Pain treatment method based on infrared photothermal coupling

Abstract

The invention discloses a pain treatment method based on infrared photothermal coupling, and belongs to the technical field of medical engineering. Comprising the following steps: acquiring physiological feature data, pathological feature data and optical feature data of a patient, and acquiring initial equipment parameters of the intelligent infrared photothermal treatment device; the method comprises the following steps: constructing an individualized photo-thermal response prediction model through Monte Carlo photon transmission simulation in combination with finite element heat conduction analysis, predicting a therapeutic effect score and an optimal parameter combination by using a deep neural network model based on a Transform architecture, generating an individualized temperature-time therapeutic curve, and determining initial illumination parameter configuration; controlling the intelligent infrared photothermal treatment device to start a treatment process according to the initial illumination parameter configuration; and according to the comprehensive biomarker score and in combination with a real-time analysis result of the neural network processing module, the output power, the pulse duty ratio and the irradiation mode of the infrared light source are automatically adjusted through the infrared light source control module.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical engineering, more particularly, to a pain treatment method based on infrared light-heat coupling. BACKGROUND

[0002] In clinical pain treatment, infrared light-heat therapy is a common non-drug method, which is widely used in the intervention of various chronic and acute pain due to its simple operation and less side effects. This method mainly acts on the skin and deep tissues through the thermal effect of infrared light, promotes local blood circulation, relieves muscle tension, and thus relieves pain symptoms. However, existing infrared light-heat therapy mostly uses fixed parameters or relies on manual adjustment of irradiation intensity and time, which is difficult to take into account the individual differences of different patients, the treatment process lacks pertinence, and the phenomena of insufficient heating or local overheating are likely to occur, affecting the actual efficacy and use safety.

[0003] In recent years, with the development of physiological monitoring equipment and data processing technology, the ability to obtain key physiological information such as skin temperature, blood flow, and muscle response has gradually been achieved, which also provides a new idea for optimizing the infrared treatment process. If the light exposure parameters can be adjusted in real time in combination with the specific physical characteristics, pain types and tissue response of the patient, it will help to improve the treatment effect. However, there is currently a lack of an infrared light-heat treatment method and system that can integrate multiple biological characteristic information, dynamically adjust the treatment process, and have recording and optimization capabilities.

[0004] In summary, how to realize the individualization and dynamization of parameter regulation in the infrared light-heat treatment process, improve the treatment effect while ensuring the safety of treatment, has become a technical problem that needs to be solved. SUMMARY

[0005] In order to overcome a series of defects existing in the prior art, the purpose of the present application is to provide a pain treatment method based on infrared light-heat coupling, which is applied to an intelligent infrared light-heat treatment device, the intelligent infrared light-heat treatment device includes an infrared light source control module, a biological signal acquisition module, a temperature monitoring module, a neural network processing module and a data management module, and the method includes the following steps:

[0006] Obtain physiological characteristic data, pathological characteristic data and optical characteristic data of the patient, and obtain initial device parameters of the intelligent infrared light-heat treatment device;

[0007] According to the patient characteristic data, an individualized light-heat response prediction model is constructed by combining Monte Carlo photon transport simulation with finite element heat conduction analysis, and a deep neural network model based on the Transformer architecture is used to predict the treatment effect score and the optimal parameter combination, to generate a personalized temperature-time treatment curve and determine the initial light exposure parameter configuration;

[0008] controlling the intelligent infrared photothermal treatment device to start a treatment process according to the initial light exposure parameters;

[0009] During the treatment process, the current treatment area is marked as a target treatment area, the skin temperature distribution, the blood perfusion index, the muscle tension, and the skin tissue conductivity are synchronously collected by using the biological signal acquisition module, and a comprehensive biological marker scoring system is constructed;

[0010] According to the comprehensive biological marker score and the real-time analysis result of the neural network processing module, the output power, the pulse duty cycle, and the irradiation mode of the infrared light source are automatically adjusted by the infrared light source control module;

[0011] During the automatic adjustment process, the temperature change of the target treatment area is monitored in real time by the temperature monitoring module, real-time biological feedback data is continuously collected by the biological signal acquisition module, and whether the preset pain relief standard is reached and kept within the safe temperature range is judged by combining the temperature change and the biological feedback data;

[0012] If the preset pain relief standard is not reached or the temperature exceeds the safe range, the optimal parameters are recalculated by the neural network processing module, and the treatment parameters are adjusted until the pain relief standard is reached and the temperature is safe;

[0013] If the preset pain relief standard is reached and the temperature is kept safe, the complete treatment parameter trajectory, biological feedback data, and efficacy evaluation result are automatically recorded by using the data management module, and the model parameters of the neural network processing module are continuously optimized based on the post-treatment patient feedback and follow-up data for incremental learning.

[0014] Further, the physiological characteristic data of the patient includes the following steps: first, the reflectance spectrum data of the patient's skin surface is collected, which covers 64 characteristic wavelength points in the wavelength range of 700-1400 nanometers; second, the blood flow velocity distribution of the treatment area and the surrounding skin tissue is measured, and the corresponding blood vessel distribution two-dimensional mapping image is obtained; then, the conductivity parameters of the skin and subcutaneous skin tissue are measured, including the skin tissue resistance value, capacitance value, and impedance phase angle; then, the basic body surface temperature distribution atlas of the patient in a resting state is obtained, and the uniformity index of the temperature distribution is calculated; in addition, the heart rate variability index, respiratory frequency mode, and skin conductance response parameters of the patient are collected, and an individualized physiological baseline database is established.

[0015] Further, the pathological feature data acquisition process includes: quantitative evaluation of skin tissue hardness in the target treatment area, specifically including measuring the Young's modulus value and constructing the spatial distribution map of the elasticity parameter; detecting the concentration changes of oxygenated hemoglobin and deoxyhemoglobin in the skin tissue, calculating the skin tissue oxygen saturation and blood volume parameters; obtaining the thickness information of each layer of skin structure, including the accurate measurement values of the epidermis layer, the dermis layer and the subcutaneous fat layer; determining the spatial coordinates of the pain sensitive area and the pain threshold distribution; combining the scores of the patient pain assessment scale and the pain nature description, establishing a multi-dimensional pain feature vector; obtaining the dielectric property parameters of the skin tissue at different depths.

