Wearable millimeter wave therapy device with MMIC microstrip

Abstract

The present application discloses a wearable millimeter wave therapeutic device with an MMIC microstrip, and relates to the technical field of wearable millimeter wave therapeutic devices. It includes: using a camera to collect skin status images of the lesion site in the millimeter wave treatment area in real time, using deep learning-based image processing technology to perform image analysis on the skin status images of the lesion site at each time point, extracting the skin status characteristics of the lesion site, and using time series analysis technology to dig out the skin status change trend of the lesion site, and then adjusting the parameters of the millimeter wave therapeutic device based on this, so that the therapeutic device can adjust the parameters in real time according to the changes in the patient's skin status during the treatment process, thereby improving the stability and safety of the treatment effect.

Description

Technical Field

[0001] The present application relates to the technical field of wearable millimeter wave therapeutic devices, and in particular, to a wearable millimeter wave therapeutic device with an MMIC microstrip. Background Art

[0002] In recent years, with the advancement of medical technology and the continuous improvement of people's health needs, non-invasive treatment methods have received widespread attention and development. Among them, millimeter wave therapy, as an emerging physical therapy method, has shown broad application prospects in pain management, inflammation treatment, tissue repair, etc. due to its non-invasive, safe and effective characteristics. Millimeter wave therapy uses electromagnetic waves in the millimeter wave band (usually refers to the 30 GHz to 300 GHz frequency band) to irradiate the human body to promote cell metabolism, accelerate blood circulation and enhance immune function.

[0003] Traditional millimeter wave therapy equipment is mostly large, fixed devices that need to be used in professional places such as hospitals or clinics, which limits its convenience and accessibility. With the development of portable electronic device technology, wearable millimeter wave therapy devices have emerged. These devices integrate microstrip integrated circuits (MMICs), making them smaller and lighter, making them easier for patients to use at home or when they are out and about. However, how to ensure that the energy distribution of millimeter waves during treatment is uniform and accurate in the target area, and how to adjust the treatment parameters in real time according to individual differences is one of the main technical challenges currently facing wearable millimeter wave therapy devices.

[0004] The parameter adjustment of millimeter wave therapy devices on the market during treatment mostly relies on the doctor's experience or the patient's subjective feedback, lacking objective and quantitative data support, which may lead to unstable treatment effects or overtreatment. In addition, as the treatment progresses, the patient's skin condition will change, and fixed treatment parameters cannot adapt to the changes in the patient's skin condition, which may affect the treatment effect.

[0005] Therefore, a wearable millimeter wave therapeutic device with MMIC microstrip is needed to solve the above technical problems. Summary of the invention

[0006] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0007] In a first aspect, the present application provides a wearable millimeter wave therapeutic device with an MMIC microstrip, the wearable millimeter wave therapeutic device comprising:

[0008] A therapeutic device and a wearable structure connected to the therapeutic device, wherein the wearable structure is suitable for being fixed to a treatment site of a human body;

[0009] The therapeutic device includes an electronic camera, a millimeter wave transmitting module, a control module and a power module. The millimeter wave transmitting module includes an MMIC microstrip antenna, and the MMIC microstrip antenna is used to generate and transmit millimeter waves; the control module is used to control the parameters of the millimeter waves, and the parameters include frequency, power and transmission time; the power module is electrically connected to the millimeter wave transmitting module and the control module, and is used to supply power to the millimeter wave transmitting module and the control module; the electronic camera is used to collect skin status images of the lesion site;

[0010] Wherein, the control module comprises:

[0011] A data receiving unit, used for receiving a time queue of skin status images of the lesion site collected by the electronic camera;

[0012] A lesion site skin state feature extraction unit, used to respectively extract the skin state features of each lesion site skin state image in the time queue of the lesion site skin state images to obtain a time queue of lesion site skin state feature images;

[0013] A feature enhancement unit, configured to perform feature gradient enhancement on the time queue of the lesion site skin state feature map to obtain a time queue of the lesion site skin state enhanced feature map;

[0014] A feature aggregation unit, used for performing temporal significance aggregation on the time queue of the skin state enhancement feature map of the lesion site to obtain a skin state temporal attenuation significant aggregation representation vector;

[0015] The millimeter wave adaptive adjustment unit is used to generate a millimeter wave adaptive adjustment instruction based on the skin state temporal attenuation significant aggregation representation vector.

[0016] Optionally, the lesion site skin state feature extraction unit is used to: input each lesion site skin state image in the time queue of the lesion site skin state images into a lesion site skin state feature extractor based on a void convolutional neural network model to obtain the time queue of the lesion site skin state feature map.

[0017] Optionally, the feature enhancement unit is used to: input the time queue of the skin state feature map of the lesion site into a gradient mask feature significance enhancement module to obtain the time queue of the skin state enhanced feature map of the lesion site.

