Control system of intelligent shoulder, neck and crotch thermal therapy equipment based on multi-modal feedback

Abstract

The invention discloses a control system of intelligent shoulder, neck and crotch thermal therapy equipment based on multi-modal feedback, and relates to the technical field of medical instruments, and the control system comprises a multi-modal sensing module, a central processing unit and an execution module; the adaptive thermal therapy scheme can be automatically generated according to the physiological data of the user and the thermal therapy rehabilitation treatment course, a doctor is assisted in performing thermal therapy operation, intelligent linkage and adaptive adjustment of the multiple thermal therapy modules for shoulders, necks, hips and the like are achieved through the cooperative control module and the comprehensive multi-mode feedback, and the thermal therapy efficiency is improved. Meanwhile, a multipoint fusion temperature monitoring technology is adopted, the advantages of a non-contact type sensor and a contact type sensor are combined, accurate control over a temperature field is achieved, local overheating or insufficient heating is effectively avoided, safety is high, a definite operation arbitration module is arranged, and the operation efficiency is improved. The problem of instruction conflict of multi-channel operation is solved, uniqueness and safety of system response are guaranteed, and man-machine interaction experience is optimized.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a control system for an intelligent shoulder, neck, and hip hyperthermia device based on multimodal feedback. Background Technology

[0002] In recent years, with the increasing awareness of health, intelligent thermotherapy devices for relieving muscle fatigue and soreness in areas such as the shoulders, neck, waist, and hips have been widely used. Existing thermotherapy devices typically integrate multiple functional modules such as heating, massage, and posture adjustment. Shortwave (frequency 3-30MHz) and ultrashortwave (frequency 30-300MHz) thermotherapy are key technologies for achieving deep tissue heating. Their core principle is that electromagnetic waves penetrate the tissue, causing internal ions and polar molecules to oscillate and rub at high speed, directly generating heat deep within the tissue. Traditional shoulder thermotherapy has limitations in depth, control precision, and regional zoning. Shortwave and ultrashortwave provide stronger deep heat and regional control capabilities. Combined with closed-loop temperature control, more uniform and safer individualized treatment can be achieved. However, existing hyperthermia equipment control systems still have many technical shortcomings in practical applications: 1. Imperfect multi-module coordination control: Existing control systems usually control each module independently or in simple linkage, lacking intelligent coordination strategies in changing user states and environments. For example, when the user adjusts their posture, each heating module cannot adaptively adjust its output power and heat application area in real time, resulting in poor heat therapy effect. 2. Inaccurate temperature monitoring and control: Traditional equipment often uses single-point contact temperature sensors or relies on preset heating curves, which makes it difficult to accurately reflect the true temperature distribution on the user's body surface and the equipment surface. This can easily lead to problems such as local overheating or insufficient heating. 3. Unreliable motion safety and positioning detection: Some devices lack effective detection of the user's seated state, posture and abnormal movement. If the system fails to respond in time when the user makes large movements on the device or leaves midway, it may lead to energy waste, or even continue to heat or run during abnormal movement, which may bring safety hazards. IV. Inadequate User Experience and Conflict Handling: Modern devices often support multiple operation methods such as touch screen, voice, and remote control. When multiple commands are generated at the same time, the existing system lacks a clear operation arbitration logic, which can easily lead to operation conflicts and chaotic system response, reducing user experience and security. To address the aforementioned technical deficiencies, a solution is proposed. Summary of the Invention