[0016] Further, the optical feature data acquisition process includes:

[0017] Accurately determining the diffuse reflectance and specular reflectance of the patient's skin in the wavelength range of 780-1400 nanometers;

[0018] Measuring the microvascular perfusion level and related hemodynamic indicators to reflect the local blood circulation state;

[0019] Obtaining the two-dimensional optical parameter distribution map of the skin tissue, including the scattering coefficient, absorption coefficient and anisotropy parameter;

[0020] Determining the average penetration depth and attenuation coefficient of photons in the skin tissue, quantifying the photon transport characteristics;

[0021] Evaluating the metabolic state and cell activity parameters of the skin tissue to judge the local physiological response ability;

[0022] Predicting the propagation path, energy deposition distribution and heat generation efficiency of different wavelength infrared light in the patient's skin tissue structure.

[0023] Further, the Monte Carlo photon transport simulation process includes:

[0024] Establishing a three-layer skin tissue optical model including the epidermis layer, the dermis layer and the subcutaneous fat layer;

[0025] Assigning corresponding optical parameters to each layer of skin tissue, including the refractive index, scattering coefficient, absorption coefficient, anisotropy parameter and layer thickness;

[0026] Using the Henyey-Greenstein phase function to describe the angular distribution characteristics of photon scattering;

[0027] Releasing not less than 10 5 photon packets in the simulation process, tracking the propagation path, scattering events and absorption process of each photon packet in the skin tissue;

[0028] The energy deposition distribution and local heating efficiency of photon packets at different skin tissue depths are calculated;

[0029] Considering the influence of skin tissue heterogeneity on light transmission, a random model of skin tissue density distribution is established to simulate the influence of light transmission;

[0030] Output two-dimensional spatial light energy density distribution map and depth profile light intensity decay curve.

[0031] Further, the finite element heat conduction analysis process includes:

[0032] Considering heat conduction, blood perfusion, metabolic heat production and external heat source terms comprehensively, a mathematical model of skin tissue temperature field distribution is established;

[0033] A two-dimensional finite element model containing multi-layer skin structure is established, and the grid of the treatment area is refined to 0.5 millimeter;

[0034] Assign appropriate thermal physical parameters to each skin tissue layer, including thermal conductivity, specific heat capacity, density, blood perfusion rate and metabolic heat production rate;

[0035] Set the boundary conditions, including the convective heat transfer coefficient and radiation heat transfer coefficient of the skin surface and the environment;

[0036] The light energy deposition distribution is input as a volumetric heat source, and the forward Euler time integration scheme is used to solve the transient heat conduction equation;

[0037] Calculate the temperature spatiotemporal distribution of the treatment area and the surrounding skin tissue, and predict the temperature rise rate and maximum temperature value;

[0038] Analyze the cumulative effect of thermal damage and the safety temperature threshold to provide safety guarantee and effect prediction for treatment parameter setting.

[0039] Further, the deep neural network model based on the Transformer architecture includes the following components and implementation methods:

[0040] The multi-head self-attention mechanism is used to process the multi-modal feature data of the patient, and the number of attention heads is set to 8, and the dimension of each head is 64;

[0041] A network architecture containing 6 layers of Transformer encoder is constructed, each layer containing a multi-head self-attention sublayer and a feedforward neural network sublayer;

[0042] The position encoding is used to process the time series biological signal data, supporting sequence input of up to 512 time steps;

[0043] Residual connection and layer normalization are used to ensure the training stability of the deep network;

[0044] The output layer is designed to predict the degree of pain relief, the length of treatment, and the optimal light power parameters.

[0045] The training is based on clinical treatment data, and the training data set contains no less than 10,000 treatment records.

[0046] Further, the personalized temperature-time treatment curve generation process includes the following steps:

[0047] According to the type and severity of the patient's pain, the target treatment temperature range is determined, acute pain is set to 42-45 degrees Celsius, and chronic pain is set to 40-43 degrees Celsius;

[0048] Calculate the upper limit of the safe treatment time and evaluate the cumulative heat damage;

[0049] A multi-stage temperature control strategy is designed, including a rapid heating stage, a constant temperature maintenance stage, and a slow cooling stage, wherein: the temperature rise rate in the heating stage is controlled at 2-4 degrees Celsius per minute to avoid heat shock response; the temperature fluctuation range in the constant temperature stage is controlled within ±0.5 degrees Celsius, and the maintenance time is adjusted between 5-20 minutes according to the type of pain;

[0050] Multiple temperature monitoring points are set to achieve accurate control of the spatial temperature distribution in the treatment area;

[0051] Based on real-time temperature deviation, the light power output is automatically adjusted, and individual difference factors including age, gender, skin type, and pain sensitivity are combined to generate a personalized treatment curve.

[0052] Further, the output power of the infrared light source is automatically adjusted, including the following steps:

[0053] An initial power adjustment instruction is generated through a power regulation system based on a proportional-integral-derivative control mechanism;

[0054] Based on a rate limiting mechanism, the initial power adjustment instruction is limited in rise rate and fall rate to obtain a speed-limited adjustment instruction;

[0055] Through a multi-stage power control strategy, the speed-limited instruction is coarsely and finely adjusted by 5% and 1% steps respectively to obtain a refined adjustment instruction;

[0056] Combined with the current temperature change trend and the target temperature, the required power adjustment amount is predicted to form a prediction adjustment instruction;

[0057] Based on a preset ±0.2℃ dead zone range, it is judged whether to execute the prediction instruction to obtain the final execution instruction;

[0058] Meanwhile, the power limit safety mechanism is used to monitor the real-time power anomaly, and when overload or abnormal risk is detected, the power is automatically reduced to a safety threshold or the light source is turned off, and a corresponding safety protection instruction is generated.

[0059] Further, the irradiation modes include four basic modes of continuous irradiation, pulsed irradiation, scanning irradiation and focused irradiation, wherein: in the continuous irradiation mode, the constant power output is maintained, which is suitable for the treatment scene requiring stable heating; in the pulsed irradiation mode, the pulse frequency is adjustable in the range of 0.1-10 Hz, and the pulse width is adjustable in the range of 0.1-5 seconds; in the scanning irradiation mode, the light spot moves in the treatment area according to the preset trajectory, and the scanning speed is 1-10 mm / s to ensure the uniformity of heating; in the focused irradiation mode, the light energy is concentrated in the specific area most sensitive to pain, and the light spot diameter is adjustable in the range of 2-20 mm.

