Intelligent four-dimensional eye physiotherapy system and method

Abstract

The application discloses an intelligent four-dimensional eye physiotherapy system and method, and relates to the technical field of eye physiotherapy.The system comprises a multi-source data acquisition module, a double-mode feature extraction module, an intelligent syndrome differentiation module, a physical field scheme decision module, a physiotherapy parameter dynamic programming module and a four-dimensional physiotherapy prescription generation module, wherein the intelligent syndrome differentiation module is used for carrying out multi-source information fusion on structured Chinese medicine symptom information, blood vessel quantitative features and thermal imaging quantitative features, and making a judgment based on a preset syndrome classification relationship; the physical field scheme decision module is used for determining a dominant physical field and at least one auxiliary physical field from a thermal physical field, a low-frequency pulse magnetic field, a specific spectrum light field and a micro-vibration sound field according to a current dominant Chinese medicine syndrome type; and the core design of multi-source information fusion and accurate syndrome type judgment is achieved, objective quantitative representation of Chinese medicine syndrome types and disease severity is realized, and a physiotherapy scheme is accurately matched with individual core pathogenesis.

Description

Technical Field

[0001] This invention relates to the field of eye physiotherapy technology, and in particular to an intelligent four-dimensional eye physiotherapy system and method. Background Technology

[0002] With the high-frequency use of electronic devices in modern lifestyles, the incidence of eye problems such as eye strain, dry eye syndrome, and refractive errors continues to rise, making eye physiotherapy an important means of maintaining eye health. Current eye physiotherapy techniques encompass various forms, including traditional Chinese medicine external therapies and single-field physical therapy devices. Traditional Chinese medicine physiotherapy relies on the theory of viscera and meridians, emphasizing "treatment based on syndrome differentiation," while modern physical field therapy focuses on improving the physiological state of the eyes through physical effects. Both techniques are widely used in clinical practice and daily health maintenance. However, as people's demands for precision and personalization in physiotherapy increase, the inherent limitations of existing technologies are gradually becoming apparent, making it difficult to meet diverse health needs.

[0003] The core deficiency of existing ocular physiotherapy technology lies in the lack of precise adaptation between physiotherapy plans and the individual syndrome types and severity of patients' conditions. The physical field types of existing physiotherapy equipment are mostly fixed configurations, without differentiated design for the core pathogenesis of different syndrome types. At the same time, key parameters such as physiotherapy intensity and duration of action are mostly standardized, lacking a mechanism for dynamic adjustment based on the severity of the condition. This makes it impossible to provide targeted intervention for the core pathogenesis and difficult to adapt to the different ocular physiological differences of different patients, ultimately resulting in inconsistent physiotherapy effects, and may even affect the treatment efficiency due to inappropriate plans. Summary of the Invention

[0004] This invention provides an intelligent four-dimensional eye therapy system and method to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an intelligent four-dimensional eye physiotherapy system, the system comprising a multi-source data acquisition module, a dual-modal feature extraction module, an intelligent syndrome differentiation module, a physical field scheme decision-making module, a physiotherapy parameter dynamic planning module, and a four-dimensional physiotherapy prescription generation module, wherein: The multi-source data acquisition module is used to acquire visible light images and periocular infrared thermal images of the user's eyes, and to collect the user's structured TCM symptom information; The dual-modal feature extraction module is used to extract blood vessel quantification features that characterize the morphology and distribution density of scleral blood vessels from visible light images of the eye, and to extract thermal imaging quantification features that characterize the temperature gradient of specific acupoint areas around the eye socket from infrared thermal imaging images of the periorbital eye. The intelligent syndrome differentiation module is used to fuse structured TCM symptom information, blood vessel quantitative features and thermal imaging quantitative features into multi-source information, and make a judgment based on the preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value. The physical field scheme decision module is used to determine a dominant physical field and at least one auxiliary physical field from thermophysical fields, low-frequency pulsed magnetic fields, specific spectral light fields, and micro-vibration sound fields based on the current dominant TCM syndrome type. The dynamic planning module for physiotherapy parameters is used to dynamically configure the output intensity ratio, action time ratio, and application time sequence of the dominant physical field and each auxiliary physical field according to the quantitative syndrome intensity value. The four-dimensional physiotherapy prescription generation module is used to generate executable, personalized four-dimensional physiotherapy prescriptions based on the configured output intensity ratio, action time ratio, and application time sequence.

[0006] Preferably, when the multi-source data acquisition module acquires the user's visible light image and periocular infrared thermal image, and collects the user's structured TCM symptom information, it is specifically used for: Segmenting the sclera region in a visible light image of the eye yields a sclera region image; Identify the vascular network in the scleral region image, analyze the morphological bifurcation features of the vascular network, and obtain the quantitative features of blood vessels; The inner and outer canthal regions in the periocular infrared thermal imaging image were calibrated, and the temperature field distribution pattern between the inner and outer canthal regions was extracted to obtain the thermal imaging quantitative features. The symptom descriptions in structured TCM symptom information are analyzed, and the symptom descriptions are associated with TCM organ differentiation items to form a standard symptom feature vector.

[0007] Preferably, when the dual-modal feature extraction module extracts vascular quantification features characterizing the morphology and distribution density of scleral vascular morphology from visible light images of the eye, it is specifically used for: Identify the vascular network in the sclera region of the eye in visible light images to obtain a vascular network structure map; By analyzing the bifurcation and terminal points of blood vessels in the vascular network structure diagram and statistically analyzing the density of bifurcation points per unit area, the characteristics of blood vessel distribution density are obtained. Assess the tortuosity and straightness characteristics of the main vessel segments in the vascular network structure diagram, quantify their tortuosity and consistency of direction, and form the morphological characteristics of blood vessels; Combining the distribution density characteristics and morphological characteristics of blood vessels, we can construct the quantitative characteristics of blood vessels.

[0008] Preferably, when the dual-modal feature extraction module extracts thermal imaging quantization features characterizing the temperature gradient of specific acupoint areas around the eye socket from the periorbital infrared thermal imaging image, it is specifically used for: In the periocular infrared thermal imaging image, the Jingming acupoint area and the Tongziliao acupoint area were marked respectively to obtain the first acupoint area and the second acupoint area; Extract the temperature distribution in the first acupoint region and the second acupoint region, calculate the average temperature decay rate between the highest temperature point and the contour boundary in each region, and obtain the local temperature gradient characteristics of the acupoint. By comparing the temperature distribution patterns of the first and second acupoint areas, the relative positional relationship between the high-temperature core and the surrounding area is analyzed to form the temperature field pattern characteristics between acupoints. The local temperature gradient characteristics of acupoints and the temperature field pattern characteristics between acupoints constitute the quantitative features of thermal imaging.