[0003] The purpose of this invention is to automatically generate adaptive thermotherapy plans based on the user's physiological data and thermotherapy rehabilitation course, assisting physicians in performing thermotherapy operations. Furthermore, through a collaborative control module and comprehensive multimodal feedback, it achieves intelligent linkage and adaptive adjustment of multiple thermotherapy modules such as the shoulder, neck, and hip, significantly improving the targeting and comfort of thermotherapy.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a control system for an intelligent shoulder, neck, and hip hyperthermia device based on multimodal feedback, comprising a multimodal sensing module, a central processing unit, a hyperthermia execution module, and a physician terminal, wherein: The multimodal sensing module is used to acquire multi-point temperature feedback signals, the user's historical thermotherapy data and real-time physiological data, as well as user operation command signals in real time and send them to the central processing unit. The central processing unit is electrically connected to the multimodal sensing module and configured to run multiple functional logic modules; The functional logic modules of the central processing unit include a temperature parameter generation module, a multi-point fusion temperature monitoring module, an operation arbitration module, and a collaborative control module, wherein: The temperature parameter generation module is used to acquire the user's historical thermotherapy data and real-time physiological data, build a plan generation model based on deep learning algorithms, and generate the current thermotherapy plan by combining historical thermotherapy data. The current thermotherapy plan is sent to the doctor's end. Based on the doctor's judgment, an executable thermotherapy plan is generated, and then a thermotherapy operation instruction is generated and sent to the operation arbitration module. At the same time, the executable thermotherapy plan is adaptively optimized based on real-time physiological data, and an optimized operation instruction is generated and sent to the operation arbitration module. The multi-point fusion temperature monitoring module is used to acquire real-time temperature data, generate a high-precision real-time temperature field distribution based on the fusion algorithm, and combine it with the PID control algorithm to achieve precise closed-loop temperature control of the thermotherapy execution module. The operation arbitration module is used to acquire thermotherapy operation instructions and receive user operation instructions from different channels. When an instruction conflict is detected, the module arbitrates the operation according to the preset safety priority rules and outputs a unique valid execution instruction to the collaborative control module. The collaborative control module is used to dynamically adjust the working parameters of each hyperthermia execution module based on the valid execution instructions output by the operation arbitration module and the temperature data provided by the multi-point fusion temperature monitoring module, so as to realize the intelligent and personalized collaborative work of multiple modules.

[0005] Furthermore, the thermotherapy execution module is electrically connected to the central processing unit and includes a shoulder heating unit, a neck heating unit, and a hip heating unit.

[0006] Furthermore, the physician terminal is used to obtain an executable hyperthermia plan, which includes a target temperature curve, heating duration, heating rate, and hyperthermia mode. Based on the patient's historical hyperthermia data, the physician makes an empirical judgment on the executable hyperthermia plan and makes auxiliary corrections to obtain an optimized executable hyperthermia plan. Based on the optimized executable hyperthermia plan, a hyperthermia operation instruction is generated and sent to the operation arbitration module.

[0007] Furthermore, the specific process of obtaining an executable hyperthermia plan and generating optimized operating instructions is as follows: S101. Obtain the user's historical thermotherapy data and real-time physiological data. The historical thermotherapy data includes the user's identification information, historical thermotherapy plan, and post-thermotherapy feedback data. The historical thermotherapy plan specifically includes the target temperature curve, heating duration, heating rate, and thermotherapy mode. The real-time physiological data is collected in real time by the PPG photoplethysmography sensor and GSR skin conductance sensor integrated into the intelligent shoulder, neck and hip thermotherapy device. The real-time physiological data includes heart rate, heart rate variability and skin conductance. Heart rate and heart rate variability are used to characterize the user's autonomic nervous system state. A decrease in heart rate variability and a sharp increase in heart rate indicate that the user is in a state of tension, discomfort or pain. Electrodermal conductance is used to characterize a user's emotional arousal and sympathetic nervous system excitation level. Sudden peaks in electrodermal conductance are a sensitive indicator of stress. S102. A scheme generation model is constructed based on a long short-term memory network. Historical hyperthermia schemes are normalized and integrated to obtain training samples. The specific training process is as follows: Training inputs include user identification information, historical hyperthermia protocols and post-hyperthermia feedback data, as well as physiological data at the beginning of the treatment course. Training tag: Obtain the optimal thermotherapy plan, which is defined as: the thermotherapy parameter curve that achieves the highest physician score, the highest patient subjective comfort, and the most significant pain relief improvement, while ensuring safety. Training objective: To learn the complex nonlinear mapping relationship from patient status to the optimal hyperthermia regimen; S103. Retrieve the patient's historical hyperthermia data and collect the real-time physiological data before the start of this hyperthermia as the baseline value. Input the above data into the pre-trained protocol generation model and output a multi-dimensional current hyperthermia protocol. S104. During the execution of the current hyperthermia protocol, the temperature parameter generation module continuously receives real-time physiological data streams from the multimodal sensing module and continuously compares the real-time physiological data with the baseline values ​​collected before the start of this hyperthermia. S105. Based on physiological feedback, set the adaptive optimization rule for optimization logic, evaluate the comparison results according to the optimization logic, and generate optimization operation instructions based on the comparison results.

