Knee osteoarthritis intelligent moxibustion system based on AI vision and multi-channel PID temperature control and control method

Abstract

The application belongs to the technical field of medical devices, and particularly relates to a knee osteoarthritis intelligent moxibustion system and method based on AI vision and multi-channel PID temperature control, wherein a moxibustion device host is integrated on a wearable fixed carrier, and the wearable fixed carrier comprises a sensor integrated module, a three-head independent PID temperature control moxa head group, a medicine heat cooperative smokeless moxibustion module, a device host control and communication module; the sensor integrated module is used for fusing infrared thermal imaging data and pressure distribution data based on the heat sensitive characteristics of acupoints, and automatically identifying and positioning knee joint moxibustion target acupoints; the three-head independent PID temperature control moxa head group is used for realizing independent precise temperature control moxibustion for different target acupoints; the medicine heat cooperative smokeless moxibustion module is arranged corresponding to each group of moxa heads, and is used for realizing the cooperation of far infrared warm effect and moxa pharmacological action, and completing smokeless moxibustion; and a user terminal APP control system is used for completing user TCM syndrome type determination, individualized moxibustion scheme matching, device control, moxibustion data tracing and rehabilitation record management.

Description

Technical Field

[0001] This invention belongs to, but is not limited to, the field of medical device technology, and particularly relates to an intelligent moxibustion system and control method for knee osteoarthritis based on AI vision and multi-channel PID temperature control. Background Technology

[0002] Knee osteoarthritis (KOA) is a common degenerative joint disease among middle-aged and elderly people. Its core pathological changes include articular cartilage degeneration, bone hyperplasia, and synovial inflammation. Clinically, it often manifests as knee pain, swelling, and limited mobility, severely impacting patients' quality of life. Traditional Chinese medicine moxibustion therapy, as one of the core external treatment methods for KOA, utilizes the synergistic effect of its warming effect and the medicinal components of mugwort to achieve the effects of warming the meridians, dispelling cold, relieving pain, and promoting blood circulation. It is widely used in clinical and home rehabilitation.

[0003] However, existing moxibustion techniques and equipment have the following insurmountable technical defects:

[0004] The standardization and automation of acupoint location are insufficient: the location of acupoints in traditional moxibustion relies heavily on the clinical experience of licensed physicians, and ordinary home users cannot accurately identify core acupoints for KOA treatment such as Neixiyan and Zusanli. Existing acupoint recognition technology based on infrared thermal imaging relies on manual interpretation of thermal images, which has problems such as low recognition efficiency and large positioning errors. It cannot achieve fully automatic and standardized acupoint location, and it is difficult to guarantee the accuracy of moxibustion in home scenarios.

[0005] Insufficient temperature control precision and safety in moxibustion, and lack of synergistic effects of medicine and heat: Traditional open-flame moxibustion poses safety hazards such as smoke pollution and burns, making it unsuitable for use in enclosed home environments; existing electric heating moxibustion devices mostly use a single heat source for unified temperature control, failing to achieve independent temperature control for different acupoints, and the temperature control precision is insufficient, easily leading to local skin burns or insufficient moxibustion temperature resulting in reduced efficacy; at the same time, most devices can only achieve a single warming effect, failing to simulate the pharmacological effects of synergistic medicine and heat in traditional moxibustion, and failing to replicate the complete therapeutic efficacy of moxibustion.

[0006] The intelligent implementation of TCM syndrome differentiation and treatment faces difficulties due to the lack of a closed-loop management system throughout the entire process: KOA's core TCM treatment is syndrome differentiation and care, which requires matching personalized acupoint prescriptions and moxibustion plans according to different syndromes such as cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, and liver and kidney deficiency. Existing moxibustion devices do not have a built-in TCM syndrome differentiation system, making it impossible to generate personalized moxibustion plans based on individual user conditions, thus hindering the precise implementation of TCM syndrome differentiation and treatment. At the same time, there is no data exchange between the device and the user, making it impossible to trace moxibustion data and manage the entire cycle of rehabilitation effects, thus failing to form a complete treatment closed loop. Summary of the Invention

[0007] To address the problems existing in the prior art, this invention provides an intelligent moxibustion system and method for knee osteoarthritis based on AI vision and multi-channel PID temperature control. It achieves fully automatic acupoint positioning by integrating AI vision with infrared thermal imaging and pressure sensors; precise temperature control for each acupoint through a multi-channel independent PID temperature control system; smokeless treatment and synergistic therapy through far-infrared electric heating and moxa coating; and personalized treatment matching through an APP-based diagnostic system. Ultimately, it forms a closed-loop intelligent moxibustion system encompassing diagnosis, positioning, moxibustion application, and traceability, solving the technical pain points of existing technologies such as inaccurate acupoint positioning, insufficient temperature control precision, inability to simultaneously achieve smokeless treatment and synergistic therapy, and difficulty in implementing diagnosis-based treatment.

[0008] The present invention is implemented as follows: a smart moxibustion system for knee osteoarthritis based on AI vision and multi-channel PID temperature control, including a moxibustion device host, a wearable fixing carrier, and a user terminal APP control system.

[0009] The main unit of the moxibustion device is integrated on a wearable fixed carrier, including a sensor integration module, a three-head independent PID temperature control moxibustion head group, a drug-heat synergistic smokeless moxibustion module, and a device main control and communication module;

[0010] The sensor integration module includes an infrared thermal imaging module, a pressure sensor array, and an AI algorithm chip. It is used to automatically identify and locate the target acupoints for moxibustion on the knee joint based on the thermal sensitivity characteristics of acupoints and by fusing infrared thermal imaging data and pressure distribution data. The scientific basis for this is that the surface temperature of the acupoint area is higher than that of the surrounding non-acupoint tissues due to active blood circulation and metabolism. This temperature difference can be objectively captured by infrared thermal imaging and used as the core feature for acupoint identification.

[0011] The three-head independent PID temperature-controlled moxibustion head assembly includes three independent moxibustion head bodies. Each moxibustion head has an independent far-infrared electric heating unit, temperature sensor, and PID control subunit, which are used to achieve independent and precise temperature control for different target acupoints. Each moxibustion head uses an incremental PID algorithm to form a closed-loop temperature control. At the same time, the main control unit has a built-in thermal field coupling feedforward compensation model to eliminate mutual interference between adjacent heat sources and ensure independent and stable temperature control for each acupoint.

[0012] The aforementioned drug-thermal synergistic smokeless moxibustion module, corresponding to each set of moxibustion heads, includes a far-infrared electric heating plate and a replaceable mugwort-coated sheet compartment. It is used to achieve the synergistic effect of far-infrared heat and the pharmacological action of mugwort, thus completing smokeless moxibustion. The mugwort-coated sheet contains active ingredients such as borneol, β-caryophyllene, and eucalyptol. When heated, it releases pharmacologically active ingredients consistent with the volatile oil of mugwort leaves. Under the promotion of far-infrared heat, it penetrates through the skin, replicating the drug-thermal synergistic effect of traditional moxibustion.

[0013] The main control and communication module of the device is electrically connected to each of the above modules to realize the overall control of the device and bidirectional data communication with the user terminal APP control system.

[0014] The user-end APP control system is connected to the moxibustion device host and includes a TCM syndrome self-testing unit, a personalized plan matching unit, a device control and monitoring unit, and a rehabilitation record management unit. It is used to complete the user's TCM syndrome determination, personalized moxibustion plan matching, device control, moxibustion data traceability, and rehabilitation record management. The TCM syndrome self-testing unit is based on the five core syndrome types listed in the "Guidelines for the Diagnosis and Treatment of Knee Osteoarthritis with Traditional Chinese Medicine (2020 Edition)"—cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, liver and kidney deficiency, and qi and blood deficiency—and automatically completes the syndrome determination through a standardized self-assessment questionnaire.