[0009] Preferably, when the intelligent syndrome differentiation module performs a judgment based on a preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value, it is specifically used for: Based on the correspondence between the internal organs and the eyes in traditional Chinese medicine, the symptom items in the structured TCM symptom information are mapped with the quantitative features of blood vessels and the quantitative features of thermal imaging to form a multi-source feature set with correlation weights. Based on the association weights of each feature item in the multi-source feature set, the single organ syndrome type pointed to by the feature combination with the highest weight is identified as the current dominant TCM syndrome type. The degree to which each characteristic item constituting the current dominant TCM syndrome deviates from its theoretical health benchmark is analyzed, and a weighted synthesis is performed based on their correlation weights to obtain a quantitative syndrome intensity value.

[0010] Preferably, when the intelligent syndrome differentiation module analyzes the degree to which each feature item constituting the current dominant TCM syndrome deviates from its theoretical health benchmark, and performs weighted synthesis based on its correlation weights to obtain a quantitative syndrome intensity value, it is specifically used for: Based on the quantitative characteristics of blood vessels, the dispersion of their blood vessel distribution density and morphological straightness relative to the baseline range was determined; Based on the quantitative characteristics of thermal imaging, the magnitude of the change in temperature gradient in the acupoint area relative to the baseline gradient was determined. Based on the severity level of each symptom item in the structured TCM symptom information, the intensity range of the symptom is determined. The discrete amplitude, the variation amplitude, and the symptom intensity amplitude are multiplied by their respective association weights in the multi-source feature set and then summed to obtain the quantitative syndrome intensity value.

[0011] Preferably, when the physical field scheme decision module determines a dominant physical field and at least one auxiliary physical field from the thermophysical field, low-frequency pulsed magnetic field, specific spectral light field, and micro-vibration sound field based on the current dominant TCM syndrome type, it is specifically used for: Analyze the core pathogenesis attributes indicated by the current dominant TCM syndrome types and classify them into either the attribute of excessive Yang or the attribute of deficiency. For the yang excess attribute, a wavelength band with a clearing and draining effect in a specific spectral light field is selected as the dominant physical field; for the deficiency attribute, a thermophysical field is selected as the dominant physical field. Based on the viscera and meridian pathways associated with the core pathogenesis attributes, at least one auxiliary physical field is selected from the micro-vibration sound field and low-frequency pulse magnetic field that complements the action site of the dominant physical field. Combine the dominant physical field with at least one auxiliary physical field to form a physical field combination scheme that is compatible with the core pathogenesis attributes and the internal organs and meridians.

[0012] Preferably, when the physiotherapy parameter dynamic planning module executes the dynamic configuration of the output intensity ratio, action time ratio, and application time sequence of the dominant physical field and each auxiliary physical field based on the quantified syndrome intensity value, it is specifically used for: Based on the numerical range of quantitative symptom intensity values, the intensity levels of the fundamental effects corresponding to the dominant physical field are classified. Based on the basic intensity level, the duration of the dominant physical field's continuous action in the total action time is set to obtain its proportion of the action time. Based on the functional synergy between each auxiliary physical field and the dominant physical field, the intervention time point and duration of each auxiliary physical field are allocated within the duration of continuous action, forming an application time sequence; Based on the intensity level of the basic action and the synergistic effect requirements of each auxiliary physical field at the corresponding intervention time point, the relative intensity relationship between the dominant physical field and each auxiliary physical field at the same moment is determined to form the output intensity ratio.

[0013] Preferably, when the four-dimensional physiotherapy prescription generation module generates an executable personalized four-dimensional physiotherapy prescription based on the configured output intensity ratio, action time ratio, and application time sequence, it is specifically used for: Based on the output intensity ratio, the relative intensity relationship between the dominant physical field and each auxiliary physical field is transformed into the independent intensity parameters corresponding to each physical field; Based on the proportion of the action time, the proportion of the dominant physical field and each auxiliary physical field in the total duration is transformed into the independent action time of each physical field. Based on the application time sequence, the intervention order and time point of the dominant physical field and each auxiliary physical field are transformed into the independent start time and duration of each physical field. By combining independent intensity parameters, independent duration of action, and independent start time and duration of action, a personalized four-dimensional physiotherapy prescription containing all physical field execution instructions is compiled.

[0014] To address the aforementioned problems, the present invention also provides an intelligent four-dimensional eye therapy method, the method comprising: S1. Acquire the user's visible light image and periocular infrared thermal image, and collect the user's structured TCM symptom information; S2. Extract the blood vessel quantification features that characterize the morphology and distribution density of scleral blood vessels from visible light images of the eye, and extract the thermal imaging quantification features that characterize the temperature gradient of specific acupoint areas around the eye socket from infrared thermal imaging images of the periorbital area. S3. The structured TCM symptom information, blood vessel quantitative features and thermal imaging quantitative features are fused from multiple sources, and the judgment is made based on the preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value. S4. Based on the current dominant TCM syndrome type, determine one dominant physical field and at least one auxiliary physical field from the thermophysical field, low-frequency pulsed magnetic field, specific spectral light field and micro-vibration sound field. S5. Based on the quantitative syndrome intensity value, dynamically configure the output intensity ratio, action time ratio and application time sequence of the dominant physical field and each auxiliary physical field within the physiotherapy cycle. S6. Based on the configured output intensity ratio, action time ratio, and application time sequence, generate an executable personalized four-dimensional physiotherapy prescription.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Through the core design of multi-source information fusion and precise syndrome differentiation, an objective quantitative representation of TCM syndrome types and disease severity is achieved, allowing physiotherapy plans to precisely match the individual's core pathogenesis. The deep integration of structured TCM symptom information and bimodal quantitative features comprehensively captures key information about eye health status. Combined with standardized syndrome differentiation logic, it not only clarifies the core orientation of the dominant syndrome type but also accurately reflects the severity of the disease through quantitative syndrome intensity values. This provides solid objective data support for the design of physiotherapy plans, significantly improving the targeting and core efficacy of physiotherapy.

