A control method and system of a laser millimeter wave therapeutic instrument for scar treatment

Abstract

The application discloses a scar treatment laser millimeter wave therapeutic instrument control method and system, and relates to the technical field of medical instrument control. The method comprises the following steps: acquiring a scar morphological feature; outputting a posterior probability distribution of laser and millimeter wave joint control parameters based on a multi-target Gaussian process regression model, quantifying parameter uncertainty and coupling relationship; combining Chebyshev inequality and parameter distribution skewness, dynamically generating a safety adjustment factor and constructing an adaptive safety feasible region to replace traditional static empirical threshold; solving optimal joint control parameters under the constraint of the safety feasible region; and controlling laser and millimeter wave to cooperatively perform scar repair treatment according to the parameters. The method solves the problems of traditional scar therapeutic instrument parameter adjustment, such as conservatism, fixed safety boundary, lack of uncertainty quantification and disconnection between laser and millimeter wave control, significantly improves treatment effect and safety, and is suitable for precise non-invasive repair of various types of hypertrophic, atrophic and pigmented scars.

Description

Technical Field

[0001] This invention relates to the field of instrument control technology, and in particular to a control method and system for a laser millimeter wave therapy device for scar treatment. Background Technology

[0002] A scar treatment device is a medical instrument primarily used to improve or remove skin scars. It typically combines technologies such as lasers and millimeter waves to selectively heat or stimulate scar tissue, promoting skin repair and remodeling collagen structure, thereby reducing the color, raisedness, or thickness of scars. The device allows adjustment of laser power, wavelength, irradiation duration, and spot size to suit different types and sizes of scars.

[0003] The control of scar treatment devices directly affects treatment effectiveness and safety. Properly controlled parameters maximize scar repair while avoiding thermal damage or irritation to surrounding healthy skin, thus reducing treatment risks. Furthermore, scar morphology is complex and varied; different patients have different skin thicknesses, scar protrusion, and textures. Precise control of the treatment device parameters enables personalized treatment, making the therapeutic effect more stable and predictable, while improving patient comfort and satisfaction. This is an indispensable key aspect of modern scar treatment.

[0004] However, the existing scar treatment devices lack quantification of the uncertainty of the control parameters and the coupling effect between the control parameters, making it difficult to reliably adjust the control parameters. Furthermore, the safety boundary setting relies on static experience thresholds, resulting in conservative parameter adjustment or insufficient risk control. In severe cases, this may cause secondary harm to the patient, making it difficult to achieve safe and efficient control of the scar treatment device. Summary of the Invention

[0005] To address the technical problems of existing scar treatment devices, such as the lack of quantification of the uncertainty of control parameters and the coupling effect between control parameters, which makes it difficult to reliably adjust the control parameters and the reliance on static empirical thresholds for safety boundary settings, leading to conservative parameter adjustments or insufficient risk control, which may even cause secondary harm to patients, this invention provides a control method and system for a laser millimeter-wave therapy device for scar treatment.

[0006] The technical solutions provided by the embodiments of the present invention are as follows: First aspect This invention provides a method for controlling a laser millimeter-wave therapy device for scar treatment, comprising: S1: Acquire images of the scar; S2: Extract scar differential features from scar images; S3: Input the scar difference features into the multi-objective Gaussian process regression model, combine the treatment response index, and output the posterior conditional distribution of the treatment device control parameters, including the posterior mean vector and the posterior covariance matrix of the treatment device control parameters. S4: Combining the posterior mean vector and the posterior covariance matrix of the control parameters of the therapeutic instrument, a safety adjustment factor for the control parameters of the therapeutic instrument is generated by Chebyshev's inequality. S5: Establish a safe and feasible domain for the control parameters of the therapeutic device regarding safety adjustment factors; S6: Under the constraint of the safe and feasible region of the control parameters of the therapeutic instrument, determine the optimal control parameters of the therapeutic instrument based on the posterior condition distribution of the control parameters of the therapeutic instrument; S7: Control the therapeutic device according to the optimal therapeutic device control parameters.