[0018] Optionally, the feature enhancement unit includes: a gradient amplitude calculation subunit, which is used to calculate the multidirectional gradient value distribution of each position in the lesion site skin state characteristic map, and determine the gradient amplitude value of each position in the lesion site skin state characteristic map based on the multidirectional gradient value distribution of each position to obtain the lesion site skin state characteristic gradient amplitude distribution map; a local significance measurement subunit, which is used to calculate the gradient amplitude local description operator of each position in the lesion site skin state characteristic gradient amplitude distribution map to obtain the lesion site skin state characteristic gradient amplitude local significant distribution map; a gated mask subunit, which is used to input the lesion site skin state characteristic gradient amplitude local significant distribution map into a gated masker based on the GELU function to obtain a gradient amplitude local significant gated mask map; a mask application subunit, which is used to calculate the position point multiplication between the gradient amplitude local significant gated mask map and the lesion site skin state characteristic map to obtain the lesion site skin state enhanced feature map.

[0019] Optionally, the local significance measurement subunit is used to: determine the scale of the local neighborhood, calculate the average of the differences between the gradient amplitude value at a predetermined position in the gradient amplitude distribution map of the skin state characteristics of the lesion site and the gradient amplitude values ​​at other positions in the local neighborhood to obtain a local description operator of the gradient amplitude corresponding to the predetermined position.

[0020] Optionally, the feature aggregation unit is used to: input the time queue of the skin state enhancement feature map of the lesion site into a feature aggregation network guided by node feature significance attenuation to obtain the skin state temporal attenuation significance aggregation representation vector.

[0021] Optionally, the feature aggregation unit comprises: a feature shape reshaping subunit, used to perform feature shape reshaping on each lesion site skin condition enhancement feature map in the time queue of the lesion site skin condition enhancement feature map to obtain a time queue of lesion site skin condition enhancement feature vectors; a feature significance calculation subunit, used to calculate the feature significance description factor of each lesion site skin condition enhancement feature vector in the time queue of the lesion site skin condition enhancement feature vectors; a feature significance attenuation factor calculation subunit, used to use the last lesion site skin condition enhancement feature vector in the time queue of the lesion site skin condition enhancement feature vector as the current lesion site skin condition enhancement feature vector, and based on the other lesion site skin condition enhancement feature vectors in the time queue of the lesion site skin condition enhancement feature vectors, calculate the feature significance description factor of each lesion site skin condition enhancement feature vector in the time queue of the lesion site skin condition enhancement feature vectors. The distance span between the feature vector and the current lesion site skin condition enhancement feature vector is used to construct the feature significance attenuation factor of the other lesion site skin condition enhancement feature vectors; a significance modulation subunit is used to calculate the product between the feature significance attenuation factor of the other lesion site skin condition enhancement feature vectors and their feature significance description factors to obtain a time queue of the feature significance attenuation description factors; a weighting subunit is used to input the time queue of the feature significance attenuation description factors into the gated mask module to obtain a time queue of the feature significance attenuation weight factors; a weight application subunit is used to calculate the weighted sum of the time queue of the lesion site skin condition enhancement feature vector based on the time queue of the feature significance attenuation weight factors to obtain the skin state temporal attenuation significant aggregation representation vector.

[0022] Optionally, the feature significance calculation subunit is used to: calculate the expected value of the fourth power of the difference between each eigenvalue in the lesion site skin condition enhanced feature vector and its feature mean, and divide the expected value by the square of the feature variance of the lesion site skin condition enhanced feature vector to obtain the feature significance description factor.

[0023] Optionally, the feature significance attenuation factor calculation subunit is used to: calculate the difference between the maximum eigenvalue of the current lesion site skin condition enhancement feature vector and the maximum eigenvalue of each other lesion site skin condition enhancement feature vector in the time queue of the lesion site skin condition enhancement feature vector, and then divide the difference by the number of eigenvectors separated by the two to obtain the feature significance attenuation factors of each other lesion site skin condition enhancement feature vector.

[0024] Optionally, the millimeter wave adaptive adjustment unit is used to: input the skin state temporal attenuation significant aggregation representation vector into a decoder-based adjustment instruction generator to obtain the adaptive adjustment instruction, and the adaptive adjustment instruction includes the decoded value of the recommended frequency of the millimeter wave at the next time point.

[0025] The present application has at least the following technical effects: compared with the prior art, the above-mentioned technical scheme is adopted to collect the skin condition images of the lesion site in the millimeter wave treatment area in real time through the camera, and the skin condition images of the lesion site at each time point are analyzed by using the image processing technology based on deep learning to extract the skin condition characteristics of the lesion site, and the skin condition change trend of the lesion site is excavated through the time series analysis technology, and then the parameters of the millimeter wave therapy device are adjusted accordingly, so that the therapy device can adjust the parameters in real time according to the changes in the patient's skin condition during the treatment process, thereby improving the stability and safety of the treatment effect.