[0016] 2. Through the scientific combination of the dominant and auxiliary physical fields, a multi-dimensional complementary therapeutic effect is formed. The output intensity ratio, action time ratio, and application time sequence, dynamically optimized based on the symptom intensity value, ensure that the physiotherapy parameters can adapt to changes in the patient's condition in real time. The synergistic effect of both maximizes the therapeutic advantages of different physical fields and achieves dynamic adaptation of the physiotherapy process, making the therapeutic effect more stable and continuous, further enhancing the overall scientific level and application value of the physiotherapy. Attached Figure Description

[0017] Figure 1 A system architecture diagram of an intelligent four-dimensional eye therapy system provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating an intelligent four-dimensional eye therapy method according to an embodiment of the present invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0021] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0022] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0023] In practice, the server-side equipment deployed in an intelligent 4D eye therapy system may consist of one or more devices. This intelligent 4D eye therapy system can be implemented as: a business instance, a virtual machine, or hardware devices. For example, this intelligent 4D eye therapy system can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, this intelligent 4D eye therapy system can be understood as software deployed on a cloud node, used to provide an intelligent 4D eye therapy system to various user terminals. Alternatively, this intelligent 4D eye therapy system can also be implemented as a virtual machine deployed on one or more devices in a cloud node. This virtual machine contains application software for managing various user terminals. Alternatively, this intelligent 4D eye therapy system can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide an intelligent 4D eye therapy system to various user terminals.

[0024] In terms of implementation, the intelligent 4D eye therapy system and the user terminal are mutually compatible. That is, if the intelligent 4D eye therapy system is implemented as an application installed on a cloud service platform, then the user terminal is a client that establishes a communication connection with the application; or if the intelligent 4D eye therapy system is implemented as a website, then the user terminal is implemented as a webpage; or if the intelligent 4D eye therapy system is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0025] Example 1, such as Figure 1 The diagram shown is a system architecture diagram of an intelligent four-dimensional eye therapy system provided in an embodiment of the present invention.

[0026] This invention discloses an intelligent four-dimensional eye therapy system that can be hosted on a cloud server. In terms of implementation, it can function as one or more service devices, or as an application installed on the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed into a website. Depending on the functions implemented, an intelligent four-dimensional eye therapy system may include a multi-source data acquisition module 101, a dual-modal feature extraction module 102, an intelligent syndrome differentiation module 103, a physical field scheme decision module 104, a therapy parameter dynamic planning module 105, and a four-dimensional therapy prescription generation module 106. The modules of this invention can also be referred to as units, which are a series of computer program segments that can be executed by an electronic device's processor and perform a fixed function, stored in the electronic device's memory.

[0027] In this embodiment of the invention, in an intelligent four-dimensional eye therapy system, each of the above-mentioned modules can be implemented independently and can call other modules. Here, "calling" can be understood as one module connecting to multiple modules of another type and providing corresponding services to those connected modules. The intelligent four-dimensional eye therapy system provided by this embodiment of the invention allows for adjustment of the system's applicability by adding modules and directly calling them without modifying the program code, achieving cluster-based horizontal expansion to quickly and flexibly expand the intelligent four-dimensional eye therapy system. In practical applications, the above modules can be set in the same device or different devices, or they can be set in a virtual device, such as a service instance in a cloud server.

[0028] The following describes, with reference to specific embodiments, each component and its specific workflow of an intelligent four-dimensional eye therapy system: The multi-source data acquisition module 101 is used to acquire the user's visible light image of the eye and the infrared thermal image of the periorbital area, and to collect the user's structured TCM symptom information. The dual-modal feature extraction module 102 is used to extract blood vessel quantification features that characterize the morphology and distribution density of scleral blood vessels from visible light images of the eye, and to extract thermal imaging quantification features that characterize the temperature gradient of specific acupoint areas around the eye socket from infrared thermal imaging images of the periorbital area. The intelligent syndrome differentiation module 103 is used to fuse structured TCM symptom information, blood vessel quantitative features and thermal imaging quantitative features into multi-source information, and make a judgment based on the preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value. The physical field scheme decision module 104 is used to determine a dominant physical field and at least one auxiliary physical field from the thermophysical field, low-frequency pulsed magnetic field, specific spectral light field and micro-vibration sound field according to the current dominant TCM syndrome type. The physiotherapy parameter dynamic planning module 105 is used to dynamically configure the output intensity ratio, action time ratio and application time sequence of the dominant physical field and each auxiliary physical field according to the quantitative syndrome intensity value. The four-dimensional physiotherapy prescription generation module 106 is used to generate an executable personalized four-dimensional physiotherapy prescription based on the configured output intensity ratio, action time ratio and application time sequence.

[0029] In this embodiment, when the multi-source data acquisition module 101 acquires the user's visible light image and periocular infrared thermal image, and collects the user's structured TCM symptom information, it is specifically used for: Segmenting the sclera region in a visible light image of the eye yields a sclera region image; Identify the vascular network in the scleral region image, analyze the morphological bifurcation features of the vascular network, and obtain the quantitative features of blood vessels; The inner and outer canthal regions in the periocular infrared thermal imaging image were calibrated, and the temperature field distribution pattern between the inner and outer canthal regions was extracted to obtain the thermal imaging quantitative features. The symptom descriptions in structured TCM symptom information are analyzed, and the symptom descriptions are associated with TCM organ differentiation items to form a standard symptom feature vector.

[0030] In practice, the visible light image of the eye is converted to grayscale. By analyzing the differences in grayscale values ​​in different regions of the image, the sclera is distinguished from other tissues such as the iris, cornea, and periocular skin. A grayscale threshold-based segmentation method is adopted. Based on the characteristic that the grayscale value of the sclera region is relatively uniform and significantly higher than that of the iris and cornea, all pixels that meet the grayscale range are selected. Scattered noise pixels at the image edges are removed, and the continuously distributed sclera pixels are integrated into a complete region to obtain the sclera region image.

[0031] Furthermore, a contrast enhancement operation is performed on the scleral region image to increase the grayscale difference between the blood vessels and the scleral background. The entire scleral region image is traversed by scanning pixels row by row and column by column. Continuous pixels with grayscale values ​​lower than the grayscale values ​​of the scleral background are identified as blood vessel pixels. A complete blood vessel network structure is constructed by connecting adjacent blood vessel pixels. Then, the extension path of each blood vessel is tracked, and the number of branches, the angle between branches, and the curvature of the blood vessel direction are counted. After systematically integrating these statistical results, the blood vessel quantitative features are obtained.