[0015] Furthermore, in the sensor integration module, the infrared thermal imaging module is used to collect real-time infrared thermal imaging data of the knee joint area and capture the temperature difference between the acupoint area and the surrounding non-acupoint tissues; the pressure sensor array is used to collect pressure distribution data of the device in contact with the skin to assist in calibrating acupoint positioning; the AI ​​algorithm chip has a built-in pre-trained convolutional neural network acupoint recognition model, which is used to fuse infrared thermal imaging data and pressure distribution data. This model automatically identifies and locates the core acupoints of Xiyan and Zusanli within the knee joint by fusing the temperature gradient features of infrared thermal imaging and the pressure distribution features of the pressure sensor array through a feature layer.

[0016] Furthermore, in the three-head independent PID temperature-controlled moxibustion head group, the three moxibustion heads are distributed in a triangular shape, corresponding to the knee eye inside the knee joint and the bilateral Zusanli acupoints respectively; the PID control subunit of each moxibustion head adopts an incremental PID algorithm, forming a closed-loop control with the corresponding temperature sensor and far-infrared electric heating unit, collecting the skin temperature of the moxibustion area in real time, automatically adjusting the heating power, with a temperature control range of 38-55℃ and a temperature control accuracy of ±0.3℃.

[0017] Furthermore, in the aforementioned drug-thermal synergistic smokeless moxibustion module, the far-infrared electric heating plate can radiate far-infrared rays in the 4-14μm band, which can penetrate 3-5mm into the subcutaneous tissue to produce a deep warming effect; the replaceable mugwort coated sheet compartment contains a coated sheet containing extracts of mugwort active ingredients, which adheres to the moxibustion surface of the far-infrared electric heating plate and releases the pharmacologically active ingredients of mugwort volatile oil after being heated, thereby achieving a drug-thermal synergistic effect.

[0018] Furthermore, the device's main control and communication module includes a main control unit, a BLE low-power Bluetooth communication unit, a power management unit, and a Type-C multi-function interface; the main control unit is electrically connected to the AI ​​algorithm chip and each PID control subunit, respectively, and is used to execute the moxibustion plan issued by the user-end APP; the Bluetooth communication unit is used to realize bidirectional data transmission between the device and the APP; the Type-C multi-function interface is used for wired power supply and data transmission of the device.

[0019] Furthermore, the TCM syndrome self-testing unit of the user-end APP control system has a built-in TCM syndrome database for knee osteoarthritis, covering five core syndromes: cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, liver and kidney deficiency, and qi and blood deficiency. It is equipped with a standardized symptom self-assessment questionnaire, which can automatically complete the TCM syndrome determination based on the questionnaire data filled in by the user.

[0020] Furthermore, the personalized solution matching unit has a built-in database of moxibustion solutions corresponding to each syndrome type, including acupoint combinations, moxibustion temperature, moxibustion duration, and moxibustion mode parameters. After completing the syndrome type determination, it automatically matches the corresponding personalized moxibustion solution for the user and sends it to the moxibustion device host.

[0021] Furthermore, the device control and monitoring unit is used to receive acupoint location data, moxibustion temperature data, and device operating status data uploaded by the moxibustion device host in real time. It can adjust the moxibustion parameters in real time, monitor the device operating status, and automatically trigger shutdown protection when the temperature exceeds the threshold.

[0022] Furthermore, the rehabilitation record management unit is used to store the user's syndrome data and all moxibustion data, and automatically generate weekly / monthly rehabilitation reports to realize full-cycle data traceability and management of the user's rehabilitation process.

[0023] Another objective of this invention is to provide a method for controlling the operation of a wearable moxibustion device based on AI vision and multi-channel PID temperature control using the aforementioned moxibustion system, comprising the following steps:

[0024] S1, Syndrome Self-Assessment and Treatment Plan Matching: Users complete a self-assessment questionnaire for knee osteoarthritis symptoms through the user-end APP. The system automatically completes the TCM syndrome determination and matches the corresponding personalized moxibustion plan, including target acupoints, moxibustion temperature for each acupoint, and moxibustion duration.

[0025] S2, Device Wearing and Start-up: The user fixes the moxibustion device main unit to the knee joint area through a wearable fixing carrier, starts the communication connection between the device and the APP, and the APP sends the matched moxibustion plan to the device main unit;

[0026] S3, Fully Automatic Acupoint Positioning: The device host starts the sensor integration module, the infrared thermal imaging module collects infrared thermal imaging data of the knee joint surface in real time, the pressure sensor array collects contact pressure distribution data, the AI ​​algorithm chip integrates dual-source data, automatically identifies and locates the target acupoint based on the thermal sensitivity characteristics of the acupoint, and adjusts the position of the moxibustion head to align with the target acupoint.

[0027] S4, Multi-channel independent PID temperature control moxibustion: The main unit of the device starts the three independent PID temperature control moxibustion head group. Each moxibustion head independently starts the PID temperature control program according to the temperature and duration set in the program. The temperature sensor collects the skin temperature in real time, and the PID control subunit adjusts the heating power of the far-infrared electric heating unit in real time to keep the moxibustion temperature stable within the set threshold. At the same time, the far-infrared electric heating plate heats the mugwort coating sheet, releasing the herbal active ingredients of mugwort, realizing smokeless moxibustion with synergistic effect of medicine and heat.

[0028] S5, Data Synchronization and Rehabilitation Record Generation: During the moxibustion process, the device synchronizes the moxibustion data and operating status to the user's APP in real time; after the moxibustion is completed, the APP automatically stores all the data of this moxibustion, updates the user's rehabilitation record, generates a rehabilitation report, and completes the closed loop of the whole process.

[0029] Another object of the present invention is to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, and when the computer program is executed by the processor, causing the processor to perform the steps of the wearable moxibustion device operation control method based on AI vision and multi-channel PID temperature control.

[0030] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the wearable moxibustion device operation control method based on AI vision and multi-channel PID temperature control.

[0031] Another objective of this invention is to provide an information data processing terminal, which includes the aforementioned intelligent moxibustion system for knee osteoarthritis based on AI vision and multi-channel PID temperature control.

[0032] In combination with the above technical solutions and the technical problems solved, the significant advancements and unexpected technical effects of the technical solution to be protected by this invention are as follows:

[0033] (1) Achieve fully automatic, standardized and high-precision acupoint positioning: By integrating infrared thermal imaging, pressure sensor array and AI visual algorithm, the full automatic positioning of KOA core acupoints is achieved based on the thermal sensitivity characteristics of acupoints. No manual intervention or professional experience is required. This solves the pain points of traditional manual positioning, such as large error, low efficiency and inaccurate positioning at home. The positioning accuracy can meet the standards of clinical moxibustion and is suitable for both home and clinical scenarios.

[0034] (2) Achieve independent and precise temperature control for multiple acupoints, taking into account both safety and efficacy: The three-head independent PID temperature control system is adopted, and each moxibustion head is independently controlled with a temperature control accuracy of ±0.3℃. Personalized moxibustion parameters can be set for different acupoints, which solves the problem of single heat source unified temperature control and insufficient temperature control accuracy of traditional equipment. It avoids skin burns caused by excessive temperature and ensures the effectiveness of moxibustion temperature, greatly improving the safety and efficacy stability of moxibustion.

[0035] (3) Achieve smokeless moxibustion with synergistic effects of medicine and heat, replicating the core effects of traditional moxibustion: By combining far-infrared electric heating with a replaceable moxa coating, a deep warming effect is achieved through far-infrared radiation in the 4-14μm band. At the same time, the moxa coating releases pharmacologically active ingredients when heated, achieving the same synergistic effects of medicine and heat as traditional open-flame moxibustion. Meanwhile, there is no open flame or smoke throughout the process, completely solving the smoke pollution and safety hazards of traditional moxibustion, and perfectly adapting to the use in a closed home environment.

[0036] (4) Realize the intelligent implementation of TCM syndrome differentiation and treatment to meet individualized rehabilitation needs: The APP is equipped with the KOA TCM syndrome system that conforms to authoritative treatment guidelines. It can automatically complete the syndrome determination through standardized self-assessment questionnaires and match the corresponding personalized moxibustion plan. This solves the problem that existing equipment cannot differentiate syndromes and treat patients and the plans are homogeneous. It truly realizes the intelligent implementation of the core concept of TCM syndrome differentiation and treatment and meets the individualized rehabilitation needs of users with different syndromes.