[0007] Second aspect An embodiment of the present invention provides a control system for a laser millimeter-wave therapy device for scar treatment, comprising: processor; The memory stores computer-readable instructions, which, when executed by a processor, implement the control method for the scar treatment laser millimeter-wave therapy device as described in the first aspect.

[0008] Third aspect The present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the scar treatment laser millimeter wave therapy device control method as described in the first aspect.

[0009] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this embodiment of the invention, a probabilistic mapping relationship between scar difference characteristics and treatment device control parameters is established through a multi-objective Gaussian process regression model. The posterior probability distribution of the treatment device control parameters is output, with its mean vector indicating the recommended value of the current optimal parameters, and the covariance matrix quantifying the uncertainty of parameter prediction and the mutual influence, i.e., coupling effect, between the parameters. Next, using Chebyshev's inequality, combined with posterior mean and covariance information, a safety adjustment factor is dynamically generated. This factor is used to construct a feasible region of safe parameters that adaptively changes with individual circumstances and prediction uncertainty, replacing the fixed static safety threshold. Under the constraint of this probabilistically guaranteed safety boundary, the optimal control parameters that balance treatment effectiveness and safety are solved from the posterior probability distribution. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic flowchart illustrating a control method for a laser millimeter-wave therapy device for scar treatment provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the control system of a laser millimeter-wave therapy device for scar treatment provided in an embodiment of the present invention. Detailed Implementation

[0012] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0013] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0014] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0015] Reference manual attached Figure 1 The diagram shows a flowchart of a control method for a laser millimeter-wave therapy device for scar treatment provided in an embodiment of the present invention.

[0016] This invention provides a control method for a laser millimeter-wave therapy device for scar treatment. This method can be implemented by a control device for the laser millimeter-wave therapy device, which can be a terminal or a server. The processing flow of the control method for the laser millimeter-wave therapy device for scar treatment may include the following steps: S1: Acquire images of the scar.

[0017] Scar images refer to visual images of a patient's scar area obtained through imaging equipment.

[0018] S2: Extract scar differential features from scar images.

[0019] Among them, scar difference features refer to a set of features extracted from scar images to quantify differences in scar morphology, size, and structure.

[0020] In one possible implementation, scar difference characteristics include scar area, scar aspect ratio, scar elevation, and scar orientation. S2 specifically includes: Scar differential features are extracted using a pre-trained convolutional neural network model.

[0021] The pre-trained convolutional neural network model specifically comprises an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a third convolutional layer, a third pooling layer, a fully connected layer, and an output layer, connected in sequence. During training, the model first collects a large number of labeled scar image samples, annotating each image with its area, aspect ratio, convexity, and orientation. These samples are then input into the convolutional neural network. By minimizing the mean squared error between the predicted features and the true annotations, the parameters of the convolutional kernels and fully connected layers are optimized, enabling the model to automatically extract key morphological features from the scar images, providing accurate input for subsequent treatment parameter prediction.

[0022] S3: Input the scar difference features into the multi-objective Gaussian process regression model, combine the treatment response index, and output the posterior conditional distribution of the treatment device control parameters, including the posterior mean vector and the posterior covariance matrix of the treatment device control parameters.

[0023] The multi-objective Gaussian process regression model is a machine learning method used to establish a probabilistic mapping relationship between input features (here, scar difference features) and multiple output variables (here, treatment device control parameters). Treatment response indicators are standardized numerical values ​​used to assess treatment effectiveness and risk. The posterior mean vector represents the initial recommended values ​​for the current optimal scar control parameters, and the posterior covariance matrix quantifies the relationship between prediction uncertainty and parameter linkage. The multi-objective Gaussian process regression model specifically includes an input layer, a kernel function layer (a composite kernel function in this scheme), and an output layer. During training, the model first collects a large number of sample scar images and corresponding treatment parameters and effectiveness / risk indicators. Then, it extracts scar difference features, combines these features with labeled data to form a training set, and optimizes the hyperparameters of the composite kernel function to enable the multi-objective Gaussian process regression model to accurately fit the relationship between input features and treatment control parameters. Simultaneously, it learns the covariance structure between parameters, achieving optimal parameter prediction and uncertainty quantification for unknown scars.