[0032] Furthermore, the infrared thermal imaging images around the eyes are normalized by thermal pixel values. Based on the inherent characteristics of the physiological structure of the human eye, the feature points at both ends of the palpebral fissure are located. The feature points closer to the bridge of the nose are determined as the inner canthus region, and the feature points closer to the temporal side are determined as the outer canthus region.

[0033] Then, a linear analysis region is constructed using these two regions as endpoints. The thermal pixel values ​​of all pixels within a certain range on both sides of this linear region are extracted. The trend of change of thermal pixel values ​​along this region, the distribution density, and the location of extreme values ​​are analyzed. After summarizing these analysis results, the thermal imaging quantitative features are obtained.

[0034] Furthermore, the textual descriptions in the structured TCM symptom information are segmented to remove meaningless auxiliary words, and core symptom keywords are accurately extracted. These core symptom keywords are then compared one by one with a pre-built TCM organ differentiation database.

[0035] It should be noted that the pre-constructed TCM Zang-Fu syndrome differentiation database can be formed by collecting the correspondence between eye-related symptoms and Zang-Fu syndrome differentiation in TCM classics throughout history, combining them with syndrome differentiation cases accumulated in clinical practice, and organizing them after expert review. The corresponding TCM Zang-Fu syndrome differentiation entries are determined according to the semantic matching degree between keywords and syndrome differentiation entries. The matching results are converted into a fixed-dimensional binary representation form, with successfully matched dimensions marked as 1 and unmatched dimensions marked as 0, ultimately forming a standard symptom feature vector.

[0036] In summary, the multi-source data acquisition module 101 integrates three core data types: visible light images of the eye, infrared thermal images of the periocular region, and structured TCM symptom information. This achieves comprehensive coverage from morphological representation and thermophysiological state to TCM syndrome information. Visible light images of the eye can accurately capture the objective morphological characteristics of scleral blood vessels, infrared thermal images of the periocular region can reflect the physiological changes in temperature gradients in acupoint areas, and structured TCM symptom information quantifies the core characteristics of users' subjective symptoms. The three types of data complement each other and corroborate each other, breaking the limitations of a single data dimension and providing rich and comprehensive basic materials for subsequent multi-source information fusion.

[0037] In summary, standardized data collection and processing procedures, such as scleral region segmentation, acupoint region calibration, and structured analysis of symptom information, effectively reduced subjective errors. Structured TCM symptom information, by associating it with TCM organ-based diagnostic criteria, forms standardized feature vectors, avoiding the ambiguity of traditional symptom descriptions. Precise region extraction from image data ensures the targeted nature of feature collection, making all types of data quantifiable and comparable, providing crucial assurance for the accuracy of subsequent syndrome differentiation.

[0038] In summary, the collected multi-source data is directly integrated with the subsequent dual-modal feature extraction and intelligent syndrome differentiation module 103, achieving end-to-end adaptation between data collection, syndrome differentiation, and treatment plan generation. Differences in ocular physiological characteristics, thermal manifestations, and symptoms among different users are accurately captured through multi-source data, providing exclusive data support for personalized syndrome determination. This ensures that subsequent physical field solutions and treatment parameters accurately match the individual user's pathogenesis, guaranteeing the personalization and effectiveness of the four-dimensional physiotherapy prescription from the source, and promoting the upgrade of ocular physiotherapy from generalization to precision and intelligence. In this embodiment, when the dual-modal feature extraction module 102 extracts vascular morphology and distribution density quantification features from visible light images of the eye, it is specifically used for: Identify the vascular network in the sclera region of the eye in visible light images to obtain a vascular network structure map; By analyzing the bifurcation and terminal points of blood vessels in the vascular network structure diagram and statistically analyzing the density of bifurcation points per unit area, the characteristics of blood vessel distribution density are obtained. Assess the tortuosity and straightness characteristics of the main vessel segments in the vascular network structure diagram, quantify their tortuosity and consistency of direction, and form the morphological characteristics of blood vessels; Combining the distribution density characteristics and morphological characteristics of blood vessels, we can construct the quantitative characteristics of blood vessels.

[0039] In practice, the visible light image of the eye is converted to grayscale. By adjusting the grayscale levels of the image, the grayscale difference between blood vessels and the scleral background is enhanced. Then, a pixel-by-pixel traversal method is used to filter out all blood vessel pixels based on the characteristic that the grayscale value of blood vessel pixels is lower than that of the scleral background. Subsequently, adjacent blood vessel pixels are connected to form continuous blood vessel lines. At the same time, isolated noise pixels in the image are removed. Finally, a vascular network structure map that can completely present the direction and connection relationship of blood vessels in the scleral region is constructed.

[0040] Furthermore, the blood vessel lines in the vascular network structure diagram are traced segment by segment. When a blood vessel line branches during its extension, i.e., a blood vessel node connects three or more blood vessel segments, the node is determined as a bifurcation point. When a blood vessel line extends to its end and does not continue to extend, i.e., a blood vessel node connects only one blood vessel segment, the node is determined as an end point.

[0041] After identifying all bifurcation points and terminal points, the effective range of the scleral region in the vascular network structure diagram is determined and the area of ​​the effective range is calculated. Then, the number of all bifurcation points within the effective range is counted. By correlating the number of bifurcation points with the area of ​​the effective range, the bifurcation point density per unit area is obtained. This density result is integrated as the core information to form the blood vessel distribution density feature.

[0042] Furthermore, based on the thickness and extension length of the vascular network structure diagram, vessels with thicker diameters and longer extension distances are selected as main vessel segments. Each main vessel segment is divided into several continuous short segments using a piecewise fitting method. The curvature characteristics of the main vessel segment are evaluated by judging the angle between each short segment. When the angle between adjacent short segments is close to a flat angle, it indicates that the vessel segment has a gentle direction and a small degree of curvature. The larger the angle deviates from a flat angle, the greater the degree of curvature. Thus, the curvature of the main vessel segment is quantified.

[0043] Simultaneously, the entire course of the main blood vessel segment is tracked and the overall deviation angle is statistically analyzed. A smaller and continuous deviation angle indicates high course consistency, while a larger deviation angle indicates low consistency. The quantified curvature and course consistency results are integrated to form the blood vessel morphology characteristics.

[0044] Finally, the obtained blood vessel distribution density features and blood vessel morphology features are fused together. The density information of bifurcation points per unit area in the blood vessel distribution density features is associated and integrated with the consistency information of curvature direction in the blood vessel morphology features. This ensures that the information of the two features is matched with each other and without redundancy, and finally forms a blood vessel quantitative feature that can comprehensively characterize the morphology and distribution density of scleral blood vessels.