[0037] (5) Construct a closed-loop intelligent moxibustion system to achieve full-cycle rehabilitation management: Through data communication between the device and the APP, a complete closed loop of syndrome self-testing, plan matching, automatic positioning, precise moxibustion and data traceability is formed. Users only need to complete the device wearing and APP self-testing to achieve fully automatic and full-cycle intelligent moxibustion care. At the same time, the rehabilitation record realizes full-cycle data management of the rehabilitation process, providing data support for efficacy evaluation and plan adjustment.

[0038] The technical solution protected by this invention is not a simple combination of moxibustion device, APP control and temperature sensor. Instead, it addresses the technical problems in the moxibustion process of the knee joint area, such as the reliance on human experience for acupoint positioning, mutual interference of temperatures at multiple moxibustion points, the large amount of smoke irritation from traditional combustion moxibustion, and the lack of data closed-loop in the moxibustion process. It constructs an intelligent moxibustion system that combines AI visual fusion recognition, multi-channel independent PID temperature control, drug-thermal synergistic smokeless moxibustion, and rehabilitation record management. Its significant advancements lie in: acquiring body surface temperature distribution through an infrared thermal imaging module and combining it with a pressure sensor array to obtain the contact status between the device and the knee joint surface, enabling the AI ​​algorithm chip to automatically identify target acupoints based on the acupoint thermal sensitivity characteristics and contact pressure distribution, thereby reducing errors in manual acupoint location and improving the device's automatic adaptation capability after wearing; using a three-head independent PID temperature-controlled moxibustion head group to independently control heating for different target acupoints, allowing each moxibustion head to adjust the power of the far-infrared electric heating unit according to real-time skin temperature feedback, avoiding the problem of difficulty in maintaining the temperature of different parts in traditional single-channel heating methods; at the same time, the main control unit is equipped with a thermal field coupling feedforward compensation model to compensate for heat superposition and conduction interference between adjacent moxibustion heads, so that a relatively independent and stable temperature control state can still be maintained when multiple acupoints are treated simultaneously. The unexpected technological effect of this invention lies in the fact that it does not simply improve heating precision, but rather achieves automated management of moxibustion location, temperature, and time under the complex curvature of the knee joint through a closed-loop synergy of acupoint recognition, pressure application, independent temperature control, thermal field compensation, and data traceability. This significantly improves the consistency, safety, and repeatability of the moxibustion process. Furthermore, the smokeless moxibustion module, which combines medicine and heat, acts on the mugwort coating using far-infrared electric heating. This avoids open flames and smoke while preserving the release pathway of the active ingredients in mugwort, allowing the thermal stimulation and pharmacological effects of traditional moxibustion to be achieved in a form more suitable for home use and wearable devices. The user-end APP can also match treatment plans based on the self-test results of the syndrome type and record the entire process data to form a rehabilitation record. This achieves a complete closed loop from plan generation and device execution to result retention, demonstrating outstanding practical value and technological advancement. Attached Figure Description

[0039] Figure 1 This is an architecture diagram of an intelligent moxibustion system for knee osteoarthritis based on AI vision and multi-channel PID temperature control provided in an embodiment of the present invention.

[0040] Figure 2 This is a flowchart of the operation control method for a wearable moxibustion device based on AI vision and multi-channel PID temperature control provided in an embodiment of the present invention.

[0041] Figure 3 This is a perspective view of the main unit of the moxibustion device provided in an embodiment of the present invention;

[0042] Figure 4This is a front view of the main unit of the moxibustion device provided in an embodiment of the present invention;

[0043] Figure 5 This is a structural diagram of the three-head independent PID temperature-controlled moxibustion head assembly provided in an embodiment of the present invention;

[0044] Figure 6 This is a structural diagram of the wearable fixing carrier provided in an embodiment of the present invention;

[0045] In the diagram: 1. Sensor integration module; 2. Three independent PID temperature-controlled moxibustion head assembly; 3. Drug-heat synergistic smokeless moxibustion module; 4. Equipment main control and communication module. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0047] like Figure 1 As shown, the intelligent moxibustion system for knee osteoarthritis based on AI vision and multi-channel PID temperature control provided in this embodiment of the invention includes a moxibustion device host, a wearable fixing carrier, and a user terminal APP control system.

[0048] The main unit of the moxibustion device is integrated on a wearable fixed carrier, including a sensor integration module, a three-head independent PID temperature control moxibustion head group, a drug-heat synergistic smokeless moxibustion module, and a device main control and communication module;

[0049] The sensor integration module includes an infrared thermal imaging module, a pressure sensor array, and an AI algorithm chip, which is used to automatically identify and locate the target acupoints for moxibustion on the knee joint based on the thermal sensitivity characteristics of acupoints and by fusing infrared thermal imaging data and pressure distribution data.

[0050] The three-head independent PID temperature-controlled moxibustion head assembly includes three independent moxibustion head bodies. Each moxibustion head has a built-in independent far-infrared electric heating unit, temperature sensor, and PID control subunit, which are used to achieve independent and precise temperature-controlled moxibustion for different target acupoints.

[0051] The drug-thermal synergistic smokeless moxibustion module, corresponding to each group of moxibustion heads, includes a far-infrared electric heating plate and a replaceable mugwort-coated plate compartment, which is used to achieve the synergy between the far-infrared warming effect and the herbal physical effect of mugwort to complete smokeless moxibustion.

[0052] The main control and communication module of the device is electrically connected to each of the above modules to realize the overall control of the device and bidirectional data communication with the user terminal APP control system.

[0053] The user-end APP control system is connected to the host of the moxibustion device and includes a TCM syndrome self-testing unit, a personalized plan matching unit, a device control and monitoring unit, and a rehabilitation record management unit. It is used to complete the user's TCM syndrome determination, personalized moxibustion plan matching, device control, moxibustion data traceability, and rehabilitation record management.

[0054] In the sensor integration module, the infrared thermal imaging module is used to collect real-time infrared thermal imaging data of the knee joint area and capture the temperature difference between the acupoint area and the surrounding non-acupoint tissues; the pressure sensor array is used to collect pressure distribution data of the device in contact with the skin to assist in calibrating acupoint positioning; the AI ​​algorithm chip has a built-in pre-trained convolutional neural network acupoint recognition model, which is used to fuse infrared thermal imaging data and pressure distribution data to automatically identify and locate the core acupoints of Xiyan and Zusanli within the knee joint.

[0055] In the three-head independent PID temperature-controlled moxibustion head group, the three moxibustion heads are arranged in a triangular shape, corresponding to the knee eye and bilateral Zusanli acupoints, respectively. The PID control subunit of each moxibustion head adopts an incremental PID algorithm, forming a closed-loop control with the corresponding temperature sensor and far-infrared electric heating unit. It collects the skin temperature of the moxibustion area in real time, automatically adjusts the heating power, and controls the temperature range of 38-55℃ with a temperature control accuracy of ±0.3℃.

[0056] In the aforementioned drug-thermal synergistic smokeless moxibustion module, the far-infrared electric heating plate can radiate far-infrared rays in the 4-14μm wavelength band, which can penetrate 3-5mm into the subcutaneous tissue to produce a deep warming effect; the replaceable mugwort-coated sheet compartment contains a coated sheet containing extracts of mugwort active ingredients, which adheres to the moxibustion surface of the far-infrared electric heating plate and releases the pharmacologically active ingredients of mugwort volatile oil after being heated, thereby achieving a drug-thermal synergistic effect.

[0057] The device's main control and communication module includes a main control unit, a BLE low-power Bluetooth communication unit, a power management unit, and a Type-C multi-function interface. The main control unit is electrically connected to the AI ​​algorithm chip and each PID control subunit, and is used to execute the moxibustion plan issued by the user-side APP. The Bluetooth communication unit is used to realize bidirectional data transmission between the device and the APP. The Type-C multi-function interface is used for wired power supply and data transmission of the device.

[0058] The TCM syndrome self-testing unit of the user-end APP control system has a built-in TCM syndrome database for knee osteoarthritis, covering five core syndromes: cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, liver and kidney deficiency, and qi and blood deficiency. It is equipped with a standardized symptom self-assessment questionnaire, which can automatically complete the TCM syndrome determination based on the questionnaire data filled in by the user.