[0060] Compared with the prior art, the present application has the following beneficial effects:

[0061] The present application collects physiological, pathological and optical characteristic data of the patient through multi-modal acquisition, constructs an individualized photothermal response model by combining Monte Carlo photon transport simulation and finite element heat conduction analysis, uses a deep neural network based on Transformer to predict and dynamically optimize treatment parameters in real time, realizes adaptive adjustment of infrared light source power and irradiation mode, ensures maximum treatment effect and precise temperature control within the safe temperature range, and dynamically improves the intelligent level of treatment individualization and efficacy evaluation through continuous biological feedback acquisition and neural network incremental learning. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 A flowchart of a pain treatment method based on infrared light-heat coupling disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme of the embodiments of the present application will be described in more detail below with reference to the drawings in the embodiments of the present application. In the drawings, the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The described embodiments are part of the embodiments of the present application, not all embodiments.

[0064] Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0065] The embodiments described below with reference to the drawings and the directional words are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.

[0066] As shown in Figure 1 An infrared light heat coupling-based pain treatment method applied to an intelligent infrared light heat treatment device, the intelligent infrared light heat treatment device comprising an infrared light source control module, a biological signal acquisition module, a temperature monitoring module, a neural network processing module and a data management module, the method comprising the following steps:

[0067] Obtaining physiological feature data, pathological feature data and optical feature data of a patient, and obtaining initial device parameters of the intelligent infrared light heat treatment device;

[0068] According to the patient feature data, an individualized light heat response prediction model is constructed by combining Monte Carlo photon transport simulation with finite element heat conduction analysis, and a deep neural network model based on a Transformer architecture is used to predict treatment effect scores and optimal parameter combinations to generate a personalized temperature-time treatment curve and determine initial light exposure parameter configurations;

[0069] Controlling the intelligent infrared light heat treatment device to start the treatment process according to the initial light exposure parameter configurations;

[0070] During the treatment process, the current treatment area is marked as a target treatment area, the biological signal acquisition module is used to synchronously acquire skin temperature distribution, blood perfusion index, muscle tension and skin tissue conductivity, and a comprehensive biomarker scoring system is constructed;

[0071] According to the comprehensive biomarker scores and the real-time analysis results of the neural network processing module, the output power, pulse duty cycle and irradiation mode of the infrared light source are automatically adjusted by the infrared light source control module;

[0072] During the automatic adjustment process, the temperature change of the target treatment area is monitored in real time by the temperature monitoring module, real-time biological feedback data is continuously acquired by the biological signal acquisition module, and it is determined whether the preset pain relief standard is reached and kept within a safe temperature range by combining the temperature change and the biological feedback data;

[0073] If the preset pain relief standard is not reached or the temperature exceeds the safe range, the optimal parameters are recalculated by the neural network processing module and the treatment parameters are adjusted until the pain relief standard is reached and the temperature is safe;

[0074] If the preset pain relief standard is reached and the temperature is kept safe, the data management module is used to automatically record the complete treatment parameter trajectory, biological feedback data and efficacy evaluation results, and the model parameters of the neural network processing module are continuously optimized based on post-treatment patient feedback and follow-up data for incremental learning.

[0075] In summary, the pain treatment method based on infrared light heat coupling forms a complete closed-loop system covering "patient feature analysis-treatment parameter setting-real-time adaptive control-therapeutic effect evaluation and learning optimization" through the collaborative application of multi-modal data acquisition, personalized modeling, real-time feedback control and intelligent learning mechanism. Its core advantage lies in the integration of physical simulation and neural network prediction, realizing individualized precision treatment; through biological feedback and multi-dimensional scoring system to strengthen the dynamic nature and safety of process control; and relying on the continuous learning mechanism driven by data, realizing the continuous evolution of system performance and the expansion of patient adaptability.

[0076] Further, the acquisition of physiological feature data of the patient includes the following steps: first, collecting the reflectance spectrum data of the patient's skin surface, which covers 64 characteristic wavelength points in the wavelength range of 700-1400 nanometers; second, measuring the blood flow velocity distribution of the treatment area and the surrounding skin tissue to obtain the corresponding two-dimensional vascular distribution map; then, measuring the conductivity parameters of the skin and subcutaneous skin tissue, including skin tissue resistance value, capacitance value and impedance phase angle; then, obtaining the basic body surface temperature distribution atlas of the patient in a resting state and calculating the uniformity index of the temperature distribution; in addition, collecting the heart rate variability index, respiratory frequency pattern and skin conductance response parameters of the patient, and then establishing an individualized physiological baseline database.

[0077] In summary, through multi-dimensional and refined physiological feature data acquisition means, a comprehensive and individualized physiological baseline database is constructed, realizing the accurate quantification and modeling of the current physiological state of the patient. Specifically, by acquiring the skin reflectance spectrum data in the range of 700-1400 nanometers, it is helpful to identify the optical absorption and scattering characteristics of the patient's skin tissue, providing a basis for subsequent simulation of infrared light energy transmission path; blood flow velocity distribution and two-dimensional vascular mapping can reveal the local circulation state and heat conduction ability, assisting in judging the dynamic response area of infrared energy distribution; the determination of conductivity parameters reflects the skin electrophysiological state and water content, etc., which are closely related to the heat sensitivity; the basic body surface temperature atlas and its uniformity index are used to evaluate the thermal environment distribution characteristics in the resting state, facilitating the establishment of individualized temperature response model; while heart rate variability, respiratory pattern and skin electrical response data reveal the overall stress level and physiological fluctuation range of the patient from the perspective of autonomic nervous system regulation.