[0045] In this embodiment, the dual-modal feature extraction module 102, when performing the extraction of thermal imaging quantization features characterizing the temperature gradient of specific acupoint areas around the eye socket from the periorbital infrared thermal imaging image, is specifically used for: In the periocular infrared thermal imaging image, the Jingming acupoint area and the Tongziliao acupoint area were marked respectively to obtain the first acupoint area and the second acupoint area; Extract the temperature distribution in the first acupoint region and the second acupoint region, calculate the average temperature decay rate between the highest temperature point and the contour boundary in each region, and obtain the local temperature gradient characteristics of the acupoint. By comparing the temperature distribution patterns of the first and second acupoint areas, the relative positional relationship between the high-temperature core and the surrounding area is analyzed to form the temperature field pattern characteristics between acupoints. The local temperature gradient characteristics of acupoints and the temperature field pattern characteristics between acupoints constitute the quantitative features of thermal imaging.

[0046] In practice, thermal enhancement processing is performed on the infrared thermal imaging image around the eye to improve the thermal signal differences in different areas of the image. Combined with the physiological structural characteristics of the human eye, the inner and outer canthi of the palpebral fissure are located in the image. The specific area located at the inner canthus and close to the bridge of the nose is the area corresponding to the Jingming acupoint, and the specific area located at the outer canthus and close to the temporal side is the area corresponding to the Tongziliao acupoint.

[0047] Then, using the center position of the two acupoints as a reference, a certain range is evenly extended in all directions to determine the effective range containing the core area of ​​the acupoints. These two effective ranges are respectively marked as the first acupoint area and the second acupoint area.

[0048] Furthermore, the first acupoint region and the second acupoint region are scanned pixel by pixel to extract the thermal signal value corresponding to each pixel and convert it into a temperature value, thereby obtaining the complete temperature distribution of the two regions. In the temperature distribution of each region, the pixel with the highest temperature value is selected as the highest temperature point.

[0049] Simultaneously, the contour boundary of each region is determined, and the temperature value change is tracked point by point from the highest temperature point towards the contour boundary. The temperature change and corresponding distance from the highest temperature point to each point on the contour boundary are recorded. By correlating the temperature change in each direction with the corresponding distance, the temperature decay rate in each direction is obtained. Then, the average value of the temperature decay rate in all directions is taken to obtain the average temperature decay rate between the highest temperature point and the contour boundary in each region. The average temperature decay rates of the two regions are integrated to form the local temperature gradient feature of the acupoint.

[0050] Furthermore, feature extraction was performed on the temperature distribution of the first and second acupoint regions respectively to determine the temperature distribution range, shape and size of the high-temperature core in each region, forming their respective temperature distribution patterns. The temperature distribution patterns of the two regions were compared to analyze the relative positions of the high-temperature cores of the first and second acupoint regions in the entire periocular infrared thermal imaging image. At the same time, it was determined whether there was any overlap or complementarity in the temperature distribution of the areas surrounding the two high-temperature cores. These relative positional relationships and temperature distribution correlations were integrated to form the temperature field pattern features between acupoints.

[0051] Finally, the obtained local temperature gradient features of acupoints and temperature field pattern features between acupoints are correlated. The average temperature decay rate information in the local temperature gradient features of acupoints is fused with the relative positional relationship and temperature distribution correlation information in the temperature field pattern features between acupoints to ensure that the information of the two features matches each other, forming a thermal imaging quantitative feature that can comprehensively characterize the temperature gradient of a specific acupoint area around the eye socket.

[0052] In summary, through standardized and refined extraction logic, the ambiguous information in visible light and infrared thermal images of the eye is transformed into quantifiable and comparable feature data. For scleral blood vessels, by identifying the vascular network, statistically analyzing the density of bifurcation points per unit area, and quantifying the curvature and consistency of the main blood vessels, the limitations of traditional Chinese medicine's "qualitative description" of eye signs are overcome. This provides objective data support for the characterization of blood vessel morphology and distribution density, avoiding errors from subjective judgment. Regarding the temperature gradient of acupoints around the eye, by calibrating specific areas such as Jingming and Tongziliao acupoints, calculating the temperature decay rate, and analyzing the temperature field patterns between acupoints, subtle changes in the physiological thermal state of the eye are accurately captured. This transforms the temperature characteristics of acupoint areas from "fuzzy perception" to "precise quantification," significantly improving the reliability of the feature data.

[0053] In summary, the extracted quantitative features of blood vessels and the quantitative features of thermal imaging form a dual "structure-function" coverage. The quantitative features of blood vessels focus on the morphological and structural state of ocular tissues, reflecting pathological changes in the vascular network; the quantitative features of thermal imaging focus on the physiological functional state of acupoints around the eyes, reflecting the flow of Qi and blood in the internal organs and meridians. These two types of features correspond to different levels of pathological changes in the eyes, mutually corroborating and complementing each other. They capture both organic structural abnormalities and functional physiological fluctuations, comprehensively restoring the user's eye health status and avoiding the omission of pathological information by single-modal features. This provides a complete feature foundation for subsequent multi-source information fusion and diagnostic analysis.

[0054] In summary, the extracted quantitative features directly address the core requirements of the intelligent syndrome differentiation module 103 and the dynamic planning module for physiotherapy parameters 105. The quantified blood vessel features and thermal imaging features can be directly used to calculate the dispersion and variation amplitude of deviations from theoretical health benchmarks, providing precise input for quantifying syndrome intensity values. The targeted nature of the features aligns closely with the logic of TCM syndrome differentiation, making the multi-source feature association mapping more scientific and thus improving the accuracy of dominant syndrome determination. Simultaneously, the objectivity of the quantitative features ensures the comparability of feature data from different users and different periods, providing a stable reference for the subsequent dynamic adjustment of physiotherapy parameters, and guaranteeing the adaptability and effectiveness of personalized four-dimensional physiotherapy prescriptions from a core perspective.

[0055] In this embodiment, when the intelligent syndrome differentiation module 103 performs a judgment based on a preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value, it is specifically used for: Based on the correspondence between the internal organs and the eyes in traditional Chinese medicine, the symptom items in the structured TCM symptom information are mapped with the quantitative features of blood vessels and the quantitative features of thermal imaging to form a multi-source feature set with correlation weights. Based on the association weights of each feature item in the multi-source feature set, the single organ syndrome type pointed to by the feature combination with the highest weight is identified as the current dominant TCM syndrome type. The degree to which each characteristic item constituting the current dominant TCM syndrome deviates from its theoretical health benchmark is analyzed, and a weighted synthesis is performed based on their correlation weights to obtain a quantitative syndrome intensity value.