[0026] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0028] Figure 1 It is a schematic diagram of the structure of a wearable millimeter wave therapeutic device with an MMIC microstrip according to an exemplary embodiment.

[0029] Figure 2 It is a block diagram of the control module of a wearable millimeter wave therapeutic device with an MMIC microstrip according to an exemplary embodiment.

[0030] Figure 3 The present invention is a flow chart showing a wearable millimeter wave treatment method with an MMIC microstrip according to an exemplary embodiment.

[0031] Figure 4 It is a diagram of an application scenario of a wearable millimeter wave therapeutic device with an MMIC microstrip according to an exemplary embodiment.

[0032] The accompanying drawings are marked as follows: 1. treatment device; 2. wearable structure; 11. camera; 12. millimeter wave transmitting module; 13. control module; 14. power module; 101. data receiving unit; 102. lesion site skin condition feature extraction unit; 103. feature enhancement unit; 104. feature aggregation unit; 105. millimeter wave adaptive adjustment unit. DETAILED DESCRIPTION

[0033] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0034] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0035] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0036] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0037] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0038] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0039] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0040] Figure 1 FIG. 1 is a schematic diagram showing a structure of a wearable millimeter wave therapeutic device with an MMIC microstrip according to an exemplary embodiment. Figure 1 As shown, the wearable millimeter wave therapeutic device includes:

[0041] A therapeutic device 1 and a wearable structure 2 connected to the therapeutic device 1, wherein the wearable structure 2 is suitable for being fixed to a human body.

[0042] Among them, the therapeutic device 1 includes an electronic camera 11, a millimeter wave transmitting module 12, a control module 13 and a power module 14, the millimeter wave transmitting module 12 includes an MMIC microstrip antenna, and the MMIC microstrip antenna is used to generate and transmit millimeter waves; the control module 13 is used to control the parameters of the millimeter waves, and the parameters include frequency, power and transmission time; the power module 14 is electrically connected to the millimeter wave transmitting module 12 and the control module 13, and is used to power the millimeter wave transmitting module 12 and the control module 13; the electronic camera 11 is used to collect skin condition images of the lesion site.

[0043] The wearable millimeter wave therapeutic device with MMIC microstrip is a new type of medical device that integrates monolithic microwave integrated circuit (MMIC) technology. The application of MMIC technology in the millimeter wave frequency band makes the therapeutic device have the characteristics of high integration, low power consumption and miniaturization, which is very suitable for wearable devices.

[0044] The millimeter wave therapy device resonates with human cells through millimeter waves, promotes cell vitality, thereby enhancing cell metabolism and shortening treatment time. This therapy device is very convenient to use and can be carried around by users. In addition, the millimeter wave therapy device is also non-invasive and has a remote effect, so it can eliminate other diseases while treating a certain disease.

[0045] The technical concept of the present application is: to collect skin condition images of the lesion site in the millimeter wave treatment area in real time through a camera, to perform image analysis on the skin condition images of the lesion site at each time point using deep learning-based image processing technology, to extract the skin condition characteristics of the lesion site, and to use time series analysis technology to dig out the skin condition change trend of the lesion site, and then use this to adjust the parameters of the millimeter wave therapy device, so that the therapy device can adjust the parameters in real time according to the changes in the patient's skin condition during the treatment process, thereby improving the stability and safety of the treatment effect.

[0046] Figure 2 is a block diagram of the control module of a wearable millimeter wave therapeutic device with an MMIC microstrip according to an exemplary embodiment, such as Figure 2 As shown, the control module 13 includes:

[0047] The data receiving unit 101 is used to receive the time queue of the skin status image of the lesion site collected by the electronic camera;

[0048] The lesion site skin state feature extraction unit 102 is used to respectively extract the skin state features of each lesion site skin state image in the time queue of the lesion site skin state images to obtain a time queue of lesion site skin state feature images;

[0049] A feature enhancement unit 103 is used to perform feature gradient enhancement on the time queue of the lesion site skin state feature map to obtain a time queue of the lesion site skin state enhanced feature map;

[0050] A feature aggregation unit 104 is used to perform temporal significance aggregation on the time queue of the skin state enhancement feature map of the lesion site to obtain a skin state temporal attenuation significant aggregation representation vector;

[0051] The millimeter wave adaptive adjustment unit 105 is used to generate a millimeter wave adaptive adjustment instruction based on the skin state temporal attenuation significant aggregation representation vector.

