system, which comprises: One or more processors; A memory for storing one or more programs, Wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of the preceding claims. The intelligent physical therapy system has the advantages that biomechanical parameters such as body pressure distribution and spine curvature of a user are captured in real time through the multi-mode sensor array, the suspension supporting force adjusting scheme for guiding spine health is dynamically generated by combining an artificial intelligent algorithm, the supporting discomfort risk caused by static adjustment logic of a traditional deep physical therapy cabin is effectively reduced, abnormal accumulation of spine pressure caused by single physiological perception is avoided, the spine health maintenance effect in s