[0024] In one possible implementation, the treatment response indicators include normalized values ​​of the treatment effect score and normalized values ​​of the tissue damage risk rate. The treatment device control parameters include the laser power, millimeter-wave frequency, treatment duration, and spot size.

[0025] Among them, the normalized value of the treatment effect score is a subjective standardized score that quantifies the repair effect after scar treatment. The normalized value of the tissue damage risk rate is a subjective score of the probability of irreversible thermal damage to normal skin under the current combination of control parameters of the treatment device.

[0026] In one possible implementation, S3 specifically includes: S301: Obtain multiple sample scar images and their corresponding label data.

[0027] S302: Extract the sample scar difference features from each sample scar image. The sample scar difference features and the corresponding label data constitute the training data. The label data includes the control parameters of the treatment device and the treatment response index.

[0028] S303: With the goal of the normalized value of the treatment effect score in the treatment response index being greater than the preset normalized value of the treatment effect score, a multi-objective Gaussian process regression model is trained using training data in combination with a composite kernel function.

[0029] It should be noted that those skilled in the art can set the magnitude of the preset treatment effect score normalization value according to actual needs, and this invention does not limit it.

[0030] The formula for a composite kernel function is as follows: .

[0031] .

[0032] in, Indicates the first i Individual sample scar differences With the j Individual sample scar differences Similarity between them Represents the natural exponential function. Represents a trainable adaptive length size that describes the sensitivity to differential features of scars. Indicates the first i Individual sample scar differences With the j Individual sample scar differences The Euclidean distance between them e Represents the natural constant. Represents the cosine function. Represents pi (π). Indicates the range of values ​​as Trainable parameters used to capture texture features, index This indicates transpose.

[0033] The trainable adaptive length dimension is a non-zero parameter, whose initial value can be directly set to 1, and its value ranges from 0 to 1. Specifically, this composite kernel function, by simultaneously combining the exponential square term, the smoothing term, and the cosine texture term, can simultaneously capture the global similarity, local smoothing changes, and subtle texture patterns of scar difference features. This allows the multi-objective Gaussian process regression model to consider both the overall trend and local features and texture information when predicting treatment parameters, thereby improving prediction accuracy and personalized adaptability.

[0034] S304: Input the scar difference features into the trained multi-objective Gaussian process regression model, and output the posterior mean vector of the treatment device control parameters and the posterior covariance matrix of the initial treatment device control parameters.

[0035] The formula for the posterior mean vector of the control parameters of the therapeutic instrument is as follows: .

[0036] .

[0037] in, Indicating the differences in scar characteristics The corresponding variables include different control parameters for different therapeutic instruments. The posterior mean vector, The first term obtained from the training data i Individual sample scar differences With the j Individual sample scar differences Similarity kernel matrix between, subscripts Indicates transpose. Indicates avoidance Irreversible perturbation noise variance Represents the identity matrix. This represents the matrix of control parameters for the therapeutic device in the training data. Indicating the differences in scar characteristics Similarity vector to the scar difference features of each sample in all training data.

[0038] The formula for the posterior covariance matrix of the initial treatment device control parameters is as follows: .

[0039] in, Indicating the differences in scar characteristics The initial posterior covariance matrix of the control parameters of the corresponding therapeutic instrument between any two variables x. Indicating the differences in scar characteristics The self-similarity.

[0040] Specifically, this method generates posterior mean and covariance matrices by kernel mapping new scar difference features with the similarity of training data and combining them with existing treatment parameter information. This provides optimal parameter recommendations and quantifies prediction uncertainty and the correlation between parameters, thereby enabling personalized and reliable treatment parameter decisions.

[0041] S305: Correct the posterior covariance matrix of the initial control parameters of the treatment device by the treatment response index, and obtain the posterior covariance matrix of the control parameters of the treatment device.