[0052] Specifically, in the control module, first, a time queue of skin status images of the lesion site collected by the electronic camera is received. It should be understood that, affected by factors such as treatment progress and individual differences, the skin status of the patient's lesion site will change dynamically, such as inflammation subsides, redness and swelling are reduced, etc. In the technical solution of the present application, by continuously monitoring the patient's skin status and obtaining a time queue of skin status images of the lesion site, the parameters of the millimeter wave therapeutic device can be dynamically adjusted according to the changing trend of the patient's skin status, thereby optimizing the treatment effect and avoiding unnecessary harm to the patient caused by improper millimeter wave parameter settings.

[0053] Next, in order to fully extract the characteristic information of the skin condition of the lesion site, the present application adopts a dilated convolutional neural network model with excellent performance in the field of image processing as a feature extractor of the skin condition of the lesion site, and performs feature extraction on each lesion site skin condition image in the time queue of the lesion site skin condition image. By utilizing the dilated structure of the dilated convolutional neural network and the sparse convolution method, the receptive field (Receptive Field) of the convolution kernel is expanded without increasing the number of parameters and computational complexity, so as to effectively capture a wider range of image context information in the lesion site skin condition image and maintain high resolution of the features, thereby mining key features such as the edge contour of the lesion site and the color texture of the skin, so as to obtain the time queue of the lesion site skin condition feature map.

[0054] In one embodiment of the present application, the lesion site skin state feature extraction unit is used to: input each lesion site skin state image in the time queue of the lesion site skin state image into a lesion site skin state feature extractor based on a void convolutional neural network model to obtain the time queue of the lesion site skin state feature map.

[0055] Then, in order to further highlight the key detail features such as the edge contour of the skin state of the lesion site, the present application introduces a gradient mask feature saliency enhancement module to process the time queue of the skin state feature map of the lesion site, and generates a mask based on the feature gradient distribution of the skin state feature map of the lesion site and applies it to the feature map, thereby improving the distinguishability of the feature and achieving the enhancement of the significance of the key features. Specifically, the module first quantifies the local change intensity of the pixels at each position by calculating the gradient amplitude value of each position in the skin state feature map of the lesion site. Then, the gradient amplitude local description operator is calculated by measuring the significance of the gradient amplitude value of each position relative to the gradient amplitude value of other positions in the local neighborhood, and the feature importance of each position is evaluated. Then, the gradient amplitude local description operator of each position is nonlinearly activated using the GELU function to generate a gated mask to guide the enhancement of the contour information in the feature map. Finally, the feature map is element-by-element weighted based on the generated gated mask, so as to focus on the contour salient area in the feature map and remove the background noise to obtain a clearer and more distinguishable time queue of the skin state enhanced feature map of the lesion site.

[0056] In one embodiment of the present application, the feature enhancement unit is used to: input the time queue of the skin state feature map of the lesion site into the gradient mask feature significance enhancement module to obtain the time queue of the skin state enhanced feature map of the lesion site.

[0057] Further, in one embodiment of the present application, the feature enhancement unit includes: a gradient amplitude calculation subunit, which is used to calculate the multidirectional gradient value distribution of each position in the lesion site skin state characteristic map, and determine the gradient amplitude value of each position in the lesion site skin state characteristic map based on the multidirectional gradient value distribution of each position to obtain the lesion site skin state characteristic gradient amplitude distribution map; a local significance measurement subunit, which is used to calculate the gradient amplitude local description operator of each position in the lesion site skin state characteristic gradient amplitude distribution map to obtain the lesion site skin state characteristic gradient amplitude local significant distribution map; a gated mask subunit, which is used to input the lesion site skin state characteristic gradient amplitude local significant distribution map into a gated masker based on the GELU function to obtain a gradient amplitude local significant gated mask map; a mask application subunit, which is used to calculate the position point multiplication between the gradient amplitude local significant gated mask map and the lesion site skin state characteristic map to obtain the lesion site skin state enhanced feature map.

[0058] Furthermore, in one embodiment of the present application, the local significance measurement subunit is used to: determine the scale of the local neighborhood, calculate the average of the differences between the gradient amplitude value at a predetermined position in the gradient amplitude distribution map of the skin condition characteristics of the lesion site and the gradient amplitude values ​​at other positions in the local neighborhood to obtain a local description operator of the gradient amplitude corresponding to the predetermined position.