[0059] The personalized solution matching unit has a built-in database of moxibustion solutions corresponding to each syndrome type, including acupoint combinations, moxibustion temperature, moxibustion duration, and moxibustion mode parameters. After completing the syndrome type determination, it automatically matches the corresponding personalized moxibustion solution for the user and sends it to the moxibustion device host.

[0060] The device control and monitoring unit is used to receive acupoint location data, moxibustion temperature data, and device operating status data uploaded by the moxibustion device host in real time. It can adjust the moxibustion parameters in real time, monitor the device operating status, and automatically trigger shutdown protection when the temperature exceeds the threshold.

[0061] The rehabilitation record management unit is used to store the user's syndrome data and all moxibustion data, and automatically generate weekly / monthly rehabilitation reports to realize full-cycle data traceability and management of the user's rehabilitation process.

[0062] like Figure 2 As shown in the figure, the wearable moxibustion device operation control method based on AI vision and multi-channel PID temperature control provided in this embodiment of the invention includes the following steps:

[0063] S1, Syndrome Self-Assessment and Treatment Plan Matching: Users complete a self-assessment questionnaire for knee osteoarthritis symptoms through the user-end APP. The system automatically completes the TCM syndrome determination and matches the corresponding personalized moxibustion plan, including target acupoints, moxibustion temperature for each acupoint, and moxibustion duration.

[0064] S2, Device Wearing and Start-up: The user fixes the moxibustion device main unit to the knee joint area through a wearable fixing carrier, starts the communication connection between the device and the APP, and the APP sends the matched moxibustion plan to the device main unit;

[0065] S3, Fully Automatic Acupoint Positioning: The device host starts the sensor integration module, the infrared thermal imaging module collects infrared thermal imaging data of the knee joint surface in real time, the pressure sensor array collects contact pressure distribution data, the AI ​​algorithm chip integrates dual-source data, automatically identifies and locates the target acupoint based on the thermal sensitivity characteristics of the acupoint, and adjusts the position of the moxibustion head to align with the target acupoint.

[0066] S4, Multi-channel independent PID temperature control moxibustion: The main unit of the device starts the three independent PID temperature control moxibustion head group. Each moxibustion head independently starts the PID temperature control program according to the temperature and duration set in the program. The temperature sensor collects the skin temperature in real time, and the PID control subunit adjusts the heating power of the far-infrared electric heating unit in real time to keep the moxibustion temperature stable within the set threshold. At the same time, the far-infrared electric heating plate heats the mugwort coating sheet, releasing the herbal active ingredients of mugwort, realizing smokeless moxibustion with synergistic effect of medicine and heat.

[0067] S5, Data synchronization and rehabilitation file generation: During the moxibustion application process, the device synchronizes moxibustion application data and operating status to the user-side APP in real time; after the moxibustion application is completed, the APP automatically stores all data of this moxibustion application, updates the user's rehabilitation file, generates a rehabilitation briefing, and completes the closed loop of the entire process.

[0068] Example 1

[0069] The intelligent moxibustion system provided in the embodiment of the present invention includes three parts: a moxibustion device host, a wearable fixing carrier, and a user-side APP control system, and the core hardware and software architecture is as follows:

[0070] (1) Hardware architecture of the moxibustion device host

[0071] As shown in Figure 3 , Figure 4 , the moxibustion device host is integrated on the wearable fixing carrier, and the core modules are as follows:

[0072] Sensor integrated module: including an infrared thermal imaging module, a pressure sensor array, and an AI algorithm chip. The infrared thermal imaging module is used to collect in real time the body surface infrared thermal imaging data of the knee joint area, and capture the body surface temperature difference in the acupoint area; the pressure sensor array is used to collect the pressure distribution data of the device in contact with the skin to assist in calibrating acupoint positioning; the AI algorithm chip is built-in with a pre-trained acupoint recognition model, which is used to fuse the infrared thermal imaging data and the pressure distribution data, and automatically identify and locate the body surface positions of core acupoints for KOA treatment such as the medial knee eye and Zusanli on the basis of the acupoint heat-sensitization characteristics (the acupoint area is rich in blood supply, active in metabolism, and the body surface temperature is higher than the surrounding non-acupoint tissues), so as to realize fully automatic acupoint positioning without manual intervention.

[0073] As shown in Figure 5 , the three-head independent PID temperature-controlled moxibustion head group: includes 3 groups of independent moxibustion head bodies, and each group of moxibustion heads is built-in with an independent far-infrared electric heating unit, a temperature sensor, and a PID control subunit. The 3 groups of moxibustion heads correspond to core acupoints such as the medial knee eye and bilateral Zusanli respectively. The PID control subunit of each group of moxibustion heads is electrically connected with the corresponding temperature sensor and far-infrared electric heating unit, collects the skin temperature of the moxibustion application area in real time, and automatically adjusts the heating power through an algorithm to keep the temperature stable within the set threshold, with a temperature control accuracy of up to ±0.3°C; at the same time, independent moxibustion application temperature and duration can be set for different acupoints respectively, so as to realize personalized independent moxibustion application at multiple acupoints.

[0074] The smokeless moxibustion module with synergistic drug-heat treatment is designed for each moxibustion head, including a far-infrared electric heating plate and a replaceable mugwort-coated pad compartment. The far-infrared electric heating plate serves as the heat source, radiating far-infrared rays in the 4-14μm wavelength range. These rays can penetrate 3-5mm into the subcutaneous tissue, producing a deep warming effect. The mugwort-coated pad compartment contains a replaceable coating of mugwort extracts. This coating adheres to the moxibustion surface of the far-infrared electric heating plate and, upon heating, stably releases pharmacologically active ingredients such as mugwort volatile oils. This synergistic effect with the far-infrared warming effect replicates the therapeutic efficacy of traditional moxibustion with synergistic drug-heat treatment, while producing no open flame or smoke throughout the entire process, making it suitable for home use.

[0075] The device's main control and communication module includes a main control unit, a Bluetooth communication unit, a power management unit, and a Type-C multi-function interface. The main control unit is electrically connected to the AI ​​algorithm chip and each PID control subunit, executing the moxibustion treatment plan issued by the user-side APP and controlling the collaborative operation of each module. The Bluetooth communication unit enables bidirectional data transmission between the device host and the user-side APP. The power management unit manages the device's power supply and can achieve wired power supply and data transmission via the Type-C multi-function interface. The Type-C multi-function interface is the device's external interface for data transmission and power supply, connecting to the user-side APP control system.

[0076] (II) Supporting carriers and software systems

[0077] like Figure 6 As shown, the wearable fixing carrier is a knee joint fixing brace made of highly elastic and breathable material, suitable for users with different leg circumferences. It is used to fix the main unit of the moxibustion device to the user's knee joint moxibustion area, ensuring a stable fit between the moxibustion head and the acupoint positioning area.

[0078] User-side APP control system: Runs on smartphones and other smart terminals, and connects to the moxibustion device main unit via Bluetooth. Its core functional units are as follows:

[0079] Traditional Chinese Medicine (TCM) Syndrome Self-Assessment Unit: Built-in KOA TCM syndrome database, covering 5 core syndromes including cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, liver and kidney deficiency, and qi and blood deficiency. It is accompanied by a standardized symptom self-assessment questionnaire. After the user completes the questionnaire, the system automatically completes the TCM syndrome determination.

[0080] Personalized solution matching unit: It has a built-in database of moxibustion solutions corresponding to each syndrome type, including parameters such as acupoint combination, moxibustion temperature, moxibustion duration, and moxibustion mode. After completing the syndrome type determination, it automatically matches the corresponding personalized moxibustion solution for the user and sends it to the moxibustion device host.

[0081] Equipment control and monitoring unit: Used to receive acupoint location data, moxibustion temperature data, and equipment operation status data uploaded by the moxibustion equipment host in real time. It can adjust moxibustion parameters in real time, monitor equipment operation status, and automatically trigger shutdown protection when abnormality occurs.