[0042] The specific formula for the posterior conditional distribution of the control parameters of the therapeutic instrument is as follows: .

[0043] .

[0044] in, Indicating the differences in scar characteristics The posterior covariance matrix of the control parameters of the corresponding treatment device, between any two variables x. This indicates finding the partial derivative. Represents a symbolic function. and These represent the normalized values ​​of the treatment effect score and the normalized values ​​of the tissue damage risk rate, respectively. Represents the weight parameters. This indicates taking the minimum value. Indicates the first m Line 1 n Column correction parameters, and These represent the first parameter in the control parameters of the therapeutic instrument. m The first variable and the second n One variable.

[0045] It should be noted that this approach modifies the initial covariance matrix by combining treatment efficacy and tissue damage risk, incorporating treatment safety and efficacy information into the parameter uncertainty quantification. This allows the model to consider the coupling relationship between parameters and dynamically reflect the balance between efficacy and risk when predicting optimal control parameters, thereby improving the safety and reliability of treatment decisions.

[0046] Optionally, the weight parameter can be set to 0.3.

[0047] Specifically, this process begins by collecting a large number of sample scar images along with corresponding treatment parameters and efficacy indicators to construct a training dataset. Then, scar difference features are extracted from each image, reflecting information such as scar area, shape, degree of protrusion, and orientation. These features are combined with treatment response indicators, and a multi-objective Gaussian process regression model is trained using a composite kernel function. This allows the model to learn the probabilistic mapping relationship between scar features and optimal treatment parameters, while quantifying the coupling relationship and uncertainty between parameters. For new scar images, the model outputs the posterior mean vector of treatment parameters as initial recommendations based on their features, and the posterior covariance matrix reflects the uncertainty of the prediction. Furthermore, the covariance matrix is ​​corrected by combining treatment efficacy and risk indicators, ensuring that the final prediction balances maximizing efficacy and safety, thus improving treatment accuracy and safety.

[0048] S4: Combining the posterior mean vector and the posterior covariance matrix of the control parameters of the therapeutic instrument, a safety adjustment factor for the control parameters of the therapeutic instrument is generated using Chebyshev's inequality.

[0049] Chebyshev's inequality, a probability inequality, provides an upper bound on the probability of a random variable deviating from its mean. In this scheme, it quantifies the maximum range by which the control parameters of the treatment device may deviate from the posterior mean vector, thus providing a safety constraint on parameter uncertainty. The safety adjustment factor, a coefficient calculated using Chebyshev's inequality, is used to scale the posterior covariance matrix, establishing a feasible region for safe parameters that adaptively varies with individual scar differences and prediction uncertainty. This factor ensures that, under high probability (e.g., 95%), the treatment parameters will not exceed the safe range, reducing the risk of skin damage. By adaptively generating the safety adjustment factor using Chebyshev's inequality, dynamic constraints on treatment parameter uncertainty can be achieved, improving treatment safety and personalized adaptability.

[0050] In one possible implementation, S4 specifically includes: S401: Calculate the skewness of each control parameter of the therapeutic instrument under the posterior mean vector and the posterior covariance matrix of the control parameters of the therapeutic instrument.

[0051] The formula for calculating skewness is as follows: .

[0052] in, Indicates skewness, Expressing expectations, This represents the control parameter vector of the four-dimensional therapeutic instrument. These represent laser power, millimeter-wave frequency, treatment duration, and spot size, respectively. Represents a four-dimensional vector of all 1s. This represents the matrix trace operation.

[0053] It should be noted that skewness can quantify the asymmetry of the distribution of control parameters of the therapeutic instrument, so that when the model calculates the safety adjustment factor, it not only considers the degree of dispersion of the parameters, but also reflects the directionality of deviation from the mean, thereby more accurately constraining the parameters within the safe range and improving the reliability and personalized adaptability of the treatment.

[0054] S402: Substitute the skewness and the posterior covariance matrix of the control parameters of the treatment device into the Chebyshev inequality to calculate the upper bound of the probability of the control parameters of the treatment device relative to the posterior mean vector of the control parameters of the treatment device under different total deviation thresholds.