[0056] The core of this determination lies in a multi-source feature decision-making model based on correlation weights. The model determines the current dominant TCM syndrome type from the candidate syndrome type set through quantitative calculation.

[0057] In the formula, This indicates the current dominant TCM syndrome type. This indicates that the single organ syndrome pointed to by the feature combination with the highest weight has been identified. Indicates the first One candidate certificate type, This represents the set of all preset certificate types.

[0058] In the formula, This indicates the data source category index. In this invention, =3, corresponding to three types of data: This represents structured TCM symptom information. Indicates the quantitative characteristics of blood vessels. This represents the quantization characteristics of thermal imaging.

[0059] In the formula, Representing the Feature entry index in class data source, Indicates the first The total number of features contained in the class data source. These represent specific feature values, forming a "multi-source feature set". For example, This may represent the "bifurcation density" in blood vessel characteristics.

[0060] In the formula, This represents the association weight. It represents the feature. For TCM syndrome types The strength of its directionality.

[0061] In the formula, This represents a similarity or matching function that calculates feature values. With certificate type Corresponding theoretical health benchmark The degree of conformity. The higher the function value, the more the feature supports the type of evidence.

[0062] In the formula, This means calculating and summing the weighted matching degrees for all data sources and all features to obtain the proof type. The overall score.

[0063] In this embodiment, when the intelligent syndrome differentiation module 103 analyzes the degree to which each feature item constituting the current dominant TCM syndrome deviates from its theoretical health benchmark, and performs weighted summation based on its correlation weights to obtain a quantitative syndrome intensity value, it is specifically used for: Based on the quantitative characteristics of blood vessels, the dispersion range of their blood vessel distribution density and morphological straightness relative to the baseline range was determined; Based on the quantization characteristics of thermal imaging, the magnitude of the change in temperature gradient in the acupoint area relative to the baseline gradient was determined. Based on the severity level of each symptom item in the structured TCM symptom information, the intensity range of the symptom is determined. The discrete amplitude, the variation amplitude, and the symptom intensity amplitude are multiplied by their respective association weights in the multi-source feature set and then summed to obtain the quantitative syndrome intensity value.

[0064] In practice, feature association mapping is carried out based on the correspondence between the internal organs and the eyes in traditional Chinese medicine. There are fixed correspondence rules between the internal organs and the eyes in traditional Chinese medicine. For example, the liver opens into the eyes, the heart governs blood vessels and the eyes are rich in blood vessels, and the spleen governs transportation and transformation to nourish the eye tissues.

[0065] First, analyze each symptom item in the structured TCM symptom information to clarify its corresponding organ-related direction. Then, associate the blood vessel morphology and distribution density features in the blood vessel quantification features, and the acupoint area temperature gradient features in the thermal imaging quantification features, with the aforementioned symptom items respectively.

[0066] Specifically, the association process is based on the pathological association logic between symptoms and ocular signs in traditional Chinese medicine theory. For example, the symptom of dry eyes is often related to liver yin deficiency. Therefore, the abnormal density of bifurcation points in the blood vessel quantification feature and the abnormal temperature attenuation feature of the inner canthus region in the thermal imaging quantification feature are prioritized for association.

[0067] Specifically, the association weights are determined by pre-setting. The pre-setting method can be to collect a large number of clinical cases of eye diseases in traditional Chinese medicine, count the frequency of different symptom items and blood vessel quantitative features and thermal imaging quantitative features pointing to the syndromes of each organ, and then organize traditional Chinese medicine experts to conduct demonstration and adjustment of the statistical results in conjunction with theoretical classics, and finally determine the association weight between each symptom item and blood vessel quantitative features and thermal imaging quantitative features.

[0068] Finally, the associated symptom entries, blood vessel quantitative features, thermal imaging quantitative features, and their corresponding association weights are integrated to form a multi-source feature set.

[0069] Furthermore, based on the association weights of each feature item in the multi-source feature set, the dominant certificate type is identified. Specifically, the source of each parameter in the formula is first clarified, and the candidate certificate type set is determined. The syndromes related to the internal organs in traditional Chinese medicine classics were compiled and combined with the syndromes corresponding to common eye diseases in clinical practice. The results were determined after screening and verification by traditional Chinese medicine experts.

[0070] Specifically, association weight That is, the numerical values ​​obtained through clinical case statistics and expert consultation in the preceding steps, specifically the characteristic values. This represents the specific characterization results of each feature item in the multi-source feature set. Theoretical health benchmark. By collecting structured TCM symptom information, blood vessel quantitative characteristics, and thermal imaging quantitative characteristics from a large number of healthy individuals, statistical analysis was conducted to determine the normal range of each characteristic under healthy conditions.

[0071] Specifically, the matching degree function The application method is to use specific feature values Corresponding certificate type Theoretical health benchmark To make a comparison, if If the function value is within the theoretical health baseline range, the function value will be lower. When the value deviates from the theoretical health benchmark range, the function value increases with the degree of deviation.

[0072] Specifically, by calculating the weighted sum of matching degrees for each candidate certificate type, the certificate type that best matches the current multi-source features is selected. The formula trend is based on the correlation weight. The higher the specific eigenvalue Compared with theoretical health benchmarks The higher the matching degree, the more relevant the feature is to the evidence type. The greater the weighted contribution, the higher the total score.

[0073] Specifically, the calculation is performed first for each feature entry of each type of data source. and The product of all features from all data sources is then summed to obtain each candidate certificate type. Based on the total score, the candidate certificate with the highest total score will be selected. As the dominant TCM syndrome type at present.

[0074] Furthermore, based on the quantitative characteristics of blood vessels that constitute the current dominant TCM syndrome types, and using their theoretical health benchmarks as a reference, the actual characterization results of blood vessel distribution density and morphological straightness are compared to determine the dispersion range of the actual results relative to the benchmark range.

[0075] It should be noted that the dispersion amplitude is determined by judging whether the actual blood vessel distribution density exceeds the upper and lower limits of the benchmark range, and whether the direction of the morphological straightness deviates from the normal direction of the benchmark. The deviation of both constitutes the dispersion amplitude.