[0078] Further, the pathological feature data acquisition process includes: quantitative evaluation of skin tissue hardness on the target treatment area, specifically including measuring Young's modulus value and constructing the spatial distribution map of elastic parameters; detecting the concentration changes of oxygenated hemoglobin and deoxygenated hemoglobin in the skin tissue, calculating the oxygen saturation and blood volume parameters of the skin tissue; obtaining the thickness information of each layer structure of the skin, including the accurate measurement values of the epidermis layer, the dermis layer and the subcutaneous fat layer; determining the spatial coordinates of the pain sensitive area and the pain threshold distribution; combining the scores of the patient pain assessment scale and the description of the pain nature, establishing a multi-dimensional pain feature vector; obtaining the dielectric property parameters of the skin tissue at different depths.

[0079] In summary, through systematic and multi-parameter pathological feature data acquisition, the overall characterization of the tissue state and pain characteristics of the target treatment area is realized, significantly enhancing the accuracy of individualized pain diagnosis and treatment parameter setting. Specifically, the quantitative evaluation of skin tissue hardness (such as Young's modulus and its spatial distribution) reflects the pathological state of tissue inflammation, fibrosis or muscle tension, which helps to judge the response elasticity of the tissue to infrared heat stimulation; the concentration changes of oxygenated and deoxygenated hemoglobin and the calculated oxygen saturation and blood volume parameters provide a dynamic monitoring basis for the metabolic level and local microcirculation state of the tissue, which can guide the energy deposition strategy in the treatment area; the accurate measurement of the thickness of each layer of the skin provides key geometric parameters for infrared penetration depth and energy deposition calculation; the spatial calibration of the pain sensitive area and its threshold distribution realizes the precise correspondence between the treatment key parts and the heat distribution key; the multi-dimensional pain feature vector not only integrates subjective evaluation (such as scale scores) and objective physiological parameters, but also can be used to train neural network models to identify different pain types; in addition, the dielectric property parameters of the skin tissue at different depths provide the basis for modeling the infrared heat-electric coupling mechanism.

[0080] Further, the optical feature data acquisition process includes:

[0081] Accurately determining the diffuse reflectance and specular reflectance of the patient's skin in the wavelength range of 780-1400 nanometers;

[0082] Measuring microvascular perfusion level and related hemodynamic indicators to reflect the local blood circulation state;

[0083] Obtaining the two-dimensional optical parameter distribution map of the skin tissue, including scattering coefficient, absorption coefficient and anisotropy parameter;

[0084] Determining the average penetration depth and attenuation coefficient of photons in the skin tissue, quantifying the photon transport characteristics;

[0085] Evaluating the metabolic state and cell activity parameters of the skin tissue to judge the local physiological response ability;

[0086] Predict the propagation path, energy deposition distribution and heat generation efficiency of different wavelengths of infrared light in the patient's skin tissue structure.

[0087] In summary, by comprehensively obtaining the key optical properties of skin tissue, a fine modeling and quantitative prediction of the propagation behavior of infrared light in the patient's body are achieved, significantly improving the accuracy and individual adaptation ability of energy regulation in infrared light thermal therapy. Specifically, the accurate determination of the diffuse reflectance and specular reflectance in the 780-1400 nm wavelength band helps to evaluate the initial response characteristics of the epidermis to incident light, guiding the optimization of light source power and angle; the microvascular perfusion level and its hemodynamic indicators reveal the activity level of subcutaneous blood circulation, directly affecting the local heat conduction and metabolic clearance ability, and are of great significance for heat damage risk assessment. The scattering coefficient, absorption coefficient and anisotropy parameter contained in the two-dimensional optical parameter distribution map can accurately depict the propagation path and energy deposition mode of light between different tissue layers, providing a basis for personalized irradiation depth control and tissue selective heating; the determination of the average penetration depth and photon attenuation coefficient further complements the quantitative evaluation of the spatial range of photothermal response, which helps to build a patient-specific thermal field simulation model. The assessment of the metabolic state and cell activity of skin tissue provides a reference for judging the self-repairing ability of the tissue after infrared light thermal stimulation, and predicts the tolerance and regeneration potential of the tissue after treatment; finally, by predicting the propagation path, energy deposition distribution and heat generation efficiency of different wavelengths of infrared light, the system can complete the energy strategy screening under multiple wavelengths and multiple parameters before treatment, thereby realizing the "tailor-made" infrared light thermal treatment plan. Overall, the optical characteristic data acquisition process plays a core supporting role in realizing the precision control of infrared irradiation, heat effect prediction and individualization of treatment effect, ensuring the efficiency, safety and targeting of the treatment.

[0088] Further, the initial device parameter acquisition process of the intelligent infrared light thermal therapy device includes:

[0089] Calibrate the spectral power density of the infrared light source and measure its light power output value per nanometer interval in the 780-1400 nm wavelength range;

[0090] Detect the spatial uniformity, divergence angle and spot shape parameters of the light beam;

[0091] Calibrate the actual light power value and stability index under different output power levels;

[0092] Measure the working temperature distribution and heat dissipation efficiency parameters of the light source;

[0093] Detect the response time characteristics of the light source, including start-up delay, shutdown delay and power modulation frequency response;

[0094] Measure the polarization characteristics and coherence parameters of the light source;

[0095] Evaluate the drift characteristics and service life prediction of the light source output power;

[0096] At the same time, record the influence coefficient of environmental temperature and humidity on equipment performance, and establish equipment performance compensation and standardized operation parameter database.

[0097] In summary, by systematically calibrating and measuring key parameters such as spectral power density, spatial uniformity, divergence angle and spot shape, actual output and stability under each power level, working temperature distribution and heat dissipation efficiency, opening / closing delay and power modulation response, polarization and coherence, as well as output drift characteristics and life prediction of infrared light source, and combining with the influence coefficient of environmental temperature and humidity, the performance compensation and standardized operation database is established, the high precision of equipment energy output, real-time performance monitoring and long-term stability guarantee are realized, and the safety, effectiveness and repeatability of the treatment process are significantly improved.

[0098] Further, the Monte Carlo photon transport simulation process includes:

[0099] Establish a three-layer skin tissue optical model including epidermis, dermis and subcutaneous fat layer;

[0100] Assign corresponding optical parameters to each layer of skin tissue, including refractive index, scattering coefficient, absorption coefficient, anisotropy parameter and layer thickness;

[0101] Use Henyey-Greenstein phase function to describe the angle distribution characteristics of photon scattering;

[0102] Release not less than 10 5 photon packets in the simulation process, track the propagation path, scattering events and absorption process of each photon packet in the skin tissue;

[0103] Calculate the energy deposition distribution and local heating efficiency of photon packets at different skin tissue depths;

[0104] Consider the influence of skin tissue inhomogeneity on light transmission, establish a random model of skin tissue density distribution to simulate the influence of light transmission;

[0105] Output two-dimensional spatial light energy density distribution map and depth profile light intensity decay curve.