[0008] Furthermore, the specific process for achieving precise closed-loop temperature control of the thermotherapy execution module is as follows: S201. Acquire real-time temperature data, which includes thermotherapy temperature data and body surface temperature data. The thermotherapy temperature data is collected by various contact sensors integrated on the base of the thermotherapy execution module, and the body surface temperature data is collected by a non-contact temperature sensor set between the intelligent shoulder, neck and hip thermotherapy device and the user's skin. S202. Obtain the equipment parameters of the intelligent shoulder, neck and hip heat therapy device, establish a heat control physical model based on heat conduction, and use the body surface temperature data as the target temperature data. The heat control physical model is used to evaluate the time and attenuation required for heat to be transferred from the heat therapy execution module to the body surface. S203. At time k, input the thermotherapy temperature data of the thermal execution unit at time k-1 into the thermal control physical model to obtain the predicted value of the body surface temperature at time k. S204. Calculate the Kalman gain between the predicted body surface temperature and the body surface temperature data based on the Kalman filter algorithm. Obtain the fused temperature at time k, i.e., the optimal target temperature value, based on the weighted combination of the Kalman gains. Adjust the temperature of the thermotherapy execution module according to the optimal target temperature value to achieve precise closed-loop temperature control.

[0009] Furthermore, the specific process of outputting a unique and valid execution instruction is as follows: S301. Obtain thermotherapy operation instructions and user operation instructions from different channels, and generate corresponding instruction tags according to the instruction content. The instruction tags include the highest priority P1, user safety veto P2, physician professional instruction P3, and user comfort request P4. The thermotherapy operation instructions are generated by the temperature parameter generation module and issued after being reviewed, confirmed, or modified by the physician. The content of the thermotherapy operation instructions is at the professional adjustment level, for example: Initial plan: "{Neck: 42°C, Shoulders: 40°C, Total duration: 20 minutes, Maximum safety threshold: 44°C}"; Mid-process adjustment (triggered by AI physiological feedback and confirmed by a physician): "{Real-time heart rate is too high, neck target temperature reduced to 41°C}"; The user operation commands are initiated directly by the user through different human-computer interaction channels. The user operation commands are instantaneous and intention-oriented, such as: "Stop", "Turn up the temperature", "Too hot", "I want shoulder mode". S302. Preset security priority rules, which are used to define the processing order between different instruction tags, specifically: The highest priority, P1, is the system-level emergency stop, which is triggered by: a physical emergency stop button or a hardware over-temperature protection signal from another module. The arbitration logic is immediate execution, meaning that the highest priority instruction unconditionally overrides any other instruction that is being executed or awaiting execution. The user can veto P2, which is triggered by a veto command initiated by the user. The arbitration logic is to be executed immediately, ensuring the user's priority of immediate safety and physical comfort; Physician Professional Instruction P3: Its trigger condition is: AI-generated and physician-confirmed thermotherapy instruction; The arbitration logic is to implement the basic protocol for hyperthermia, which defines the baseline and safety limits of the treatment course; User comfort request P4: Its trigger condition is: a comfort request command initiated by the user; The arbitration logic is to enforce the binding order, compare it with the safety limit set by the physician professional instruction P3, and give way to the physician professional instruction; Example: P3 sets the upper limit to 44°C, the current temperature is 42°C, P4 requests "heating up", the arbitration result is: execution is allowed, and the command "heat up 1°C" is output; Example: P3 sets the upper limit to 44°C, the current temperature is 44°C, P4 requests "heating up", the arbitration result is: refuse to execute, and outputs "the maximum set temperature has been reached" to the HMI; S303. Conflict detection is performed based on an arbitration mechanism that is triggered by events and time windows, including instantaneous conflicts and state conflicts. Instantaneous conflicts are when two or more instructions from different sources with the same target parameters are received within a preset time window. State conflicts are when the received instructions conflict with the safety limit set by the physician's professional instructions. Specifically: Scenario 1: Doctor (P3) vs. User's Temperature Reduction (P2) t=0ms: Received physician P3 instruction: "[Neck] Target temperature set to 43°C"; t=50ms: Received user P2 command (voice): "[Neck] is too hot"; Arbitration process: When a conflict is detected by the module, the priority is compared: P2 > P3; Arbitration result: P2 instruction is adopted, P3 instruction is discarded; Output: = "{Execution: [Neck] temperature decreased by 1°C}", the only valid execution instruction.