[0059] Specifically, the skin condition feature map of the lesion site is processed by the following feature distribution gradient mask formula to obtain the enhanced skin condition feature map of the lesion site, wherein the feature distribution gradient mask formula is:

[0060]

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067] in, , , , , and respectively represent the grayscale values ​​of the corresponding positions of the skin state characteristic map of the lesion site, In the skin condition characteristic diagram of the lesion site The gradient value in the horizontal direction of the pixel point. In the skin condition characteristic diagram of the lesion site The gradient value in the ordinate direction of the pixel point. The skin condition characteristic diagram of the lesion site is shown in FIG. The channel direction gradient value of the position pixel point, The first characterization diagram of the skin condition at the lesion site is represented by The gradient magnitude value at the position, The characteristic gradient amplitude distribution diagram of the skin state at the lesion site is represented by The set of gradient magnitude values ​​in the local neighborhood centered at position, represents the number of gradient magnitude values ​​in the set of gradient magnitude values ​​in the local neighborhood, , and Respectively represent the offset in the horizontal axis direction, the vertical axis direction and the channel direction, The first characterization diagram of the skin condition at the lesion site is represented by The local description operator of the gradient magnitude of the position, represents the GELU function, The first part of the gradient magnitude local significant gated mask map is represented by The local description operator of the gradient magnitude of the position, represents the local significant gated mask map of the gradient magnitude, A characteristic diagram of the skin condition at the lesion site is shown, It means point multiplication by position. The enhanced characteristic map of the skin condition at the lesion site is represented.

[0068] Furthermore, in order to fully understand the dynamic change trend of the skin condition of the lesion site, it is necessary to further perform a time-series correlation analysis on the time queue of the enhanced feature map of the skin condition of the lesion site. In particular, considering that during the treatment process, the influence of the skin condition information at the historical time point on the skin condition and millimeter wave parameter adjustment at the current moment will gradually weaken with the passage of time, therefore, the present application introduces a feature aggregation network guided by the node feature significance attenuation to perform a time-series correlation analysis on the time queue of the enhanced feature map of the skin condition of the lesion site. Specifically, this module first reshapes the features of the enhanced feature map of the skin condition of each lesion site into a vector form, and then calculates its feature significance description factor based on the mean and variance of the enhanced feature vector of the skin condition of each lesion site to quantitatively represent the importance of the feature. Next, a feature significance attenuation factor is constructed based on the distance span between the skin state enhancement feature vector of each lesion site and the current skin state enhancement feature vector of the lesion site (the end of the time series), and its feature significance description factor is modulated by this factor, so that the operation data information that is closer to the current moment in the feature space is given a higher weight, thereby more accurately reflecting the relative importance of the operation data features at each time point in the overall sequence. Then, the modulated description factor is further screened using the activation function and the gating mechanism, and the time queue of the skin state enhancement feature vector of the lesion site is weighted and summed using the screened description factor as the weight, so that in the process of temporal information aggregation, the early feature information gradually decays, while the recent feature information is given a higher weight, so as to obtain a skin state temporal attenuation significant aggregation representation vector with time correlation. In this way, it can be ensured that the therapeutic device pays more attention to the changes in the current skin state when adjusting the parameters, thereby improving the real-time and adaptability of the treatment.

[0069] In one embodiment of the present application, the feature aggregation unit is used to: input the time queue of the skin state enhancement feature map of the lesion site into a feature aggregation network guided by node feature significance attenuation to obtain the skin state temporal attenuation significance aggregation representation vector.

[0070] Wherein, the feature aggregation unit comprises: a feature shape reshaping subunit, used for performing feature shape reshaping on each lesion site skin state enhancement feature map in the time queue of the lesion site skin state enhancement feature map to obtain a time queue of lesion site skin state enhancement feature vectors; a feature significance calculation subunit, used for calculating the feature significance description factor of each lesion site skin state enhancement feature vector in the time queue of the lesion site skin state enhancement feature vectors; a feature significance attenuation factor calculation subunit, used for taking the last lesion site skin state enhancement feature vector in the time queue of the lesion site skin state enhancement feature vector as the current lesion site skin state enhancement feature vector, and calculating the feature significance attenuation factor calculation subunit based on the other lesion site skin state enhancement feature vectors in the time queue of the lesion site skin state enhancement feature vectors. The distance span between the eigenvector and the current lesion site skin condition enhancement feature vector is used to construct the feature significance attenuation factor of the other lesion site skin condition enhancement feature vectors; a significance modulation subunit is used to calculate the product between the feature significance attenuation factor of the other lesion site skin condition enhancement feature vectors and their feature significance description factors to obtain a time queue of the feature significance attenuation description factors; a weighting subunit is used to input the time queue of the feature significance attenuation description factors into the gated mask module to obtain a time queue of the feature significance attenuation weight factors; a weight application subunit is used to calculate the weighted sum of the time queue of the lesion site skin condition enhancement feature vector based on the time queue of the feature significance attenuation weight factors to obtain the skin state temporal attenuation significant aggregation representation vector.

[0071] In one embodiment of the present application, the feature significance calculation subunit is used to: calculate the expected value of the fourth power of the difference between each eigenvalue in the lesion site skin condition enhanced feature vector and its feature mean, and divide the expected value by the square of the feature variance of the lesion site skin condition enhanced feature vector to obtain the feature significance description factor.