[0106] In summary, by constructing a multi-layer skin optical model containing epidermis, dermis and subcutaneous fat layers, assigning accurate optical parameters to each layer, and modeling scattering behavior with the Henyey-Greenstein phase function, combined with high-order photon packet tracing simulation, not only can the light energy deposition and local heating efficiency at different tissue depths be accurately calculated, but also the influence of tissue heterogeneity on light transmission is fully considered, thereby realizing high-fidelity simulation of light propagation in skin tissue, improving the accuracy of individualized light-thermal response prediction and the scientificity of treatment parameter setting.

[0107] Further, the finite element heat conduction analysis process includes:

[0108] Considering heat conduction, blood perfusion, metabolic heat production and external heat source terms, a mathematical model of skin tissue temperature field distribution is established;

[0109] A two-dimensional finite element model containing multi-layer skin structure is established, and the mesh of the treatment area is refined to 0.5 millimeters;

[0110] Assign appropriate thermal physical parameters to each skin tissue layer, including thermal conductivity, specific heat capacity, density, blood perfusion rate and metabolic heat production rate;

[0111] Set boundary conditions, including convective heat transfer coefficient and radiation heat transfer coefficient between skin surface and environment;

[0112] Input light energy deposition distribution as volumetric heat source, and use forward Euler time integration scheme to solve transient heat conduction equation;

[0113] Calculate the spatio-temporal distribution of temperature in the treatment area and surrounding skin tissue, and predict the temperature rise rate and maximum temperature value;

[0114] Analyze the cumulative effect of thermal damage and the safety temperature threshold to provide safety guarantee and effect prediction for treatment parameter setting.

[0115] In summary, by constructing a multi-layer skin tissue temperature field model considering the effects of heat conduction, blood perfusion, metabolic heat production and external heat source, combined with refined grid two-dimensional finite element modeling, accurately inputting light energy deposition distribution as volumetric heat source, and solving the transient heat conduction equation, the dynamic simulation and accurate prediction of temperature changes in the treatment area and surrounding tissue can be realized, thereby quantifying the temperature rise rate, peak temperature and cumulative effect of thermal damage, providing scientific basis for treatment parameter optimization, and significantly improving the safety and accuracy of infrared photothermal therapy process.

[0116] Further, the deep neural network model based on the Transformer architecture includes the following components and implementation methods:

[0117] The multi-head self-attention mechanism is used to process the multi-modal feature data of the patient, and the number of attention heads is set to 8, and the dimension of each head is 64;

[0118] A network architecture including 6 layers of Transformer encoders is constructed, each layer including a multi-head self-attention sublayer and a feedforward neural network sublayer;

[0119] The position encoding is used to process the time-series biological signal data, and a sequence input of up to 512 time steps is supported;

[0120] Residual connection and layer normalization are used to ensure the stability of the training of the deep network;

[0121] An output layer is designed to predict the pain relief degree, treatment duration and optimal light power parameters;

[0122] Based on the clinical treatment data, the training data set includes no less than 10,000 treatment records.

[0123] In summary, by constructing a deep neural network model based on the Transformer architecture, combining the multi-head self-attention mechanism to extract and correlate the patient's multi-modal features, and enhancing the processing capability of time-series biological signals through position encoding, the pain relief degree, treatment duration and optimal light power parameters can be accurately predicted while maintaining the stability of the training. With the help of large-scale clinical treatment data training, this model significantly improves the individualized parameter regulation ability and treatment effect prediction accuracy in photothermal therapy.

[0124] Further, the personalized temperature-time treatment curve generation process includes the following steps:

[0125] According to the pain type and severity of the patient, the target treatment temperature range is determined, acute pain is set to 42-45 degrees Celsius, and chronic pain is set to 40-43 degrees Celsius;

[0126] The upper limit of the safe treatment time is calculated and the cumulative heat damage is evaluated;

[0127] A multi-stage temperature control strategy is designed, including a rapid heating stage, a constant temperature maintenance stage and a slow cooling stage, wherein: the temperature rise rate in the heating stage is controlled at 2-4 degrees Celsius per minute to avoid heat shock response; the temperature fluctuation range in the constant temperature stage is controlled within ±0.5 degrees Celsius, and the maintenance time is adjusted between 5-20 minutes according to the pain type;

[0128] Multiple temperature monitoring points are set to achieve accurate control of the spatial temperature distribution in the treatment area;

[0129] Based on real-time temperature deviation, the light power output is automatically adjusted, and combined with individual difference factors including age, gender, skin type and pain sensitivity, a personalized treatment curve is generated.

[0130] In summary, by combining the patient's pain type and severity to set the target treatment temperature range, and on this basis, designing a dynamic temperature control strategy including three stages of heating, constant temperature and cooling, while introducing multi-point temperature monitoring and real-time power regulation mechanism, the personalized temperature-time treatment curve generation process can realize fine control of temperature change rate, constant temperature stability and heat damage risk, and integrate individual physiological differences to generate the most suitable heat treatment path for patient characteristics, thereby improving the safety, effectiveness and individual precision of treatment.

[0131] Further, the initial light parameter configuration determination process includes:

[0132] Calculate the required surface light power density of the treatment area;

[0133] Consider the penetration depth and moisture absorption characteristics of skin tissue to select the optimal infrared wavelength;

[0134] Design the shape and size of the light spot according to the size of the treatment area;

[0135] Determine the pulse mode parameters, including pulse width, pulse interval and duty cycle;

[0136] Calculate the optimal distance between the light source and the skin surface to ensure uniformity of light energy distribution.