[0010] Scenario 2: User Growth (P4) vs. Doctor Limit (P3) t=0ms: P3 command has taken effect, setting "[Shoulder] maximum safety threshold = 45°C"; t=5min: [Shoulder area] Current temperature has reached 45°C; t=5min,1s: Received user P4 command (touchscreen): "[Shoulder] Temperature Increase"; Arbitration process: The module detected request P4. It retrieved the state limits set by P3 and found that request P4 (heating) violated the upper limit of P3 (45°C). Arbitration result: P4 order rejected; Output: "{Execution: Null}" (i.e., no command is sent to the collaborative control module), while "{Feedback: Maximum temperature reached}" is sent to the HMI module, which is the only valid execution command.

[0011] Scenario 3: Multi-channel user command conflict (P4 vs P4) t=0ms: Received user P4 command (touchscreen): "[Shoulder] Temperature Increase"; t=30ms: Received user P4 command (voice): "[Shoulder] mode switch"; Arbitration process: A conflict with the same P4 priority was detected. At this point, the channel sub-priority rule is activated; Sub-rule (preferred): Physical touch / button (high intent) > Voice (may be accidentally triggered); Arbitration result: The "touchscreen" command is adopted; Output: = "{Execution: [Shoulder] Temperature rises by 1°C}", the only valid execution instruction.

[0012] S304. After arbitration logic, a winning instruction is generated in each preset arbitration cycle. The winning instruction is formatted into a standardized instruction set that the collaborative control module can understand, i.e., a unique valid execution instruction.

[0013] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: The control system of this intelligent shoulder, neck, and hip hyperthermia device based on multimodal feedback can automatically generate adaptive hyperthermia plans according to the user's physiological data and hyperthermia rehabilitation course, assisting physicians in hyperthermia operations. Furthermore, through a collaborative control module, it integrates multimodal feedback to achieve intelligent linkage and adaptive adjustment of multiple hyperthermia modules such as the shoulder, neck, and hip, significantly improving the targeting and comfort of hyperthermia. At the same time, it adopts multi-point fusion temperature monitoring technology, combining the advantages of non-contact and contact sensors, to achieve precise control of the temperature field, effectively avoiding local overheating or underheating, ensuring high safety. Moreover, it has a clear operation arbitration module to solve the problem of command conflicts in multi-channel operation, ensuring the uniqueness and safety of system response, and optimizing the human-computer interaction experience. Attached Figure Description Figure 1 A schematic diagram of the overall external structure of the present invention is shown. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Example: like Figure 1 As shown, the control system of the intelligent shoulder, neck, and hip hyperthermia device based on multimodal feedback includes a multimodal sensing module, a central processing unit, a hyperthermia execution module, and a physician terminal, wherein: The multimodal sensing module is used to acquire multi-point temperature feedback signals, the user's historical thermotherapy data and real-time physiological data, as well as user operation command signals in real time and send them to the central processing unit. The central processing unit is electrically connected to the multimodal sensing module and configured to run multiple functional logic modules; The functional logic modules of the central processing unit include a temperature parameter generation module, a multi-point fusion temperature monitoring module, an operation arbitration module, and a collaborative control module, among which: The temperature parameter generation module is used to acquire the user's historical thermotherapy data and real-time physiological data, build a plan generation model based on deep learning algorithms, and generate the current thermotherapy plan by combining historical thermotherapy data. The current thermotherapy plan is sent to the doctor's end. Based on the doctor's judgment, an executable thermotherapy plan is generated, and then a thermotherapy operation instruction is generated and sent to the operation arbitration module. At the same time, the executable thermotherapy plan is adaptively optimized based on real-time physiological data, and an optimized operation instruction is generated and sent to the operation arbitration module. The multi-point fusion temperature monitoring module is used to acquire real-time temperature data, generate a high-precision real-time temperature field distribution based on the fusion algorithm, and combine it with the PID control algorithm to achieve precise closed-loop temperature control of the thermotherapy execution module. The operation arbitration module is used to acquire thermotherapy operation instructions and receive user operation instructions from different channels. When an instruction conflict is detected, the module arbitrates the operation according to the preset safety priority rules and outputs a unique valid execution instruction to the collaborative control module. The collaborative control module is used to dynamically adjust the working parameters of each hyperthermia execution module based on the valid execution instructions output by the operation arbitration module and the temperature data provided by the multi-point fusion temperature monitoring module, so as to realize the intelligent and personalized collaborative work of multiple modules.