[0055] It should be noted that those skilled in the art can set the total deviation threshold according to actual needs, and this invention does not limit it.

[0056] The formula for calculating the upper bound of probability is as follows: .

[0057] in, This indicates that the total deviation of the four-dimensional therapeutic instrument's control parameter vector A from the posterior mean vector of the control parameters exceeds any arbitrary total deviation threshold. t The upper bound of the probability.

[0058] It should be noted that this upper bound of probability can quantify the risk of treatment parameters deviating from the recommended mean, providing a clear probabilistic constraint on uncertainty. Thus, when formulating the safe and feasible domain, both the overall fluctuation of parameters and the skewness reflecting the asymmetry of the distribution are considered, so that treatment decisions remain safe and reliable under high probability.

[0059] S403: Replace the total deviation threshold with a factor term relating to the posterior covariance matrix of the safety adjustment factor and the control parameters of the treatment device.

[0060] It is understandable that by binding the total deviation threshold and the total uncertainty together to obtain the factor term, we can obtain the adjustment coefficient of the safe range by scaling the total uncertainty, which is the safety adjustment factor.

[0061] S404: Set the upper bound of the probability to be less than the preset upper bound of the probability, and then use the factor terms to deduce the safety adjustment factor.

[0062] It should be noted that those skilled in the art can set the size of the preset probability upper bound according to actual needs, and this invention does not limit it.

[0063] The formula for the safety adjustment factor is as follows: .

[0064] in, Indicates skewness Safety adjustment factor below, This indicates the upper bound of the preset probability.

[0065] It should be noted that those skilled in the art can set the size of the preset probability upper bound according to actual needs, and this invention does not limit this. Optionally, it can be set to 0.05, that is, at least 95% probability to ensure that the control parameters of the treatment device fall within the safe range. The safety adjustment factor dynamically scales the covariance matrix by combining the probability upper bound with the skewness and uncertainty of the parameter distribution, thereby constraining the treatment parameters to fall within the safe range with a high probability, realizing adaptive control of uncertainty and individual differences, and improving the safety and reliability of treatment.

[0066] The derivation process of this safety adjustment factor is as follows: Factors Substituting these values ​​into the formula for calculating the upper bound of probability, we can eliminate the parameters t and y. This yields a quadratic equation in one variable. Solving this equation yields the formula for calculating the safety adjustment factor. Replacing the factor term with the preset probability upper bound signifies binding the total deviation threshold to the total uncertainty, effectively scaling the total uncertainty to obtain the safety adjustment factor within the safe range. This process applies Chebyshev's inequality, combined with the predicted posterior covariance matrix of the treatment device's control parameters, to set an upper bound constraint on the deviation probability for the uncertainty of the treatment device's control parameters, defining a safety adjustment factor related to skewness to ensure that parameter deviations remain within the safe range.

[0067] Specifically, this method first calculates the skewness of each parameter based on the posterior mean vector and covariance matrix of the control parameters of the treatment device obtained through training, to measure the asymmetry of the parameter distribution. Then, the skewness and covariance information are substituted into Chebyshev's inequality to calculate the upper bound of the probability of the parameter deviating from the mean. By binding the total deviation threshold to the covariance matrix, a safety adjustment factor is derived. This factor is used to scale the covariance matrix, thereby generating a safe and feasible region that adaptively changes with individual scar differences and prediction uncertainty. This method utilizes statistical probability constraints and parameter distribution characteristics to dynamically control uncertainty, keeping treatment parameters safe within a high probability range. It can automatically prevent parameters from exceeding the safety boundary while ensuring efficacy, dynamically adapting to different scar characteristics, and improving the safety and reliability of personalized treatment.

[0068] S5: Establish a safe and feasible domain for the control parameters of the therapeutic device regarding safety adjustment factors.