[0137] In summary, by calculating the required surface light power density of the treatment area, optimizing the selection of infrared wavelength based on the penetration depth and moisture absorption characteristics of skin tissue, and determining the shape and size of the light spot according to the size of the treatment area, further configuring the pulse mode parameters and adjusting the optimal distance between the light source and the skin, the initial light parameter configuration process can realize effective transmission and uniform distribution of light energy in the tissue, ensuring the precision and safety of energy input during the treatment process, and providing a high-quality initial setting basis for the subsequent personalized photothermal treatment plan.

[0138] Further, the blood perfusion index acquisition includes:

[0139] Detect the microcirculation blood flow changes within 1-2 mm below the skin to obtain raw blood flow data from multiple measurement points;

[0140] Calculate key hemodynamic parameters, including mean blood flow velocity, pulsatility index and blood flow resistance coefficient;

[0141] Combine the changes in oxygenated and deoxygenated hemoglobin concentration, vascular volume changes and heart rate variability to comprehensively generate the blood perfusion index;

[0142] The blood perfusion index is quantified as 0-100 points based on a standardized scoring system, and the normal reference interval is set as 50-80 points;

[0143] The dynamic changes in perfusion state during treatment are predicted, and the change trend of blood perfusion index before and after treatment is recorded.

[0144] In summary, through multi-point monitoring of subcutaneous microcirculation blood flow, key hemodynamic parameters such as average blood flow velocity, pulsatility index and resistance coefficient are extracted, combined with multi-dimensional physiological information such as hemoglobin concentration, blood vessel volume and heart rate variability, the blood perfusion index acquisition process realizes the comprehensive evaluation and standardized quantification of the perfusion state of local tissue; dynamically tracking the perfusion trend during treatment helps to judge the regulation effect of photothermal stimulation on microcirculation in real time, thereby improving the safety and accuracy of treatment feedback.

[0145] Further, the comprehensive biomarker scoring system comprises the following steps:

[0146] A multi-dimensional scoring matrix covering three dimensions of physiological indicators, physical indicators and subjective feelings is established to obtain preliminary scoring data of the comprehensive biomarker;

[0147] Based on the principal component analysis method, the scoring matrix is reduced in dimension to reduce the redundancy between indicators and extract key feature parameters;

[0148] Combined with the individual characteristics and pain types of the patient, the weights of each indicator are adaptively adjusted to form a weighted comprehensive score, wherein: in the physiological indicator dimension, heart rate variability, blood pressure change, respiratory rate and skin electrical response parameters are integrated, and the weights are set as 30%, 25%, 20% and 25% respectively; in the physical indicator dimension, based on the temperature distribution, blood perfusion, muscle tension and skin tissue conductivity change, the physical score is obtained by applying the standardized scoring algorithm; in the subjective feeling dimension, visual analog pain score, simplified pain questionnaire and comfort scale are used for evaluation; finally, the scoring update frequency is set as every 30 to 60 seconds to realize real-time monitoring and dynamic adjustment of the treatment process.

[0149] In summary, by constructing a multi-indicator scoring system covering three dimensions of physiology, physics and subjective feeling, combined with principal component analysis for feature extraction and dimension reduction, and based on the individual characteristics and pain types of the patient, the comprehensive biomarker scoring system can accurately quantify the comprehensive physiological response state of the individual during treatment, and dynamically update the score at a frequency of 30 to 60 seconds to support real-time optimization and individualized regulation of treatment strategies, significantly improving the feedback loop efficiency of photothermal therapy and the objectivity of efficacy evaluation.

[0150] Further, the output power automatic adjustment of the infrared light source comprises the following steps:

[0151] generating initial power adjustment instructions through a power adjustment system based on a proportional-integral-derivative control mechanism;

[0152] limiting the initial power adjustment instructions in terms of rising rate and falling rate based on a rate limiting mechanism to obtain speed-limited adjustment instructions;

[0153] coarsely adjusting and finely adjusting the speed-limited instructions by 5% and 1% steps respectively through a multi-stage power control strategy to obtain refined adjustment instructions;

[0154] combining the current temperature change trend and the target temperature to predict the required power adjustment amount and form prediction adjustment instructions;

[0155] judging whether to execute the prediction instructions based on a preset ±0.2℃ dead zone range to obtain final execution instructions;

[0156] Meanwhile, monitoring real-time power abnormality based on a power limit safety mechanism, and automatically reducing the power to a safety threshold or shutting down the light source when overload or abnormal risk is detected to generate corresponding safety protection instructions.

[0157] In summary, through the introduction of a proportional-integral-derivative (PID) control mechanism to generate initial adjustment instructions, and in combination with rate limiting, multi-stage adjustment (coarse adjustment and fine adjustment), temperature trend prediction, and dead zone judgment strategies, the output power automatic adjustment process of the infrared light source realizes fine and stable control of power output; meanwhile, supplemented by power abnormality monitoring and safety protection mechanisms, the system can automatically respond when overload or abnormal risk occurs, adjusting the power to a safety threshold or shutting down the light source, thereby effectively improving the temperature control accuracy and safety during infrared treatment.

[0158] Further, the pulse duty cycle adjustment comprises:

[0159] calculating optimal duty cycle parameters based on the deviation between the actual temperature and the target temperature to form preliminary duty cycle settings;

[0160] generating pulse signals matching the preliminary settings with an adjustable pulse period of 1-10 seconds and a real-time calculated pulse width to realize duty cycle adjustment;

[0161] predicting future temperature trends based on the current duty cycle and temperature change trend, and dynamically adjusting the duty cycle to maintain temperature stability accordingly;

[0162] considering the delay effect of thermal inertia of skin tissue on temperature response to further optimize the response efficiency and control effect of duty cycle adjustment;

[0163] Adopting multi-region independent control strategy, according to the temperature demand of different positions in the treatment area, differential duty cycle adjustment is implemented to obtain local precise control parameters;

[0164] The duty cycle change rate limit is set to ensure that the single adjustment amplitude does not exceed 5%, so as to avoid the temperature fluctuation caused by the excessive adjustment amplitude;

[0165] Combined with historical treatment data and efficacy feedback, the duty cycle adjustment model is continuously optimized to form an adaptive control strategy with learning ability;

[0166] Through the abnormality detection mechanism, whether the duty cycle deviates from the set range is monitored in real time, and if an abnormality occurs, an alarm is automatically sent and the corresponding safety protection measures are started.