[0016] The hyperthermia execution module is electrically connected to the central processing unit and includes a shoulder heating unit, a neck heating unit, and a hip heating unit.

[0017] The physician's end is used to obtain executable hyperthermia plans, which include target temperature curves, heating duration, heating rate, and hyperthermia mode. Based on the patient's historical hyperthermia data, the physician makes an empirical judgment on the executable hyperthermia plans and makes auxiliary corrections to obtain optimized executable hyperthermia plans. Based on the optimized executable hyperthermia plans, the physician generates hyperthermia operation instructions and sends them to the operation arbitration module.

[0018] The specific process of obtaining an executable hyperthermia plan and generating optimized operating instructions is as follows: S101. Obtain the user's historical hyperthermia data and real-time physiological data. The historical hyperthermia data includes the user's identification information, historical hyperthermia plans, and post-hyperthermia feedback data. The historical hyperthermia plans specifically include the target temperature curve, heating duration, heating rate, and hyperthermia mode. Real-time physiological data is collected in real time by the PPG photoplethysmography sensor and GSR skin conductance sensor integrated into the intelligent shoulder, neck and hip thermotherapy device. The real-time physiological data includes heart rate, heart rate variability and skin conductance. Heart rate and heart rate variability are used to characterize the user's autonomic nervous system state. A decrease in heart rate variability and a sharp increase in heart rate indicate that the user is in a state of tension, discomfort or pain. Electrodermal conductance is used to characterize a user's emotional arousal and sympathetic nervous system excitation level. Sudden peaks in electrodermal conductance are a sensitive indicator of stress. S102. A scheme generation model is constructed based on a long short-term memory network. Historical hyperthermia schemes are normalized and integrated to obtain training samples. The specific training process is as follows: Training inputs include user identification information, historical hyperthermia protocols and post-hyperthermia feedback data, as well as physiological data at the beginning of the treatment course. Training tag: Obtain the optimal thermotherapy plan, which is defined as: the thermotherapy parameter curve that achieves the highest physician score, the highest patient subjective comfort, and the most significant pain relief improvement, while ensuring safety. Training objective: To learn the complex nonlinear mapping relationship from patient status to the optimal hyperthermia regimen; S103. Retrieve the patient's historical hyperthermia data and collect the real-time physiological data before the start of this hyperthermia as the baseline value. Input the above data into the pre-trained protocol generation model and output a multi-dimensional current hyperthermia protocol. S104. During the execution of the current hyperthermia protocol, the temperature parameter generation module continuously receives real-time physiological data streams from the multimodal sensing module and continuously compares the real-time physiological data with the baseline values ​​collected before the start of this hyperthermia. S105. Based on physiological feedback, set the adaptive optimization rule for optimization logic, evaluate the comparison results according to the optimization logic, and generate optimization operation instructions based on the comparison results.

[0019] The specific process for achieving precise closed-loop temperature control of the hyperthermia execution module is as follows: S201. Acquire real-time temperature data, which includes thermotherapy temperature data and body surface temperature data. The thermotherapy temperature data is collected by various contact sensors integrated on the base of the thermotherapy execution module, and the body surface temperature data is collected by non-contact temperature sensors set between the intelligent shoulder, neck and hip thermotherapy device and the user's skin. S202. Obtain the equipment parameters of the intelligent shoulder, neck and hip heat therapy device, establish a heat control physical model based on heat conduction, use the body surface temperature data as the target temperature data, and use the heat control physical model to evaluate the time and attenuation required for heat to be transferred from the heat therapy execution module to the body surface. S203. At time k, input the thermotherapy temperature data of the thermal execution unit at time k-1 into the thermal control physical model to obtain the predicted value of the body surface temperature at time k. S204. Calculate the Kalman gain between the predicted body surface temperature and the body surface temperature data based on the Kalman filter algorithm. Obtain the fused temperature at time k, i.e., the optimal target temperature value, based on the weighted combination of the Kalman gains. Adjust the temperature of the thermotherapy execution module according to the optimal target temperature value to achieve precise closed-loop temperature control.