[0069] It should be noted that by using safety adjustment factors to construct a safe and feasible domain for the control parameters of the treatment device, a clear parameter boundary is provided for each treatment. This ensures that the therapeutic effect can be optimized and skin damage can be prevented under uncertain conditions, thereby improving the reliability and safety of the treatment.

[0070] In one possible implementation, S5 specifically includes: Based on the safety adjustment factor, the posterior mean vector of the control parameters of the therapeutic instrument, and the posterior covariance matrix of the control parameters of the therapeutic instrument, a safe and feasible region for the control parameters of the therapeutic instrument is established.

[0071] The formula for the safe and feasible region of the control parameters of the therapeutic instrument is as follows: .

[0072] in, This indicates the safe and feasible range of control parameters for the therapeutic device.

[0073] The safe and feasible domain of the control parameters of the treatment device defines the safe boundary of the parameters under high probability by combining the posterior mean, covariance matrix and safety adjustment factor, so that the treatment parameters can maintain the efficacy and avoid exceeding the safety limit, thereby achieving a reliable and controllable treatment plan under the conditions of individual differences and prediction uncertainty.

[0074] S6: Under the constraint of the safe and feasible region of the control parameters of the therapeutic instrument, determine the optimal control parameters of the therapeutic instrument based on the posterior condition distribution of the control parameters of the therapeutic instrument.

[0075] It should be noted that selecting the optimal treatment parameters under the constraints of the safe and feasible domain ensures that the final parameters not only meet the safety limits but also make full use of the predicted posterior information, thereby achieving a balance between maximizing treatment effectiveness and minimizing risk, and improving the accuracy and reliability of personalized treatment.

[0076] In one possible implementation, S6 specifically includes: S601: Determine whether all the control parameters in the posterior mean vector of the control parameters of the treatment device are within the safe and feasible region of the control parameters of the treatment device. If so, update the posterior mean vector of the control parameters of the treatment device to the optimal control parameters of the treatment device and proceed to step S603. Otherwise, proceed to step S602.

[0077] S602: Correct the control parameters of the therapeutic instrument that exceed the safe and feasible range of the control parameters, obtain the corrected control parameters of the therapeutic instrument, and update the corrected control parameters of the therapeutic instrument to the optimal control parameters of the therapeutic instrument.

[0078] The revised formula is as follows: .

[0079] in, This represents the corrected control parameter vector for the therapeutic device. This indicates that the control parameters of the therapeutic device are within the safe and feasible range. The vector of control parameters for the therapeutic device that minimizes the function.

[0080] S603: Outputs optimal control parameters for the therapeutic device.

[0081] It should be noted that this process first determines whether the posterior mean of the predicted treatment device control parameters falls entirely within the safe and feasible region. If the condition is met, these parameters are directly used as the optimal treatment parameters. If some parameters exceed the safe region, the excess portion is projected into the safe boundary to obtain the corrected optimal parameters. This process combines the posterior mean and safety constraints, considering both efficacy and risk. Through optimization, it ensures that the final parameters are as close as possible to the optimal prediction without violating safety limits. It can dynamically adjust parameters to ensure treatment is conducted within a safe range while maximizing efficacy and personalized adaptability.

[0082] S7: Control the therapeutic device according to the optimal therapeutic device control parameters.

[0083] In one possible implementation, it also includes: The multi-objective Gaussian process regression model and the pre-trained convolutional neural network model are updated at preset intervals.

[0084] It should be noted that those skilled in the art can set the preset duration according to actual needs, and this invention does not limit this.

[0085] Specifically, during the update process, based on the newly collected data, the multi-objective Gaussian process regression model and the pre-trained convolutional neural network model are retrained at preset time intervals. By adjusting the model parameters using the latest observation data, their predictive and feature extraction capabilities are kept consistent with the actual environment, thereby improving the overall prediction accuracy and response performance.