[0167] In summary, by combining the real-time temperature deviation to calculate the optimal duty cycle parameter, and generating dynamic matching pulse signals with adjustable pulse period, the pulse duty cycle adjustment process realizes high responsiveness adjustment of temperature control; At the same time, the temperature change trend prediction, thermal inertia compensation, multi-region independent regulation and change rate limitation strategy not only enhance the local precision and stability of the control, but also effectively suppress the temperature fluctuation risk; Further, the historical data optimization and abnormality detection mechanism are introduced to build an adaptive control framework with learning ability, thereby significantly improving the intelligentization, safety and individualized regulation level of infrared light heat treatment.

[0168] Further, the irradiation mode includes four basic modes of continuous irradiation, pulse irradiation, scanning irradiation and focused irradiation, wherein: in the continuous irradiation mode, the constant power output is maintained, which is suitable for treatment scenarios requiring stable heating; In the pulse irradiation mode, the pulse frequency is adjustable in the range of 0.1-10 Hz, and the pulse width is adjustable in the range of 0.1-5 seconds; In the scanning irradiation mode, the light spot moves in the treatment area according to the preset trajectory, and the scanning speed is 1-10 mm / s to ensure the uniformity of heating; In the focused irradiation mode, the light energy is concentrated in the specific area most sensitive to pain, and the light spot diameter is adjustable in the range of 2-20 mm.

[0169] In summary, the irradiation mode covers four basic modes of continuous, pulse, scanning and focusing, and the continuous irradiation meets the stable heating demand through constant power output; The pulse irradiation supports flexible adjustment of 0.1-10 Hz frequency and 0.1-5 seconds width to realize dynamic thermal stimulation; The scanning irradiation moves the light spot according to the preset trajectory at a speed of 1-10 mm / s to ensure the uniformity of heating in the treatment area; The focused irradiation adjusts the light spot diameter (2-20 mm) for the pain sensitive part to realize precise energy concentration and improve the treatment specificity and effect.

[0170] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the same. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An infrared light-heat coupling-based pain treatment method applied to an intelligent infrared light-heat treatment device, the intelligent infrared light-heat treatment device comprising an infrared light source control module, a biological signal acquisition module, a temperature monitoring module, a neural network processing module and a data management module, characterized in that, The method comprises the following steps: acquiring physiological characteristic data, pathological characteristic data and optical characteristic data of a patient, and acquiring initial device parameters of the intelligent infrared photothermal treatment device; constructing an individualized photothermal response prediction model by combining Monte Carlo photon transport simulation with finite element heat conduction analysis according to the patient characteristic data, while predicting treatment effect scores and optimal parameter combinations by using a deep neural network model based on a Transformer architecture, generating a personalized temperature-time treatment curve and determining initial light exposure parameter configurations; controlling the intelligent infrared photothermal treatment device to start a treatment process according to the initial light exposure parameter configurations; In the treatment process, the current treatment area is marked as a target treatment area, and the skin temperature distribution, blood perfusion index, muscle tension and skin tissue conductivity are synchronously collected by using the biological signal acquisition module, and a comprehensive biomarker scoring system is constructed; According to the comprehensive biomarker score and the real-time analysis result of the neural network processing module, the output power, pulse duty cycle and irradiation mode of the infrared light source are automatically adjusted by the infrared light source control module; In the automatic adjustment process, the temperature change of the target treatment area is monitored in real time by the temperature monitoring module, and real-time biological feedback data is continuously collected by the biological signal acquisition module, and whether the preset pain relief standard is reached and kept within the safe temperature range is judged by combining the temperature change and the biological feedback data; If the preset pain relief standard is not reached or the temperature exceeds the safe range, the optimal parameters are recalculated by the neural network processing module and the treatment parameters are adjusted until the pain relief standard is reached and the temperature is safe; If the preset pain relief standard is reached and the temperature is kept safe, the data management module is used to automatically record the complete treatment parameter trajectory, biological feedback data and efficacy evaluation results, and based on the post-treatment patient feedback and follow-up data, incremental learning is carried out to continuously optimize the model parameters of the neural network processing module.

2. The method for pain treatment based on infrared light-heat coupling according to claim 1, characterized in that, The acquisition of the physiological characteristic data of the patient comprises the following steps: first, collecting the reflectance spectrum data of the patient's skin surface, which covers 64 characteristic wavelength points in the wavelength range of 700-1400 nanometers; second, measuring the blood flow velocity distribution of the treatment area and the surrounding skin tissue, and obtaining the corresponding vascular distribution two-dimensional mapping image; then, determining the conductivity parameters of the skin and subcutaneous skin tissue, including skin tissue resistance value, capacitance value and impedance phase angle; then, obtaining the basic body surface temperature distribution atlas of the patient in a resting state, and calculating the uniformity index of the temperature distribution; in addition, collecting the heart rate variability index, respiratory frequency pattern and skin conductance response parameters of the patient, and establishing an individualized physiological baseline database.

3. A method of pain treatment based on infrared light-heat coupling according to claim 2, characterized in that, The acquisition process of the pathological feature data includes: quantitative evaluation of skin tissue hardness in the target treatment area, specifically including measuring the Young's modulus value and constructing the spatial distribution map of the elastic parameter; detecting the concentration change of oxygenated hemoglobin and deoxyhemoglobin in the skin tissue, calculating the oxygen saturation and blood volume parameters of the skin tissue; obtaining the thickness information of each layer of the skin structure, including the accurate measurement values of the epidermis layer, the dermis layer and the subcutaneous fat layer; determining the spatial coordinates of the pain sensitive area and the pain threshold distribution; combining the score of the patient's pain assessment scale and the description of the pain nature, establishing a multi-dimensional pain feature vector; and obtaining the dielectric property parameters of the skin tissue at different depths.

4. The method for pain treatment based on infrared light-heat coupling according to claim 3, characterized in that, The optical feature data acquisition process includes: accurately determining the diffuse reflectance and specular reflectance of the patient's skin in the wavelength range of 780-1400 nanometers; measuring the microvascular perfusion level and related hemodynamic indicators to reflect the local blood circulation state; obtaining the two-dimensional optical parameter distribution map of the skin tissue, including the scattering coefficient, absorption coefficient and anisotropy parameter; determining the average penetration depth and attenuation coefficient of photons in the skin tissue, quantifying the photon transport characteristics; evaluating the metabolic state and cell activity parameters of the skin tissue to judge the local physiological response ability; predicting the propagation path, energy deposition distribution and heat generation efficiency of different wavelength infrared light in the patient's skin tissue structure.