[0076] Furthermore, based on the thermal imaging quantitative characteristics that constitute the current dominant TCM syndrome types, the actual results of the temperature gradient in the acupoint area are compared with the baseline gradient to determine the magnitude of the change in the actual results relative to the baseline gradient.

[0077] It should be noted that the method for determining the magnitude of change is to track the changing trend of the actual temperature gradient, compare it with the changing trend of the reference gradient, and determine the direction and degree of deviation between the actual trend and the reference trend, thereby determining the magnitude of change.

[0078] Furthermore, for each symptom item in the structured TCM symptom information that constitutes the current dominant TCM syndrome type, the severity level is determined based on the degree of manifestation reflected in the symptom description. For example, a description of continuous onset indicates a higher severity level, while a description of occasional onset indicates a lower severity level. The corresponding symptom intensity range is then determined based on the severity level; the higher the severity level, the greater the symptom intensity range.

[0079] Finally, the obtained discrete amplitude, variation amplitude, and symptom intensity amplitude are multiplied by their respective association weights in the multi-source feature set to obtain a weighted value for each amplitude. All weighted values ​​are then summed to obtain the quantitative symptom intensity value.

[0080] In summary, the deep integration of multi-source information involves mapping structured TCM symptom information, blood vessel quantitative features, and thermal imaging quantitative features according to the correspondence between TCM organs and the eyes, forming a multi-source feature set with associated weights. This approach takes into account both the user's subjective symptoms and integrates objective images and physiological features, avoiding diagnostic biases caused by a single data dimension, and making the syndrome differentiation more comprehensive and convincing.

[0081] In summary, based on a pre-defined syndrome classification relationship and a multi-source feature decision model, the system accurately identifies a single dominant TCM syndrome by selecting the feature combination with the highest weight through weighted matching degree calculation. This eliminates the reliance on experience in traditional syndrome differentiation and reduces human error. Simultaneously, by quantifying the deviation of each feature from the theoretical health benchmark and combining this with correlation weights to obtain a syndrome intensity value, the abstract syndrome is transformed into a quantifiable numerical value, making the severity of the pathogenesis intuitively assessable and providing a precise basis for the tiered adjustment of subsequent treatment plans.

[0082] In summary, the dominant syndrome type and quantitative syndrome intensity values ​​output by the module directly connect to the physical field scheme decision-making and physiotherapy parameter dynamic planning module 105, achieving seamless integration between syndrome differentiation results and physiotherapy plans. Its quantitative output ensures the comparability of syndrome differentiation results for different users and different physiotherapy cycles, laying a core foundation for the generation of personalized and dynamic four-dimensional physiotherapy prescriptions and promoting the upgrade of eye physiotherapy from "experience-based conditioning" to "precision syndrome differentiation and treatment".

[0083] In this embodiment, when the physical field scheme decision module 104 determines a dominant physical field and at least one auxiliary physical field from the thermophysical field, low-frequency pulsed magnetic field, specific spectral light field, and micro-vibration sound field based on the current dominant TCM syndrome type, it is specifically used for: Analyze the core pathogenesis attributes indicated by the current dominant TCM syndrome types and classify them into either the attribute of excessive Yang or the attribute of deficiency. For the yang excess attribute, a wavelength band with a clearing and draining effect in a specific spectral light field is selected as the dominant physical field; for the deficiency attribute, a thermophysical field is selected as the dominant physical field. Based on the viscera and meridian pathways associated with the core pathogenesis attributes, at least one auxiliary physical field is selected from the micro-vibration sound field and low-frequency pulse magnetic field that complements the action site of the dominant physical field. Combine the dominant physical field with at least one auxiliary physical field to form a physical field combination scheme that is compatible with the core pathogenesis attributes and the internal organs and meridians.

[0084] In practice, the core pathogenesis attributes indicated by the dominant TCM syndrome type are analyzed. First, the correspondence between the typical pathological manifestations of the syndrome type and the TCM pathogenesis theory is sorted out. The essential attributes of the core pathogenesis are then deduced from the pathological manifestations. If the pathological manifestations of the syndrome type conform to the core characteristics of Yang hyperactivity, that is, excessive Yang Qi in the body leads to hyperactivity of function, then it is judged to be of the Yang hyperactivity attribute. If the pathological manifestations of the syndrome type conform to the core characteristics of deficiency and depletion, that is, insufficient Zheng Qi in the body leads to decreased function, then it is judged to be of the deficiency and depletion attribute.

[0085] Furthermore, targeting the excessive Yang attribute, based on the TCM principle of clearing heat and eliminating turbidity, a wavelength of light field with a clearing and eliminating effect was selected from a specific spectral light field as the dominant physical field. The selection criteria were that the wavelength of light field could inhibit the excessive Yang function by acting on the eye tissue, which met the pathogenesis regulation needs of the excessive Yang attribute.

[0086] Furthermore, considering the deficiency-related characteristics, and based on the TCM principle of warming yang and replenishing qi, a thermophysical field was selected as the dominant physical field. This is because the thermophysical field can nourish the vital energy of the eyes and related organs through heat conduction, improve the state of functional decline, and match the treatment requirements for the deficiency-related characteristics.

[0087] Furthermore, based on the viscera and meridian pathways associated with the core pathogenesis attributes, we first clarify the specific pathways and key areas of action of the viscera and meridians in the eyes and throughout the body, then analyze the range and target points of the dominant physical field, and determine the meridian pathways not covered by the dominant physical field or the auxiliary areas that need to be strengthened and regulated.

[0088] Furthermore, from the micro-vibration sound field and the low-frequency pulse magnetic field, a physical field whose action area can cover the above-mentioned auxiliary parts is selected as an auxiliary physical field to ensure that the action area of ​​the auxiliary physical field complements the action area of ​​the dominant physical field, thereby achieving comprehensive regulation of the meridian pathways of the internal organs.

[0089] Finally, the selected dominant physical field is combined with at least one auxiliary physical field. During the combination process, the order of action and the synergistic logic of each physical field are clarified to ensure that the dominant physical field plays a major regulatory role in the core pathogenesis, while the auxiliary physical field plays a supplementary regulatory role in the related organs and meridians. In the end, a physical field combination scheme that is compatible with the core pathogenesis attributes and organs and meridians is formed.