[0086] In practical applications, this laser millimeter-wave therapy method for scar treatment first acquires images of the patient's scar using an imaging device. A pre-trained convolutional neural network is then used to extract differential scar features, quantifying information such as scar morphology, size, prominence, and orientation. These features are then input into a multi-objective Gaussian process regression model, combined with treatment efficacy and tissue damage risk indicators, to predict the posterior mean and covariance matrix of the therapy device's control parameters, while simultaneously quantifying the uncertainty and interrelationships of the parameters. Subsequently, by calculating skewness and applying Chebyshev's inequality, a safety adjustment factor is generated to construct a safe and feasible region that adaptively changes with individual scar differences and prediction uncertainty. Under the constraints of the safe and feasible region, the system determines whether the posterior mean is within the safe range; if it exceeds this range, corrections are made to obtain the final optimal therapy device control parameters, ensuring maximum therapeutic effect while preventing skin damage. The entire process allows for dynamic model updates to adapt to scar changes. By combining probabilistic statistics and machine learning methods to quantify parameter uncertainty and impose safety constraints, it provides personalized, precise, and safe therapy device control, effectively improving scar repair efficacy and treatment reliability.

[0087] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this embodiment of the invention, a probabilistic mapping relationship between scar difference characteristics and treatment device control parameters is established through a multi-objective Gaussian process regression model. The posterior probability distribution of the treatment device control parameters is output, with its mean vector indicating the recommended value of the current optimal parameters, and the covariance matrix quantifying the uncertainty of parameter prediction and the mutual influence, i.e., coupling effect, between the parameters. Next, using Chebyshev's inequality, combined with posterior mean and covariance information, a safety adjustment factor is dynamically generated. This factor is used to construct a feasible region of safe parameters that adaptively changes with individual circumstances and prediction uncertainty, replacing the fixed static safety threshold. Under the constraint of this probabilistically guaranteed safety boundary, the optimal control parameters that balance treatment effectiveness and safety are solved from the posterior probability distribution.

[0088] Reference manual attached Figure 2 The diagram shows a structural schematic of a control system for a laser millimeter-wave therapy device for scar treatment provided by the present invention.

[0089] The present invention also provides a control system 20 for a scar treatment laser millimeter wave therapy device, applied to the above-mentioned scar treatment laser millimeter wave therapy device control method, comprising: Processor 201.

[0090] The memory 202 stores computer-readable instructions, which, when executed by the processor 201, implement the scar treatment laser millimeter wave therapy device control method as described in the method embodiment.

[0091] The scar treatment laser millimeter wave therapy instrument control system 20 provided by the present invention can execute the above-mentioned scar treatment laser millimeter wave therapy instrument control method and achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate further.

[0092] It should be understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0093] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0094] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0095] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0096] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0097] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0098] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0099] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0100] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0101] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0103] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0104] This invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the scar treatment laser millimeter wave therapy device control method as described in the method embodiment.

[0105] The present invention provides a computer-readable storage medium that can implement the steps and effects of the scar treatment laser millimeter wave therapy device control method of the above method embodiments. To avoid repetition, the present invention will not repeat them.

[0106] 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 variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0107] The following points need to be explained: (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0108] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the invention, i.e., these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element or there may be intermediate elements.

[0109] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0110] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A control method for a laser millimeter-wave therapy device for scar treatment, characterized in that, include: S1: Acquire images of the scar; S2: Extract the scar difference features from the scar image; S3: Input the scar difference features into a multi-objective Gaussian process regression model, combine it with the treatment response index, and output the posterior conditional distribution of the treatment device control parameters, including the posterior mean vector of the treatment device control parameters and the posterior covariance matrix of the treatment device control parameters. S4: Combining the posterior mean vector of the control parameters of the therapeutic instrument and the posterior covariance matrix of the control parameters of the therapeutic instrument, a safety adjustment factor for the control parameters of the therapeutic instrument is generated by Chebyshev's inequality. S5: Establish a safe and feasible domain for the control parameters of the therapeutic device regarding the safety adjustment factor; S6: Under the constraint of the safe and feasible region of the control parameters of the therapeutic instrument, determine the optimal control parameters of the therapeutic instrument according to the posterior condition distribution of the control parameters of the therapeutic instrument; S7: Control the therapeutic instrument according to the optimal therapeutic instrument control parameters.