5. The method of claim 1, wherein the infrared light is applied to the skin of the patient for a period of time sufficient to cause a temperature increase in the skin of the patient of at least 1.5°C. The Monte Carlo photon transport simulation process includes: establishing a three-layer skin tissue optical model including the epidermis layer, the dermis layer and the subcutaneous fat layer; assigning corresponding optical parameters to each layer of skin tissue, including refractive index, scattering coefficient, absorption coefficient, anisotropy parameter and layer thickness; using the Henyey-Greenstein phase function to describe the angular distribution characteristics of photon scattering; Release no less than 10 5 Photon packets in the simulation process, tracking each photon packet in the skin tissue propagation path, scattering events and absorption process; calculating the energy deposition distribution and local heating efficiency of photon packets at different depths of the skin tissue; considering the influence of skin tissue heterogeneity on light transport, establishing a random model of skin tissue density distribution to simulate the influence of light transport; outputting the two-dimensional spatial light energy density distribution map and the depth profile light intensity decay curve.

6. The method for pain treatment based on infrared light-heat coupling according to claim 5, characterized in that, The finite element heat conduction analysis process includes: comprehensively considering heat conduction, blood perfusion, metabolic heat production and external heat source terms to establish a mathematical model of the temperature field distribution of the skin tissue; establishing a two-dimensional finite element model including multiple layers of skin structure, and refining the grid of the treatment area to 0.5 millimeters; assigning corresponding thermal physical parameters to each layer of skin tissue, including thermal conductivity, specific heat capacity, density, blood perfusion rate and metabolic heat generation rate; setting boundary conditions, including the convective heat transfer coefficient and radiative heat transfer coefficient between the skin surface and the environment; inputting the light energy deposition distribution as a volumetric heat source, and using the forward Euler time integration scheme to solve the transient heat conduction equation; calculating the temperature and space distribution of the treatment area and the surrounding skin tissue, predicting the temperature rise rate and the maximum temperature value; analyzing the cumulative effect of thermal damage and the safety temperature threshold to provide safety guarantee and effect prediction for the treatment parameter setting.

7. The method of claim 1, wherein the infrared light is applied to the skin of the patient for a period of time ranging from 1 to 30 minutes. The deep neural network model based on the Transformer architecture includes the following components and implementation methods: The multi-head self-attention mechanism is used to process the multi-modal feature data of the patient, and the number of attention heads is set to 8, and the dimension of each head is 64; A network architecture including 6 layers of Transformer encoders is constructed, each layer including a multi-head self-attention sublayer and a feedforward neural network sublayer; The position encoding is used to process the time-series biological signal data, and the sequence input of up to 512 time steps is supported; Residual connection and layer normalization are used to ensure the stability of the deep network training; An output layer is designed to predict the pain relief degree, treatment duration and optimal light power parameters; Based on the clinical treatment data, the training data set contains no less than 10,000 treatment records.

8. The method of claim 1, wherein the infrared light is applied to the skin of the patient for a period of time sufficient to cause a temperature increase in the skin of the patient of at least 1.5°C. The personalized temperature-time treatment curve generation process includes the following steps: According to the pain type and severity of the patient, the target treatment temperature range is determined, acute pain is set to 42-45℃, and chronic pain is set to 40-43℃; The upper limit of the safe treatment time is calculated and the cumulative heat damage is evaluated; A multi-stage temperature control strategy is designed, including a rapid heating stage, a constant temperature maintenance stage and a slow cooling stage, wherein: the temperature rise rate in the heating stage is controlled at 2-4℃ per minute to avoid heat shock reaction; the temperature fluctuation range in the constant temperature stage is controlled within ±0.5℃, and the maintenance time is adjusted between 5-20 minutes according to the pain type; Multiple temperature monitoring points are set to realize accurate control of the spatial temperature distribution in the treatment area; Based on the real-time temperature deviation, the light power output is automatically adjusted, and individual difference factors including age, gender, skin type and pain sensitivity are combined to generate a personalized treatment curve.

9. The method of claim 1, wherein the infrared light is applied to the skin of the patient for a period of time ranging from about 5 minutes to about 30 minutes. The output power automatic adjustment of the infrared light source includes the following steps: An initial power adjustment instruction is generated by a power regulation system based on a proportional-integral-derivative control mechanism; The initial power adjustment instruction is subjected to rate limiting mechanism to limit the rise rate and the fall rate, and the limited speed adjustment instruction is obtained; Through a multi-stage power control strategy, the limited speed instruction is coarsely adjusted and finely adjusted by 5% and 1% steps respectively to obtain a refined adjustment instruction; Combined with the current temperature change trend and the target temperature, the required power adjustment amount is predicted to form a predicted adjustment instruction; Based on the preset ±0.2℃ dead zone range, it is judged whether to execute the predicted instruction, so as to obtain the final execution instruction; At the same time, based on the power limit safety mechanism, real-time monitoring power abnormality is monitored, and when overload or abnormal risk is detected, the power is automatically reduced to a safety threshold or the light source is turned off, and the corresponding safety protection instruction is generated.

10. The method of claim 1, wherein the infrared light is applied to the skin of the patient for a period of time sufficient to cause a temperature increase in the skin of the patient of at least 1.5°C. The irradiation mode includes four basic modes of continuous irradiation, pulsed irradiation, scanning irradiation and focused irradiation, wherein: in the continuous irradiation mode, the constant power output is maintained, which is suitable for the treatment scene requiring stable heating; in the pulsed irradiation mode, the pulse frequency is adjustable in the range of 0.1-10 Hz, and the pulse width is adjustable in the range of 0.1-5 seconds; in the scanning irradiation mode, the light spot moves in the preset track in the treatment area, and the scanning speed is 1-10 mm / s to ensure the uniformity of heating; in the focused irradiation mode, the light energy is concentrated in the specific area most sensitive to pain, and the light spot diameter is adjustable in the range of 2-20 mm.

Images