[0020] The specific process of outputting a unique and valid instruction is as follows: S301. Obtain thermotherapy operation instructions and user operation instructions from different channels, and generate corresponding instruction tags according to the instruction content. The instruction tags include the highest priority P1, user safety veto P2, physician professional instructions P3, and user comfort request P4. The hyperthermia operation instructions are generated by the temperature parameter generation module and issued after being reviewed, confirmed, or modified by the physician. The content of the hyperthermia operation instructions is at the professional adjustment level, for example: Initial plan: "{Neck: 42°C, Shoulders: 40°C, Total duration: 20 minutes, Maximum safety threshold: 44°C}"; Mid-process adjustment (triggered by AI physiological feedback and confirmed by a physician): "{Real-time heart rate is too high, neck target temperature reduced to 41°C}"; User operation commands are initiated directly by the user through different human-computer interaction channels. User operation commands are instantaneous and intention-oriented, such as: "Stop", "Turn up the temperature", "It's too hot", "I want shoulder mode". S302. Preset security priority rules. Security priority rules are used to define the processing order between different instruction tags. Specifically: The highest priority, P1, is the system-level emergency stop, which is triggered by: a physical emergency stop button or a hardware over-temperature protection signal from another module. The arbitration logic is immediate execution, meaning that the highest priority instruction unconditionally overrides any other instruction that is being executed or awaiting execution. The user can veto P2, which is triggered by a veto command initiated by the user. The arbitration logic is to be executed immediately, ensuring the user's priority of immediate safety and physical comfort; Physician Professional Instruction P3: Its trigger condition is: AI-generated and physician-confirmed thermotherapy instruction; The arbitration logic is to implement the basic protocol for hyperthermia, which defines the baseline and safety limits of the treatment course; User comfort request P4: Its trigger condition is: a comfort request command initiated by the user; The arbitration logic is to enforce the binding order, compare it with the safety limit set by the physician professional instruction P3, and give way to the physician professional instruction; Example: P3 sets the upper limit to 44°C, the current temperature is 42°C, P4 requests "heating up", the arbitration result is: execution is allowed, and the command "heat up 1°C" is output; Example: P3 sets the upper limit to 44°C, the current temperature is 44°C, P4 requests "heating up", the arbitration result is: refuse to execute, and outputs "the maximum set temperature has been reached" to the HMI; S303. Conflict detection is performed based on an arbitration mechanism that is triggered by events and time windows, including instantaneous conflicts and state conflicts. Instantaneous conflicts are when two or more instructions from different sources with the same target parameters are received within a preset time window. State conflicts are when the received instructions conflict with the safety limit set by the physician's professional instructions. Specifically: Scenario 1: Doctor (P3) vs. User's Temperature Reduction (P2) t=0ms: Received physician P3 instruction: "[Neck] Target temperature set to 43°C"; t=50ms: Received user P2 command (voice): "[Neck] is too hot"; Arbitration process: When a conflict is detected by the module, the priority is compared: P2 > P3; Arbitration result: P2 instruction is adopted, P3 instruction is discarded; Output: = "{Execution: [Neck] temperature decreased by 1°C}", the only valid execution instruction.

[0021] Scenario 2: User Growth (P4) vs. Doctor Limit (P3) t=0ms: P3 command has taken effect, setting "[Shoulder] maximum safety threshold = 45°C"; t=5min: [Shoulder area] Current temperature has reached 45°C; t=5min,1s: Received user P4 command (touchscreen): "[Shoulder] Temperature Increase"; Arbitration process: The module detected request P4. It retrieved the state limits set by P3 and found that request P4 (heating) violated the upper limit of P3 (45°C). Arbitration result: P4 order rejected; Output: "{Execution: Null}" (i.e., no command is sent to the collaborative control module), while "{Feedback: Maximum temperature reached}" is sent to the HMI module, which is the only valid execution command.

[0022] Scenario 3: Multi-channel user command conflict (P4 vs P4) t=0ms: Received user P4 command (touchscreen): "[Shoulder] Temperature Increase"; t=30ms: Received user P4 command (voice): "[Shoulder] mode switch"; Arbitration process: A conflict with the same P4 priority was detected. At this point, the channel sub-priority rule is activated; Sub-rule (preferred): Physical touch / button (high intent) > Voice (may be accidentally triggered); Arbitration result: The "touchscreen" command is adopted; Output: = "{Execution: [Shoulder] Temperature rises by 1°C}", the only valid execution instruction.