[0072] Furthermore, in one embodiment of the present application, the feature significance attenuation factor calculation subunit is used to: calculate the difference between the maximum eigenvalue of the current lesion site skin condition enhancement feature vector and the maximum eigenvalue of each other lesion site skin condition enhancement feature vector in the time queue of the lesion site skin condition enhancement feature vector, and then divide the difference by the number of eigenvectors separated by the two to obtain the feature significance attenuation factor of each other lesion site skin condition enhancement feature vector.

[0073] Specifically, the time queue of the skin state enhancement feature vector of the lesion site is processed by the following feature significance attenuation guided fusion formula to obtain the skin state time series attenuation significant aggregation representation vector, wherein the feature significance attenuation guided fusion formula is:

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081] in, A time queue representing the enhanced feature vector of the skin state of the lesion site, The value of is the number of feature vectors in the time queue of the skin condition enhancement feature vector of the lesion site, , , , and The first, second, and third time series of the skin state enhancement feature vector of the lesion site are represented respectively. , and the current lesion site skin status enhanced feature vector, represents the maximum eigenvalue of the eigenvector, Indicates The first enhanced feature vector of the skin condition of the lesion eigenvalues, and Respectively represent the The feature mean and square of the feature variance of the enhanced feature vector of the skin condition of each lesion site, Indicates the The fourth-order central moment of the enhanced feature vector of the skin condition of the lesion site, that is, The expected value of the fourth power of the difference between each eigenvalue and its eigenmean in the enhanced eigenvector of the skin condition of the lesion site, , Indicates the The characteristic significance attenuation factor of the enhanced feature vector of the skin condition of the lesion site, Indicates The feature significance decay factor is described by Indicates Normalized feature significance decay description factor, Indicates feature significance decay weight factor, is the preset threshold, Indicates mask processing, A temporal attenuation significant aggregation representation vector representing the skin state.

[0082] Then, the skin state temporal attenuation significant aggregation representation vector is input into a decoder-based adjustment instruction generator to obtain an adaptive adjustment instruction, wherein the adaptive adjustment instruction includes a decoded value of a recommended frequency of a millimeter wave at a next time point.

[0083] In one embodiment of the present application, the millimeter wave adaptive adjustment unit is used to: input the skin state temporal attenuation significant aggregation representation vector into a decoder-based adjustment instruction generator to obtain the adaptive adjustment instruction, and the adaptive adjustment instruction includes the decoded value of the recommended frequency of the millimeter wave at the next time point.

[0084] That is, considering that each lesion site skin state enhancement feature map in the time queue of the lesion site skin state enhancement feature map respectively represents the image semantic enhancement features of the lesion site skin state image, therefore, when the time queue of the lesion site skin state enhancement feature map is subjected to feature aggregation guided by the node feature significance attenuation, the difference in the image semantic feature expression dimension of each lesion site skin state enhancement feature map in the time queue of the lesion site skin state enhancement feature map will cause the significance attenuation measurement deviation, thereby causing the richness of long-distance feature aggregation, resulting in the repetition of unstructured probability mapping, and affecting the expression effect of the skin state temporal attenuation significance aggregation representation vector.

[0085] In a preferred example, the step of inputting the skin state temporal attenuation significant aggregation representation vector into a decoder-based adjustment instruction generator to obtain an adaptive adjustment instruction includes the following steps:

[0086] Determine the number of zero eigenvalues ​​in the skin state temporal attenuation significant aggregation representation vector and zero eigenvalue , and the zero eigenvalue and the number of zero eigenvalues Multiply and divide them respectively to obtain the first skin state time series attenuation significant aggregation field weight value and the second skin state temporal decay significant aggregate field weight value ;

[0087] Calculate the square root of the sum of squares of all eigenvalues ​​of the skin state temporal attenuation significant aggregation representation vector to obtain the skin state temporal attenuation significant aggregation full mode weight ,in, Each eigenvalue of the skin state temporal attenuation significant aggregation representation vector is represented;

[0088] With the zero characteristic factor value The skin state temporal decay is calculated as an index for each eigenvalue of the significant aggregate representation vector The power function , and multiplied by the first skin state time series attenuation significant aggregate field weight value To obtain the first skin state time series attenuation significant aggregate median ;

[0089] Calculate each eigenvalue of the skin state time series attenuation significant aggregation representation vector The second skin state timing decays significantly with the aggregate field weight value and the skin state temporal decay significantly aggregates the full mode weights The product of obtains the significant aggregate intermediate value of the second skin state time series attenuation ;

[0090] The first skin state time series is attenuated to significantly aggregate the intermediate value The second skin state has a significant attenuation of the aggregate median value The weighted difference Composition optimized skin state temporal decay significant aggregate representation vector, and represents different weight hyperparameters;

[0091] The optimized skin state temporal attenuation significant aggregation representation vector is input into the decoder-based adjustment instruction generator to obtain an adaptive adjustment instruction.