2. The control method for the laser millimeter-wave therapy device for scar treatment according to claim 1, characterized in that, The scar difference characteristics include scar area, scar aspect ratio, scar protrusion, and scar direction; S2 specifically refers to: The differential features of the scars were extracted using a pre-trained convolutional neural network model.

3. The control method for the laser millimeter-wave therapy device for scar treatment according to claim 1, characterized in that, The treatment response indicators include the normalized value of the treatment effect score and the normalized value of the tissue damage risk rate; the control parameters of the treatment device include the laser power, millimeter wave frequency, treatment duration and spot size of the treatment device.

4. The control method for the laser millimeter-wave therapy device for scar treatment according to claim 1, characterized in that, S3 specifically includes: S301: Acquire multiple sample scar images and corresponding label data; S302: Extract the sample scar difference features of each of the sample scar images, wherein the sample scar difference features and the corresponding label data form training data, wherein the label data includes the control parameters of the treatment device and the treatment response index; S303: With the goal of the normalized value of the treatment effect score in the treatment response index being greater than the preset normalized value of the treatment effect score, the multi-objective Gaussian process regression model is trained using the training data in combination with a composite kernel function; S304: Input the scar difference features into the trained multi-objective Gaussian process regression model, and output the posterior mean vector of the treatment device control parameters and the initial posterior covariance matrix of the treatment device control parameters; S305: Correct the posterior covariance matrix of the initial treatment instrument control parameters using the treatment response index to obtain the posterior covariance matrix of the treatment instrument control parameters.

5. The control method for the laser millimeter-wave therapy device for scar treatment according to claim 1, characterized in that, S4 specifically includes: S401: Calculate the skewness of each of the treatment instrument control parameters under the posterior mean vector of the treatment instrument control parameters and the posterior covariance matrix of the treatment instrument control parameters; S402: Substitute the skewness and the posterior covariance matrix of the control parameters of the treatment device into the Chebyshev inequality to calculate the upper bound of the probability of the control parameters of the treatment device relative to the posterior mean vector of the control parameters of the treatment device under different total deviation thresholds. S403: Replace the total deviation threshold with a factor term relating to the posterior covariance matrix of the safety adjustment factor and the control parameters of the treatment device; S404: Let the upper bound of the probability be less than the preset upper bound of the probability, and then deduce the safety adjustment factor by combining the factor terms.

6. The control method for the laser millimeter-wave therapy device for scar treatment according to claim 1, characterized in that, Specifically, S5 is: Based on the safety adjustment factor, the posterior mean vector of the control parameters of the therapeutic instrument, and the posterior covariance matrix of the control parameters of the therapeutic instrument, a safe and feasible region for the control parameters of the therapeutic instrument is established.

7. The control method for the laser millimeter-wave therapy device for scar treatment according to claim 1, characterized in that, S6 specifically includes: S601: Determine whether all the control parameters in the posterior mean vector of the control parameters of the treatment device are within the safe and feasible region of the control parameters of the treatment device; if yes, update the posterior mean vector of the control parameters of the treatment device to the optimal control parameters of the treatment device and proceed to step S603; otherwise, proceed to step S602. S602: Correct the control parameters of the therapeutic instrument that exceed the safe and feasible range of the control parameters of the therapeutic instrument to obtain the corrected control parameters of the therapeutic instrument, and update the corrected control parameters of the therapeutic instrument to the optimal control parameters of the therapeutic instrument. S603: Output the optimal control parameters of the therapeutic instrument.

8. The control method for a laser millimeter-wave therapy device for scar treatment according to claim 2, characterized in that, Also includes: The multi-objective Gaussian process regression model and the pre-trained convolutional neural network model are updated at preset intervals.

9. A control system for a laser millimeter-wave therapy device for scar treatment, characterized in that, include: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the scar treatment laser millimeter-wave therapy device control method as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the scar treatment laser millimeter wave therapy device control method as described in any one of claims 1 to 8.

Images