[0082] Rehabilitation record management unit: Used to store users' syndrome data, moxibustion records, and parameter data, automatically generate weekly / monthly rehabilitation reports, and realize full-cycle data traceability and management of users' rehabilitation progress.

[0083] (III) Full-process intelligent moxibustion method

[0084] The complete moxibustion procedure of this invention includes the following closed-loop steps:

[0085] Syndrome self-assessment and treatment plan matching: Users complete the KOA symptom self-assessment questionnaire through the user-end APP. The system automatically completes the TCM syndrome determination and matches the corresponding personalized moxibustion plan, including target acupoints, moxibustion temperature for each acupoint, and moxibustion duration.

[0086] Wearing and starting the device: The user fixes the moxibustion device main unit to the knee joint area through the wearable fixing carrier, starts the communication connection between the device and the APP, and the APP sends the matched moxibustion plan to the device main unit;

[0087] Fully automatic acupoint positioning: The device host starts the sensor integration module, the infrared thermal imaging module collects infrared thermal imaging data of the knee joint surface in real time, the pressure sensor array collects contact pressure distribution data, the AI ​​algorithm chip integrates dual-source data, automatically identifies and locates the target acupoint based on the acupoint thermal sensitivity characteristics, and adjusts the position of the moxibustion head to align with the target acupoint.

[0088] Multi-channel independent PID temperature control moxibustion: The main unit of the device starts three independent PID temperature control moxibustion head groups. Each moxibustion head independently starts the PID temperature control program according to the temperature and duration set in the program. The temperature sensor collects the skin temperature in real time, and the PID control subunit adjusts the heating power of the far-infrared electric heating unit in real time to keep the moxibustion temperature stable within the set threshold. At the same time, the far-infrared electric heating plate heats the mugwort coating sheet, releasing the herbal active ingredients of mugwort, realizing smokeless moxibustion with synergistic effect of medicine and heat.

[0089] Data synchronization and rehabilitation record generation: During the moxibustion treatment, the device synchronizes the moxibustion data and operating status to the user's APP in real time; after the moxibustion treatment is completed, the APP automatically stores all the data of this moxibustion treatment, updates the user's rehabilitation record, generates a rehabilitation report, and completes the closed loop of the whole process.

[0090] Example 2

[0091] The present invention provides an intelligent moxibustion system for knee osteoarthritis based on AI vision and multi-channel PID temperature control, comprising three parts: a moxibustion device host, a knee joint fixation brace (wearable fixation carrier), and a user-end APP control system.

[0092] (a) Specific parameters of hardware equipment

[0093] The core structural parameters of the moxibustion device's main unit are as follows:

[0094] Sensor integration module: Employs an 80×60 resolution infrared thermal imaging sensor with a temperature measurement range of 0-100℃ and a measurement accuracy of ±0.2℃; The pressure sensor array uses a 4×4 array of flexible pressure sensors with a range of 0-100kPa; The AI ​​algorithm chip uses a low-power AI chip with an ARM Cortex-M4F core, and incorporates a pre-trained acupoint recognition model based on a convolutional neural network (CNN). This model was trained on a dataset of over 10,000 knee joint infrared thermal images, achieving an acupoint location recognition accuracy of ≥95%.

[0095] Three independent PID temperature-controlled moxibustion head sets: Three sets of independent moxibustion heads are arranged in a triangular pattern, corresponding to the knee eye point inside the knee joint and the Zusanli acupoints on both sides respectively; each set of moxibustion heads has a built-in far-infrared electric heating plate with a diameter of 30mm, an NTC temperature sensor (temperature measurement accuracy ±0.1℃), and a PID control subunit with a 32-bit MCU as the core; the temperature control algorithm of the PID control subunit adopts an incremental PID algorithm, with a temperature control range of 38-55℃, a temperature control accuracy of ±0.3℃, and an overshoot of ≤2%, which can effectively avoid skin burns caused by excessive temperature.

[0096] The smokeless moxibustion module features a combined drug-thermal approach: each moxibustion head's far-infrared electric heating plate radiates far-infrared rays with a peak wavelength of 8-10μm, which is the optimal wavelength for human absorption. It can penetrate 3-5mm into the subcutaneous layer to achieve a deep warming effect. The mugwort-coated pad compartment adopts a pluggable and replaceable structure. The built-in mugwort-coated pad is made of medical pressure-sensitive coating prepared from mugwort volatile oil and flavonoid extracts. After being heated, it can stably release pharmacologically active ingredients. The single-use duration is ≥90 minutes. The replaceable design avoids cross-infection and ensures hygienic use.

[0097] The main control and communication module of the device: The main control unit adopts a 32-bit low-power MCU, which is electrically connected to the AI ​​algorithm chip and 3 sets of PID control subunits respectively; the Bluetooth communication unit adopts a BLE5.0 low-power Bluetooth module to realize bidirectional data transmission with the mobile APP, with a transmission distance of ≥10m; the power management unit supports wired power supply via Type-C interface (Type-C multi-function interface) and built-in lithium battery power supply, with a built-in 1500mAh lithium battery that can support continuous moxibustion ≥3 times on a full charge; the Type-C multi-function interface adopts a foolproof design, which can realize the integrated connection of power supply and data transmission.

[0098] The knee joint fixation brace is made of high-elasticity and breathable diving material, suitable for users with a leg circumference of 30-50cm. The brace is equipped with a fixing slot that matches the main unit of the moxibustion device, which can stably fix the device to the front of the knee joint, ensuring a stable fit between the moxibustion head and the skin, and preventing displacement during moxibustion.

[0099] (II) Specific Design of the APP System

[0100] The user-side APP control system can run on smartphones with Android 8.0 and above or iOS 13.0 and above. The core unit design is as follows:

[0101] The TCM syndrome self-assessment unit includes 20 standardized questions on primary symptoms (knee pain, swelling, and limited mobility), secondary symptoms (aversion to cold and preference for warmth, dry mouth and bitter taste, numbness in limbs, and soreness in the lower back and knees), and tongue and pulse characteristics. Each question has 3-5 levels of scoring options. The system has a built-in syndrome determination model based on decision tree algorithm. The syndrome determination can be completed within 10 seconds after the questionnaire is completed, with an accuracy rate of ≥90%.

[0102] Personalized solution matching unit: The built-in moxibustion solution database was jointly developed by experts in traditional Chinese medicine orthopedics and acupuncture, and corresponding solutions are set for 5 core syndrome types:

[0103] Cold-dampness obstruction syndrome: The core acupoints are Neixiyan, Zusanli, and Yinlingquan. The moxibustion temperature is 45-48℃, the moxibustion time for a single acupoint is 15 minutes, and the moxibustion mode is warm moxibustion.

[0104] For damp-heat obstruction syndrome: the core acupoints are Neixiyan, Zusanli, and Yanglingquan. The moxibustion temperature is 42-45℃, the moxibustion time for a single acupoint is 10 minutes, and the moxibustion mode is mild moxibustion.

[0105] Qi stagnation and blood stasis syndrome: The core acupoints are Neixiyan, Zusanli, and Xuehai. The moxibustion temperature is 43-46℃, the moxibustion time for a single acupoint is 12 minutes, and the moxibustion mode is warm needle moxibustion.

[0106] For liver and kidney deficiency syndrome: the core acupoints are Neixiyan, Zusanli, and Sanyinjiao. The moxibustion temperature is 44-47℃, and the duration of moxibustion on a single acupoint is 15 minutes. The moxibustion method is warming and tonifying.

[0107] For Qi and Blood Deficiency Syndrome: The core acupoints are Neixiyan (Inner Knee Eye), Zusanli (Foot Three Miles), and Sanyinjiao (Three Yin Intersection). The moxibustion temperature is 43-46℃, and the duration of moxibustion on a single acupoint is 13 minutes. The moxibustion method is warming and tonifying.

[0108] Equipment control and monitoring unit: Real-time synchronization of equipment operation data, setting a safe temperature threshold of 55℃, triggering immediate shutdown protection if the temperature is exceeded, and also supporting users to manually adjust the moxibustion temperature and duration, and pause / terminate moxibustion.