[0092] Therefore, based on the vector field fitting law of high-dimensional manifold, the isolated zeros and their differentiable factors in the skin state temporal decay significant aggregation representation vector are used as the differentiable dimensions of the high-dimensional manifold, so as to fix the normal direction pointing to the center of the feature distribution near the local coordinates represented by the eigenvalues ​​of the skin state temporal decay significant aggregation representation vector, thereby achieving effective alignment of the feature distribution pattern of the skin state temporal decay significant aggregation representation vector to the probability density differentiable field to avoid invalid repetition in the probability mapping process caused by the unstructured characteristics of the distribution of the skin state temporal decay significant aggregation representation vector with rich interactive responses, thereby improving the accuracy of the adaptive adjustment instructions obtained by the decoder-based adjustment instruction generator based on the skin state temporal decay significant aggregation representation vector input.

[0093] In summary, the above scheme is adopted to collect skin status images of the lesion site in the millimeter wave treatment area in real time through the camera, and use the image processing technology based on deep learning to analyze the skin status images of the lesion site at each time point, extract the skin status characteristics of the lesion site, and use the time series analysis technology to dig out the skin status change trend of the lesion site, and then use this to adjust the parameters of the millimeter wave therapeutic device. In this way, the therapeutic device can adjust the parameters in real time according to the changes in the patient's skin status during the treatment process, thereby improving the stability and safety of the treatment effect.

[0094] Figure 3 is a flow chart showing a wearable millimeter wave treatment method with an MMIC microstrip according to an exemplary embodiment, such as Figure 3 As shown, the method includes:

[0095] Step S201, receiving a time queue of skin status images of lesion sites collected by the electronic camera;

[0096] Step S202, respectively extracting the skin state features of each lesion site skin state image in the time queue of the lesion site skin state images to obtain a time queue of lesion site skin state feature graphs;

[0097] Step S203, performing feature gradient enhancement on the time queue of the skin state characteristic graph of the lesion site to obtain a time queue of the skin state enhanced characteristic graph of the lesion site;

[0098] Step S204, performing temporal significance aggregation on the time queue of the skin state enhancement feature map of the lesion site to obtain a skin state temporal attenuation significant aggregation representation vector;

[0099] Step S205: Generate a millimeter wave adaptive adjustment instruction based on the skin state temporal attenuation significant aggregation representation vector.

[0100] Figure 4 FIG. 1 is an application scenario diagram of a wearable millimeter wave therapeutic device with an MMIC microstrip according to an exemplary embodiment. Figure 4 As shown, in this application scenario, first, the time queue of the skin status images of the lesion site collected by the electronic camera (for example, Figure 4 Then, the acquired skin condition image of the lesion site is input to a server (for example, Figure 4 In S) shown in , the server is capable of processing the skin condition image of the lesion site based on the wearable millimeter wave therapeutic device algorithm with MMIC microstrip to generate adaptive adjustment instructions for the millimeter wave.

[0101] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0102] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0103] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A wearable millimeter wave therapeutic device with MMIC microstrip, characterized in that: include: A therapeutic device and a wearable structure connected to the therapeutic device, wherein the wearable structure is suitable for being fixed to a treatment site of a human body; The therapeutic device includes an electronic camera, a millimeter wave transmitting module, a control module and a power module. The millimeter wave transmitting module includes an MMIC microstrip antenna, and the MMIC microstrip antenna is used to generate and transmit millimeter waves; the control module is used to control the parameters of the millimeter waves, and the parameters include frequency, power and transmission time; the power module is electrically connected to the millimeter wave transmitting module and the control module, and is used to supply power to the millimeter wave transmitting module and the control module; the electronic camera is used to collect skin status images of the lesion site; Wherein, the control module includes: A data receiving unit, used for receiving a time queue of skin status images of the lesion site collected by the electronic camera; A lesion site skin state feature extraction unit, used to respectively extract the skin state features of each lesion site skin state image in the time queue of the lesion site skin state images to obtain a time queue of lesion site skin state feature images; A feature enhancement unit, configured to perform feature gradient enhancement on the time queue of the lesion site skin state feature map to obtain a time queue of the lesion site skin state enhanced feature map; A feature aggregation unit, used for performing temporal significance aggregation on the time queue of the skin state enhancement feature map of the lesion site to obtain a skin state temporal attenuation significant aggregation representation vector; Among them, the feature enhancement unit includes: A characteristic shape reshaping subunit is used to perform characteristic shape reshaping on each lesion site skin state enhancement feature map in the time queue of the lesion site skin state enhancement feature map to obtain a time queue of lesion site skin state enhancement feature vectors; a feature significance attenuation factor calculation subunit, configured to use the last lesion site skin state enhancement feature vector in the time queue of the lesion site skin state enhancement feature vector as the current lesion site skin state enhancement feature vector, and construct feature significance attenuation factors of the other lesion site skin state enhancement feature vectors based on the distance span between the other lesion site skin state enhancement feature vectors in the time queue of the lesion site skin state enhancement feature vector and the current lesion site skin state enhancement feature vector; The millimeter wave adaptive adjustment unit is used to generate a millimeter wave adaptive adjustment instruction based on the skin state temporal attenuation significant aggregation representation vector.

2. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 1, characterized in that: The lesion site skin state feature extraction unit is used to: Each lesion site skin state image in the time queue of the lesion site skin state image is input into a lesion site skin state feature extractor based on a hole convolutional neural network model to obtain a time queue of the lesion site skin state feature map.

3. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 2, characterized in that: The feature enhancement unit is used to: The time queue of the skin state feature map of the lesion site is input into the gradient mask feature significance enhancement module to obtain the time queue of the skin state enhanced feature map of the lesion site.

4. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 3, characterized in that: The feature enhancement unit comprises: a gradient amplitude calculation subunit, configured to calculate the multi-directional gradient value distribution of each position in the lesion site skin state characteristic map, and determine the gradient amplitude value of each position in the lesion site skin state characteristic map based on the multi-directional gradient value distribution of each position to obtain a lesion site skin state characteristic gradient amplitude distribution map; A local significance measurement subunit, used for calculating the local description operator of the gradient amplitude at each position in the gradient amplitude distribution map of the skin state characteristic of the lesion site to obtain a local significant distribution map of the gradient amplitude of the skin state characteristic of the lesion site; A gated mask subunit, used for inputting the local significant distribution map of the gradient amplitude of the skin state characteristics of the lesion site into a gated masker based on the GELU function to obtain a local significant gated mask map of the gradient amplitude; The mask applying subunit is used to calculate the position point multiplication between the gradient amplitude local significant gated mask map and the skin state feature map of the lesion site to obtain the skin state enhanced feature map of the lesion site.

5. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 4, characterized in that: The local saliency measurement subunit is used to: Determine the scale of the local neighborhood, calculate the average of the difference between the gradient amplitude value at a predetermined position in the gradient amplitude distribution map of the skin state characteristics of the lesion site and the gradient amplitude values ​​at other positions in the local neighborhood to obtain a gradient amplitude local description operator corresponding to the predetermined position.

6. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 5, characterized in that: The feature aggregation unit is used to: The time queue of the enhanced feature map of the skin state of the lesion site is input into a feature aggregation network guided by the node feature significance attenuation to obtain the skin state temporal attenuation significant aggregation representation vector.

7. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 6, characterized in that: The feature aggregation unit comprises: A feature significance calculation subunit, used for calculating a feature significance description factor of each lesion site skin condition enhancement feature vector in the time queue of the lesion site skin condition enhancement feature vector; A significance modulation subunit, used for calculating the product between the characteristic significance attenuation factor of the skin state enhancement characteristic vector of each other lesion site and its characteristic significance description factor to obtain a time queue of the characteristic significance attenuation description factor; A weighting subunit, configured to input the time queue of the feature significance decay description factor into a gated mask module to obtain a time queue of the feature significance decay weight factor; The weight applying subunit is used to calculate the weighted sum of the time queue of the skin state enhancement feature vector of the lesion site based on the time queue of the feature significance attenuation weight factor to obtain the skin state temporal attenuation significant aggregation representation vector.

8. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 7, characterized in that: The feature significance calculation subunit is used to: The expected value of the fourth power of the difference between each eigenvalue in the lesion site skin condition enhanced feature vector and its feature mean is calculated, and the expected value is divided by the square of the feature variance of the lesion site skin condition enhanced feature vector to obtain the feature significance description factor.

9. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 8, characterized in that: The feature significance attenuation factor calculation subunit is used to: Calculate the difference between the maximum eigenvalue of the current lesion site skin condition enhanced feature vector and the maximum eigenvalue of each other lesion site skin condition enhanced feature vector in the time queue of the lesion site skin condition enhanced feature vector, and then divide the difference by the number of eigenvectors between the two to obtain the characteristic significance attenuation factor of each other lesion site skin condition enhanced feature vector.

10. The wearable millimeter wave therapeutic device with MMIC microstrip according to claim 9, characterized in that: The millimeter wave adaptive adjustment unit is used to: The skin state temporal attenuation significant aggregation representation vector is input into a decoder-based adjustment instruction generator to obtain the adaptive adjustment instruction, wherein the adaptive adjustment instruction includes a decoded value of the recommended frequency of the millimeter wave at the next time point.

Images