[0023] S304. After arbitration, a winning instruction is generated within each preset arbitration cycle. The winning instruction is formatted into a standardized instruction set that the collaborative control module can understand, i.e., a unique and valid execution instruction.

[0024] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value.

[0025] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A control system for an intelligent shoulder, neck, and hip hyperthermia device based on multimodal feedback, characterized in that: It includes a multimodal sensing module, a central processing unit, a hyperthermia execution module, and a physician terminal, among which: The multimodal sensing module is used to acquire multi-point temperature feedback signals, the user's historical thermotherapy data and real-time physiological data, as well as user operation command signals in real time and send them to the central processing unit. The central processing unit is electrically connected to the multimodal sensing module and configured to run multiple functional logic modules; The functional logic modules of the central processing unit include a temperature parameter generation module, a multi-point fusion temperature monitoring module, an operation arbitration module, and a collaborative control module, wherein: The temperature parameter generation module is used to acquire the user's historical thermotherapy data and real-time physiological data, build a plan generation model based on deep learning algorithms, and generate the current thermotherapy plan by combining historical thermotherapy data. The current thermotherapy plan is sent to the doctor's end. Based on the doctor's judgment, an executable thermotherapy plan is generated, and then a thermotherapy operation instruction is generated and sent to the operation arbitration module. At the same time, the executable thermotherapy plan is adaptively optimized based on real-time physiological data, and an optimized operation instruction is generated and sent to the operation arbitration module. The multi-point fusion temperature monitoring module is used to acquire real-time temperature data, generate a high-precision real-time temperature field distribution based on the fusion algorithm, and combine it with the PID control algorithm to achieve precise closed-loop temperature control of the thermotherapy execution module. The operation arbitration module is used to acquire thermotherapy operation instructions and receive user operation instructions from different channels. When an instruction conflict is detected, the module arbitrates the operation according to the preset safety priority rules and outputs a unique valid execution instruction to the collaborative control module. The collaborative control module is used to dynamically adjust the working parameters of each hyperthermia execution module based on the valid execution instructions output by the operation arbitration module and the temperature data provided by the multi-point fusion temperature monitoring module, so as to realize the intelligent and personalized collaborative work of multiple modules.

2. The control system of the intelligent shoulder, neck, and hip thermotherapy device based on multimodal feedback according to claim 1, characterized in that, The hyperthermia execution module is electrically connected to the central processing unit and includes a shoulder heating unit, a neck heating unit, and a hip heating unit.

3. The control system of the intelligent shoulder, neck, and hip thermotherapy device based on multimodal feedback according to claim 1, characterized in that, The physician terminal is used to obtain an executable hyperthermia plan, which includes a target temperature curve, heating duration, heating rate, and hyperthermia mode. Based on the patient's historical hyperthermia data, the physician makes an empirical judgment on the executable hyperthermia plan and makes auxiliary corrections to obtain an optimized executable hyperthermia plan. Based on the optimized executable hyperthermia plan, a hyperthermia operation instruction is generated and sent to the operation arbitration module.

4. The control system of the intelligent shoulder, neck, and hip thermotherapy device based on multimodal feedback according to claim 1, characterized in that, The specific process of obtaining an executable hyperthermia plan and generating optimized operating instructions is as follows: S101. Obtain the user's historical thermotherapy data and real-time physiological data. The historical thermotherapy data includes the user's identification information, historical thermotherapy plan, and post-thermotherapy feedback data. The historical thermotherapy plan specifically includes the target temperature curve, heating duration, heating rate, and thermotherapy mode. The real-time physiological data is collected in real time by the PPG photoplethysmography sensor and GSR skin conductance sensor integrated into the intelligent shoulder, neck and hip thermotherapy device. The real-time physiological data includes heart rate, heart rate variability and skin conductance. S102. A scheme generation model is constructed based on a long short-term memory network. Historical hyperthermia schemes are normalized and integrated to obtain training samples. The specific training process is as follows: Training inputs include user identification information, historical hyperthermia protocols and post-hyperthermia feedback data, as well as physiological data at the beginning of the treatment course. Training tag: Obtain the optimal thermotherapy plan, which is defined as: the thermotherapy parameter curve that achieves the highest physician score, the highest patient subjective comfort, and the most significant pain relief improvement, while ensuring safety. Training objective: To learn the complex nonlinear mapping relationship from patient status to the optimal hyperthermia regimen; S103. Retrieve the patient's historical hyperthermia data and collect the real-time physiological data before the start of this hyperthermia as the baseline value. Input the above data into the pre-trained protocol generation model and output a multi-dimensional current hyperthermia protocol. S104. During the execution of the current hyperthermia protocol, the temperature parameter generation module continuously receives real-time physiological data streams from the multimodal sensing module and continuously compares the real-time physiological data with the baseline values ​​collected before the start of this hyperthermia. S105. Based on physiological feedback, set the adaptive optimization rule for optimization logic, evaluate the comparison results according to the optimization logic, and generate optimization operation instructions based on the comparison results.