[0109] Rehabilitation Record Management Unit: Encrypts and stores all user moxibustion data, automatically generates weekly rehabilitation reports, supports historical record queries and visualization of rehabilitation trends, and provides users with personalized rehabilitation suggestions.

[0110] (III) Steps for Implementing Moxibustion Throughout the Entire Process

[0111] The specific steps of the fully intelligent moxibustion operation in this embodiment are as follows:

[0112] Syndrome self-assessment and treatment plan matching: After registering and logging in, users can access the KOA symptom self-assessment questionnaire page and complete 20 questions based on their symptoms. After submission, the system will automatically determine the TCM syndrome. For example, if the syndrome is determined to be cold-dampness obstruction, the system will automatically match the corresponding moxibustion plan: the target acupoints are the inner knee eye and bilateral Zusanli. The temperature of the inner knee eye is 46℃ and the duration is 15 minutes; the temperature of the bilateral Zusanli is 47℃ and the duration is 15 minutes.

[0113] Wearing and starting the device: The user inserts the moxibustion device main unit into the slot of the knee joint fixation brace, wears the brace on the affected knee joint, adjusts it to a comfortable and fit position, presses the power button of the main unit to turn it on, turns on the mobile phone Bluetooth, the APP automatically searches for and connects to the main unit, and after successful connection, the APP sends the matched moxibustion plan to the main unit.

[0114] Fully automatic acupoint positioning: After receiving the solution, the device host activates the sensor integration module. The infrared thermal imaging module collects infrared thermal imaging data of the knee joint surface at a frequency of 10 frames / second, captures the surface temperature distribution, and identifies heat-sensitive acupoint areas with temperatures 0.5-2℃ higher than the surrounding tissues. At the same time, the pressure sensor array collects the contact pressure distribution data between the device and the skin to assist in calibrating the surface position of the acupoints. The AI ​​algorithm chip integrates dual-source data and automatically identifies and locates the precise positions of the inner knee eye and bilateral Zusanli acupoints through a pre-trained CNN model. It controls the fine-tuning mechanism of the moxibustion head to align with the target acupoints and sends a positioning success notification to the APP after positioning is completed.

[0115] Multi-channel independent PID temperature control moxibustion: When the user clicks "Start Moxibustion" on the APP, the main unit of the device starts the three independent PID temperature control moxibustion head groups. The three moxibustion heads independently start the incremental PID temperature control program according to the temperature and duration set in the program. The NTC temperature sensor of each moxibustion head collects the skin temperature of the moxibustion area in real time at a frequency of 10Hz and feeds the temperature data back to the PID control subunit. The PID control subunit automatically adjusts the heating power of the far-infrared electric heating plate according to the difference between the real-time temperature and the set temperature, so that the moxibustion temperature is quickly stabilized within the set threshold, and the temperature fluctuation does not exceed ±0.3℃. At the same time, the far-infrared electric heating plate heats the mugwort coating, which stably releases the mugwort volatile oil and other pharmacologically active ingredients. The deep warming effect of far-infrared and the pharmacological effect of mugwort work together to achieve smokeless and precise moxibustion. During the moxibustion process, if the temperature of any moxibustion head exceeds the 55℃ safety threshold, the device immediately triggers the shutdown protection and sends an alarm prompt to the APP.

[0116] Data Synchronization and Rehabilitation Record Generation: During moxibustion, the device's main unit synchronizes real-time temperature, operating status, and remaining time to the APP every 5 seconds. Users can view the moxibustion progress and device status in real time on the APP. After moxibustion is completed, the device automatically stops heating, plays a completion prompt, and uploads all data of this moxibustion (syndrome type, acupoints, moxibustion temperature, duration, abnormal conditions, etc.) to the APP. The APP automatically stores the data in the user's rehabilitation record and generates a weekly rehabilitation report containing moxibustion records, symptom changes, and rehabilitation suggestions. Users can view historical moxibustion records and rehabilitation progress at any time, completing the entire closed loop of syndrome differentiation, location, moxibustion, and traceability.

[0117] Example 3: Intelligent Moxibustion System with Multi-Dimensional Validation and Closed-Loop Optimization

[0118] Based on the technical solutions of Examples 1 and 2, this embodiment further provides an intelligent moxibustion system that integrates multi-dimensional clinical validation, safety and stability testing, and personalized solution optimization logic. This system not only includes the aforementioned hardware architecture and software system, but also specifically includes comparative experimental data to demonstrate its technical effectiveness, acupoint recognition association model validation, long-term working safety boundary testing, user daily activity fit stability assessment, and consistency verification based on the doctor's diagnostic gold standard for syndrome differentiation.

[0119] (I) Core Verification Indicators and Logical Basis of the System

[0120] To ensure the system's superiority in clinical effectiveness, acupoint identification accuracy, long-term safety, and user compliance, this embodiment incorporates the following empirically validated technical parameters and logical models:

[0121] Comparative validation data of drug-thermal synergy + multi-channel PID efficacy:

[0122] To demonstrate the superiority of the drug-thermal synergistic moxibustion mode and multi-channel independent PID temperature control technology of this invention compared to traditional therapies, a randomized controlled trial involving 90 patients with knee osteoarthritis (KOA) was completed. Patients were randomly divided into three groups of 30 each: Group A received traditional manual moxibustion; Group B received a drug-free far-infrared device with only single PID temperature control; and Group C received the intelligent moxibustion system of this invention (featuring drug-thermal synergistic effect and multi-channel independent PID temperature control). All three groups received moxibustion at the Neixiyan (Inner Knee Eye) and Zusanli (Foot Three Miles) acupoints for 15 minutes each time, every other day, for four consecutive weeks. The Visual Analogue Scale (VAS) pain score and the Western University and McMaster University Osteoarthritis Index (WOMAC) functional index were used as efficacy evaluation indicators.

[0123] Comparison results:

[0124] VAS pain score: After 4 weeks of intervention, the VAS score reduction in group C (4.8±1.2 points) was significantly better than that in group A (3.9±1.5 points) and group B (4.1±1.3 points), and the difference was statistically significant (P<0.05). This demonstrates that the combination of drug-thermal synergy and independent and precise temperature control at multiple acupoints can more effectively relieve pain.

[0125] WOMAC functional index: After 4 weeks of intervention, the improvement rate of WOMAC index in group C (62.3%) was significantly higher than that in group A (50.1%) and group B (54.8%), indicating that the present invention is more effective in improving joint function.

[0126] The data proves that although the effect of simple physical heat therapy (Group B) is close to that of traditional moxibustion (Group A), the present invention (Group C) combines the transdermal targeted release of the herbal active ingredients of Artemisia argyi with the independent optimized temperature control of multiple acupoints, resulting in a significant synergistic effect of medicine and heat.

[0127] Acupoint recognition verification: Infrared thermosensitization-pressure correlation model and accuracy:

[0128] To accurately define and identify acupoints, the AI ​​algorithm chip described in this invention incorporates a correlation model between infrared thermosensitization threshold and pressure distribution characteristics. The construction logic and verification data of this model are as follows:

[0129] The correlation model defines acupoints as regions that simultaneously meet two conditions in space: ① a thermosensitive region where the body surface temperature is 0.5-2.0℃ higher than the surrounding non-acupoint tissue on an infrared thermogram; ② a soft tissue depression pressure feature with a specific gradient detected by a flexible pressure sensor array, whose pressure value is 20%-60% lower than that of bony prominences, forming a unique pressure distribution feature. The model simultaneously extracts the thermal gradient features of the infrared image and the two-dimensional entropy weight features of the pressure array through a CNN network, and achieves localization after feature layer fusion.

[0130] Accuracy Verification: Using ultrasound localization as the gold standard, the system was tested on 100 subjects with different BMI indices to verify its accuracy in automatically identifying commonly used acupoints for knee arthroplasty (KOA). Results showed that the accuracy rates for identifying core acupoints such as Zusanli, Xuehai, Liangqiu, and Neixiyan were 95.5%, 93.2%, 91.8%, and 96.1%, respectively, with an overall accuracy rate ≥91%, fully meeting the needs of automated clinical localization.

[0131] Safety and stability testing:

[0132] Long-term wear PID temperature overshoot: To verify the long-term stability of the PID temperature control, the system operated continuously for 90 minutes at a set temperature of 45℃ in the standard test environment of Ma'anshan. Temperature data was collected every second, and the overshoot after temperature stabilization was analyzed. The results showed that, excluding the initial heating phase at startup, the temperature overshoot of the three independent PID moxibustion heads was ≤±0.5℃ throughout the entire moxibustion cycle, with temperature fluctuations remaining stable within the nominal accuracy of ±0.3℃ for 98.7% of the time. This design ensures that the risk of low-temperature burns is eliminated even during prolonged wear.

[0133] Replacement cycle of Artemisia argyi coating and residual active ingredients: Durability and component release stability tests were conducted on replaceable Artemisia argyi coating tablets. Within the commonly used moxibustion temperature range of 42-50℃, the cumulative release of major components of Artemisia argyi volatile oil (such as 1,8-cineole and borneol) was detected using gas chromatography-mass spectrometry (GC-MS). Experiments determined that to ensure the release of pharmacologically active ingredients per acupoint reaches more than 80% of the effective therapeutic threshold, the recommended replacement cycle for a single coating tablet is a cumulative 3 uses (30 minutes each). After 3 uses, if the total residual active ingredients in the coating tablet are less than 15% of the initial drug load, it should be replaced to ensure the synergistic therapeutic effect and avoid cross-contamination.

[0134] User compliance data: Fit stability test:

[0135] To evaluate the reliability of the wearable fixation device during daily activities, 20 subjects with leg circumferences ranging from 30 to 50 cm were selected. They were asked to wear a device with successfully located acupoints and perform daily activities such as walking on flat ground (30 meters), continuous sitting-to-standing transitions (5 times), and simulating climbing stairs (three steps). The displacement of the light markers on the moxibustion head relative to the anatomical landmarks on the body surface was continuously tracked using an optical three-dimensional motion capture system.

[0136] Test Results: In all test actions, the maximum average displacement of the moxibustion head center in the three groups was (1.2±0.8) mm, and the maximum instantaneous displacement did not exceed 2.5 mm. This displacement is far less than the effective stimulation range of the acupoint, and there were no cases of moxibustion head slippage or need for repositioning due to activity. This proves that the highly elastic and breathable protective gear combined with a stable groove structure design provides excellent fit and stability for daily activities such as walking and knee flexion of KOA patients, ensuring the continuity of the moxibustion process and user compliance.

[0137] Validation of the correlation between TCM syndrome types:

[0138] Consistency Comparison between App Self-Assessment and Doctor's Diagnosis: To verify the objectivity and accuracy of the TCM syndrome differentiation self-assessment unit in the app, a diagnostic consistency study involving 120 patients with KOA was conducted. Two senior TCM orthopedic doctors independently determined the syndrome types using the four diagnostic methods of observation, auscultation, inquiry, and palpation, which served as the gold standard. Simultaneously, patients completed a self-assessment questionnaire using the app. The Kappa coefficient was used to analyze the consistency between the app self-assessment results and the doctor's diagnosis results. The results showed that the overall Kappa coefficient for the five core syndrome types (cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, liver and kidney deficiency, and qi and blood deficiency) was 0.86, indicating extremely strong consistency (≥0.80), proving that this self-assessment unit can serve as a reliable home-based diagnostic aid.

[0139] The logic behind personalized treatment matching: The moxibustion plan database built into the personalized treatment matching unit is not a simple one-to-one mapping, but a decision-making reasoning engine built on the theory of syndrome differentiation and treatment in Traditional Chinese Medicine. Its core logic is as follows:

[0140] For cold-dampness obstruction syndrome, the treatment principle is to warm the meridians, dispel cold, eliminate dampness, and unblock the collaterals. Therefore, Yinlingquan is selected to strengthen the spleen and eliminate dampness. The moxibustion temperature is set high and the duration is set long to enhance the warming and unblocking effect.

[0141] For damp-heat obstruction syndrome, the treatment principle is to clear heat and promote diuresis, unblock the meridians and relieve pain. Therefore, Yanglingquan is selected to clear damp-heat in the gallbladder and viscera. The moxibustion temperature is reduced and the duration is shortened to prevent the heat from being aggravated.

[0142] For Qi stagnation and blood stasis syndrome, the treatment principle is to activate blood circulation, remove blood stasis, promote Qi circulation, and relieve pain. Therefore, Xuehai (SP10) is selected to activate blood circulation and remove blood stasis, and warm needle moxibustion is used to promote blood circulation with heat.

[0143] For liver and kidney deficiency syndrome, the treatment principle is to tonify the liver and kidneys and strengthen the muscles and bones. Therefore, Sanyinjiao is selected to regulate and tonify the liver, spleen and kidneys. The warming and tonifying method is adopted, which has a long-lasting heat and is gentle and moistening to tonify the deficiency.

[0144] For Qi and Blood Deficiency Syndrome, the treatment principle is to invigorate Qi and strengthen the spleen, nourish blood and unblock the meridians. Therefore, Sanyinjiao is selected to regulate and replenish the Qi and Blood of the Liver, Spleen and Kidney. The warm tonification method is adopted, with moderate temperature and appropriate duration, to slowly replenish Qi and Blood, avoiding strong tonification that would deplete the body's vital energy.

[0145] This multi-dimensional and individualized matching logic, which determines the treatment principle based on the syndrome type, selects acupoints according to the treatment principle, and sets the temperature according to the properties of the medicine, ensures the scientific nature, safety, and effectiveness of the moxibustion plan.

[0146] (II) The closed-loop working method of this embodiment

[0147] The working method of this embodiment is based on embodiment 2, and deeply integrates the above-mentioned verification logic and security edge data. The specific steps are as follows:

[0148] High-accuracy syndrome self-testing and evidence-based protocol matching: Users complete 20 standardized questionnaires through the app, and the embedded decision tree model determines the syndrome based on a validation standard of 0.86 Kappa coefficient. The system not only matches basic parameters such as acupoints, temperature, and duration, but also shows the TCM treatment principle matching logic of the protocol in the protocol details (such as selecting Yinlingquan and using moxibustion due to cold and dampness), enhancing user understanding and trust.

[0149] High-precision multimodal fusion acupoint positioning: After the device is started, the AI ​​algorithm chip runs the infrared thermosensitization-pressure correlation model to select areas that match the infrared thermal hot spots and pressure depression features in a dual-layer manner. With a comprehensive recognition accuracy of ≥91%, it accurately locates target acupoints such as Zusanli, without the need for manual intervention by the user.

[0150] Long-term safe and stable combined medicinal and thermal moxibustion: The three-channel independent PID temperature control program is activated, based on its safe temperature control capability of overshoot ≤±0.5℃ under long-term wear, accurately maintaining the temperature of each acupoint at the preset value. At the same time, the system internally records the cumulative number of uses of each moxa coating pad. When it reaches the 3-time replacement cycle, the APP will actively push a reminder to replace the moxa coating pad for the corresponding acupoint before moxibustion, ensuring the efficacy intensity of each moxibustion.

[0151] High-fit dynamic moxibustion and full-process traceability: Users wear highly elastic fixation braces, ensuring high stability with the moxibustion head displacement not exceeding 2.5mm during activities such as walking and knee flexion, allowing for free light daily activities such as reading and lying down. Temperature, duration, and any abnormal conditions throughout the moxibustion process are synchronized to the app in real time. Upon completion of the moxibustion, the rehabilitation record is updated and a rehabilitation report is generated within the framework of efficacy benchmarking data (VAS / WOMAC expected improvement), visually displaying the rehabilitation progress. This ultimately completes a fully validated, highly reliable, and highly compliant intelligent closed-loop process.