5. The control system of the intelligent shoulder, neck, and hip thermotherapy device based on multimodal feedback according to claim 1, characterized in that, The specific process for achieving precise closed-loop temperature control of the thermotherapy execution module is as follows: S201. Acquire real-time temperature data, which includes thermotherapy temperature data and body surface temperature data. The thermotherapy temperature data is collected by various contact sensors integrated on the base of the thermotherapy execution module, and the body surface temperature data is collected by a non-contact temperature sensor set between the intelligent shoulder, neck and hip thermotherapy device and the user's skin. S202. Obtain the equipment parameters of the intelligent shoulder, neck and hip heat therapy device, establish a heat control physical model based on heat conduction, and use the body surface temperature data as the target temperature data. The heat control physical model is used to evaluate the time and attenuation required for heat to be transferred from the heat therapy execution module to the body surface. S203. At time k, input the thermotherapy temperature data of the thermal execution unit at time k-1 into the thermal control physical model to obtain the predicted value of the body surface temperature at time k. S204. Calculate the Kalman gain between the predicted body surface temperature and the body surface temperature data based on the Kalman filter algorithm. Obtain the fused temperature at time k, i.e., the optimal target temperature value, based on the weighted combination of the Kalman gains. Adjust the temperature of the thermotherapy execution module according to the optimal target temperature value to achieve precise closed-loop temperature control.

6. The control system of the intelligent shoulder, neck, and hip thermotherapy device based on multimodal feedback according to claim 1, characterized in that, The specific process of outputting a unique and valid instruction is as follows: S301. Obtain thermotherapy operation instructions and user operation instructions from different channels, and generate corresponding instruction tags according to the instruction content. The instruction tags include the highest priority P1, user safety veto P2, physician professional instruction P3, and user comfort request P4. The thermotherapy operation instructions are generated by the temperature parameter generation module and issued after being reviewed, confirmed or modified by the physician. The content of the thermotherapy operation instructions is at the professional adjustment level. S302. Preset security priority rules, which are used to define the processing order between different instruction tags, specifically: The highest priority, P1, is the system-level emergency stop, which is triggered by: a physical emergency stop button or a hardware over-temperature protection signal from another module. The arbitration logic is immediate execution, meaning that the highest priority instruction unconditionally overrides any other instruction that is being executed or awaiting execution. The user can veto P2, which is triggered by a veto command initiated by the user. The arbitration logic is to be executed immediately, ensuring the user's priority of immediate safety and physical comfort; Physician Professional Instruction P3: Its trigger condition is: AI-generated and physician-confirmed thermotherapy instruction; The arbitration logic is to implement the basic protocol for hyperthermia, which defines the baseline and safety limits of the treatment course; User comfort request P4: Its trigger condition is: a comfort request command initiated by the user; The arbitration logic is to enforce the binding order, compare it with the safety limit set by the physician professional instruction P3, and give way to the physician professional instruction; S303. Conflict detection is performed based on an arbitration mechanism that is triggered by events and time windows, including instantaneous conflicts and state conflicts. Instantaneous conflicts are when two or more instructions from different sources with the same target parameters are received within a preset time window. State conflicts are when the received instructions conflict with the safety limit set by the physician's professional instructions. S304. After arbitration logic, a winning instruction is generated in each preset arbitration cycle. The winning instruction is formatted into a standardized instruction set that the collaborative control module can understand, i.e., a unique valid execution instruction.

Images