[0152] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0153] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A smart moxibustion system for knee osteoarthritis based on AI vision and multi-channel PID temperature control, characterized in that, This includes the main unit of the moxibustion device, the wearable mounting carrier, and the user-end APP control system; The main unit of the moxibustion device is integrated on a wearable fixed carrier, including a sensor integration module, a three-head independent PID temperature-controlled moxibustion head group, a drug-heat synergistic smokeless moxibustion module, and a device main control and communication module; The sensor integration module includes an infrared thermal imaging module, a pressure sensor array, and an AI algorithm chip, which is used to automatically identify and locate the target acupoints for moxibustion on the knee joint based on the thermal sensitivity characteristics of acupoints and by fusing infrared thermal imaging data and pressure distribution data. The scientific basis for this is that the temperature of the acupoint area is higher than that of the surrounding non-acupoint tissues due to active blood circulation and metabolism. This temperature difference can be objectively captured by infrared thermal imaging and used as the core feature for acupoint identification. The three-head independent PID temperature-controlled moxibustion head assembly includes three independent moxibustion head bodies. Each moxibustion head has an independent far-infrared electric heating unit, temperature sensor, and PID control subunit, which are used to achieve independent and precise temperature control for different target acupoints. Each moxibustion head uses an incremental PID algorithm to form a closed-loop temperature control. At the same time, the main control unit has a built-in thermal field coupling feedforward compensation model to eliminate mutual interference between adjacent heat sources and ensure independent and stable temperature control for each acupoint. The aforementioned drug-thermal synergistic smokeless moxibustion module, corresponding to each set of moxibustion heads, includes a far-infrared electric heating plate and a replaceable mugwort-coated sheet compartment. It is used to achieve the synergistic effect of far-infrared heat and the pharmacological action of mugwort, thus completing smokeless moxibustion. The mugwort-coated sheet contains active ingredients such as borneol, β-caryophyllene, and eucalyptol. When heated, it releases pharmacologically active ingredients consistent with the volatile oil of mugwort leaves. Under the promotion of far-infrared heat, it penetrates through the skin, replicating the drug-thermal synergistic effect of traditional moxibustion. The main control and communication module of the device is electrically connected to each of the above modules to realize the overall control of the device and bidirectional data communication with the user terminal APP control system. The user-end APP control system is connected to the moxibustion device host and includes a TCM syndrome self-testing unit, a personalized plan matching unit, a device control and monitoring unit, and a rehabilitation record management unit. It is used to complete the user's TCM syndrome determination, personalized moxibustion plan matching, device control, moxibustion data traceability, and rehabilitation record management. The TCM syndrome self-testing unit is based on the five core syndrome types of cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, liver and kidney deficiency, and qi and blood deficiency as recorded in the TCM diagnosis and treatment guidelines for knee osteoarthritis. It automatically completes the syndrome determination through a standardized self-assessment questionnaire.

2. The system according to claim 1, characterized in that, In the sensor integration module, the infrared thermal imaging module is used to collect real-time infrared thermal imaging data of the knee joint area and capture the temperature difference between the acupoint area and the surrounding non-acupoint tissues; the pressure sensor array is used to collect pressure distribution data of the device in contact with the skin to assist in calibrating acupoint positioning; the AI ​​algorithm chip has a built-in pre-trained convolutional neural network acupoint recognition model, which is used to fuse infrared thermal imaging data and pressure distribution data to automatically identify and locate the core acupoints of Xiyan and Zusanli within the knee joint.

3. The system according to claim 1, characterized in that, In the three-head independent PID temperature-controlled moxibustion head group, the three moxibustion heads are arranged in a triangular shape, corresponding to the knee eye and bilateral Zusanli acupoints, respectively. The PID control subunit of each moxibustion head adopts an incremental PID algorithm, forming a closed-loop control with the corresponding temperature sensor and far-infrared electric heating unit. It collects the skin temperature of the moxibustion area in real time, automatically adjusts the heating power, and controls the temperature range of 38-55℃ with a temperature control accuracy of ±0.3℃.

4. The system according to claim 1, characterized in that, In the aforementioned drug-thermal synergistic smokeless moxibustion module, the far-infrared electric heating plate can radiate far-infrared rays in the 4-14μm wavelength band, which can penetrate 3-5mm into the subcutaneous tissue to produce a deep warming effect; the replaceable mugwort-coated sheet compartment contains a coated sheet containing extracts of mugwort active ingredients, which adheres to the moxibustion surface of the far-infrared electric heating plate and releases the pharmacologically active ingredients of mugwort volatile oil after being heated, thereby achieving a drug-thermal synergistic effect.

5. The system according to claim 1, characterized in that, The device's main control and communication module includes a main control unit, a BLE low-power Bluetooth communication unit, a power management unit, and a Type-C multi-function interface. The main control unit is electrically connected to the AI ​​algorithm chip and each PID control subunit, and is used to execute the moxibustion plan issued by the user-side APP. The Bluetooth communication unit is used to realize bidirectional data transmission between the device and the APP. The Type-C multi-function interface is used for wired power supply and data transmission of the device.

6. The system according to claim 1, characterized in that, The TCM syndrome self-testing unit of the user-end APP control system has a built-in TCM syndrome database for knee osteoarthritis, covering five core syndromes: cold-dampness obstruction, damp-heat obstruction, qi stagnation and blood stasis, liver and kidney deficiency, and qi and blood deficiency. It is equipped with a standardized symptom self-assessment questionnaire, which can automatically complete the TCM syndrome determination based on the questionnaire data filled in by the user.

7. The system according to claim 1, characterized in that, The personalized solution matching unit has a built-in database of moxibustion solutions corresponding to each syndrome type, including acupoint combinations, moxibustion temperature, moxibustion duration, and moxibustion mode parameters. After completing the syndrome type determination, it automatically matches the corresponding personalized moxibustion solution for the user and sends it to the moxibustion device host.

8. The system according to claim 1, characterized in that, The device control and monitoring unit is used to receive acupoint location data, moxibustion temperature data, and device operating status data uploaded by the moxibustion device host in real time. It can adjust the moxibustion parameters in real time, monitor the device operating status, and automatically trigger shutdown protection when the temperature exceeds the threshold.

9. The system according to claim 1, characterized in that, The rehabilitation record management unit is used to store the user's syndrome data and all moxibustion data, and automatically generate weekly / monthly rehabilitation reports to realize full-cycle data traceability and management of the user's rehabilitation process.

10. A method for controlling the operation of a wearable moxibustion device based on AI vision and multi-channel PID temperature control using the system described in any one of claims 1 to 9, characterized in that, Includes the following steps: S1, Syndrome Self-Assessment and Treatment Plan Matching: Users complete a self-assessment questionnaire for knee osteoarthritis symptoms through the user-end APP. The system automatically completes the TCM syndrome determination and matches the corresponding personalized moxibustion plan, including target acupoints, moxibustion temperature for each acupoint, and moxibustion duration. S2, Device Wearing and Start-up: The user fixes the moxibustion device main unit to the knee joint area through a wearable fixing carrier, starts the communication connection between the device and the APP, and the APP sends the matched moxibustion plan to the device main unit; S3, Fully Automatic Acupoint Positioning: The device host starts the sensor integration module, the infrared thermal imaging module collects infrared thermal imaging data of the knee joint surface in real time, the pressure sensor array collects contact pressure distribution data, the AI ​​algorithm chip integrates dual-source data, automatically identifies and locates the target acupoint based on the thermal sensitivity characteristics of the acupoint, and adjusts the position of the moxibustion head to align with the target acupoint. S4, Multi-channel independent PID temperature control moxibustion: The main unit of the device starts the three independent PID temperature control moxibustion head group. Each moxibustion head independently starts the PID temperature control program according to the temperature and duration set in the program. The temperature sensor collects the skin temperature in real time, and the PID control subunit adjusts the heating power of the far-infrared electric heating unit in real time to keep the moxibustion temperature stable within the set threshold. At the same time, the far-infrared electric heating plate heats the mugwort coating sheet, releasing the herbal active ingredients of mugwort, realizing smokeless moxibustion with synergistic effect of medicine and heat. S5, Data Synchronization and Rehabilitation Record Generation: During the moxibustion process, the device synchronizes the moxibustion data and operating status to the user's APP in real time; after the moxibustion is completed, the APP automatically stores all the data of this moxibustion, updates the user's rehabilitation record, generates a rehabilitation report, and completes the closed loop of the whole process.

Images