[0090] In summary, based on the current dominant TCM syndrome types, the dominant physical field is selected by analyzing the core pathogenesis attributes: a specific spectral light field with a clearing and purging effect is suitable for those with excessive Yang, and a thermophysical field with warming and nourishing effects is suitable for those with deficiency. This is in line with the core logic of TCM "differentiation of syndromes and treatment", ensuring that the dominant physical field directly targets the core pathogenesis and avoiding the problem of inefficient treatment caused by the physical field not matching the pathogenesis.

[0091] In summary, the module does not select a single physical field, but rather chooses auxiliary physical fields that complement the areas of action of the dominant physical field, based on the pathways of the internal organs and meridians associated with the core pathogenesis, forming a synergistic combination of "dominant + auxiliary". This combination ensures targeted treatment of the core pathogenesis while achieving comprehensive coverage of the areas of action of the internal organs and meridians, thus overcoming the limitations of a single physical field in terms of its limited range and singular treatment dimension, making the therapeutic effect more holistic.

[0092] In summary, the module receives the output from the intelligent syndrome differentiation module 103, transforming abstract TCM syndrome types into concrete and executable physical field combination schemes, providing a clear adjustment target for the subsequent dynamic planning module 105 for physiotherapy parameters. Its decision-making logic is deeply integrated with TCM theory and physical field characteristics, ensuring a seamless connection from syndrome type determination to physiotherapy implementation. Simultaneously, it tailors exclusive physical field combinations for each user, maximizing the scientific rigor and adaptability of personalized physiotherapy, and significantly improving the accuracy and effectiveness of eye physiotherapy.

[0093] In this embodiment, when the physiotherapy parameter dynamic planning module 105 executes the dynamic configuration of the output intensity ratio, action time ratio, and application time sequence of the dominant physical field and each auxiliary physical field based on the quantified syndrome intensity value, it is specifically used for: Based on the numerical range of quantitative symptom intensity values, the intensity levels of the fundamental effects corresponding to the dominant physical field are classified. Based on the basic intensity level, the duration of the dominant physical field's continuous action in the total action time is set to obtain its proportion of the action time. Based on the functional synergy between each auxiliary physical field and the dominant physical field, the intervention time point and duration of each auxiliary physical field are allocated within the duration of continuous action, forming an application time sequence; Based on the intensity level of the basic action and the synergistic effect requirements of each auxiliary physical field at the corresponding intervention time point, the relative intensity relationship between the dominant physical field and each auxiliary physical field at the same moment is determined to form the output intensity ratio.

[0094] In practice, the intensity level of the basic action corresponding to the dominant physical field is divided according to the numerical range of the quantitative syndrome intensity value. First, the correspondence between the quantitative syndrome intensity value and the severity of the pathogenesis is clarified. The higher the syndrome intensity value, the more severe the pathogenesis, and the higher the intensity of physical field conditioning required.

[0095] Furthermore, by analyzing the effective conditioning intensity data corresponding to different quantitative syndrome intensity values ​​in a large number of clinical physiotherapy cases, and combining the gradient principle of TCM pathogenesis conditioning, several continuous and non-overlapping numerical intervals were divided. Each numerical interval corresponds to a fixed basic effect intensity level. The basic effect intensity level of the dominant physical field is determined as the level corresponding to the interval in which the quantitative syndrome intensity value falls.

[0096] Furthermore, based on the basic intensity level, the duration of the dominant physical field's continuous action in the total action time is set to obtain the proportion of action time. First, it is clarified that the total action time is the complete duration of a single physiotherapy session.

[0097] It should be noted that there is a positive correlation between the basic action intensity level and the duration of the action. The higher the level, the longer the dominant physical field needs to take effect. Based on the conditioning requirements corresponding to the basic action intensity level, the duration of the dominant physical field's action is determined. This duration is then correlated with the total action duration to obtain the proportion of the dominant physical field in the total action duration, i.e., the action time proportion.

[0098] Furthermore, intervention time points and durations are allocated based on the functional synergy between each auxiliary physical field and the dominant physical field to form an application time sequence. First, it is clarified that the functional synergy relationship is that the auxiliary physical field needs to enhance the conditioning effect of the dominant physical field or make up for the limitations of the dominant physical field's role.

[0099] Furthermore, within the duration of the dominant physical field's continuous action, the timing of the intervention of auxiliary physical fields is planned according to the conditioning logic to ensure that the intervention time is within the effective stage of the dominant physical field's action. Then, the duration of each auxiliary physical field is determined according to the synergistic requirements. The order of action, intervention time, and duration of all physical fields are arranged in chronological order to form an application time sequence.

[0100] Furthermore, the relative intensity relationship is determined based on the basic action intensity level and the synergistic action requirements of each auxiliary physical field at the corresponding intervention time point to form the output intensity ratio. First, the benchmark action intensity of the dominant physical field at that level is determined according to the basic action intensity level.

[0101] Finally, when each auxiliary physical field intervenes, the required intensity of the auxiliary physical field is determined based on the synergistic effect requirement. If the synergistic requirement is to enhance the dominant effect, the intensity of the auxiliary physical field is kept in a suitable ratio with the intensity of the dominant physical field. If the synergistic requirement is to supplement and regulate, the intensity of the auxiliary physical field is set to a suitable supplementary intensity. The intensity relationship between the dominant physical field and each auxiliary physical field at the same time is clarified, and these relationships are integrated to form the output intensity ratio.

[0102] In summary, the dynamic programming module 105 for physiotherapy parameters uses the quantification of syndrome intensity as its core basis. By dividing numerical ranges, it determines the basic intensity level of the dominant physical field, directly linking the physiotherapy intensity to the severity of the user's condition. The higher the syndrome intensity, the more reasonable the ratio of basic intensity level to duration of action. This avoids both inefficiency due to insufficient intensity and discomfort caused by excessive intensity, achieving precise adaptation of "differentiated dosage" and ensuring that parameter configuration fits individual differences.

[0103] In summary, the dynamic planning module 105 for physiotherapy parameters fully considers the functional synergy between the dominant and auxiliary physical fields, scientifically allocating the intervention time, duration, and intensity ratio of the auxiliary physical field. By optimizing the application time sequence, it ensures that the auxiliary physical field intervenes at the key stages of the dominant physical field's action, forming a complementary and enhancing effect; at the same time, it clarifies the relative intensity relationship at the same moment, avoiding conflicts between the actions of different physical fields, maximizing the synergistic therapeutic value of "dominant + auxiliary," and improving the overall effect of physiotherapy.

[0104] In summary, the dynamic programming module 105 for physiotherapy parameters transforms the abstract intensity of symptoms into concrete and executable parameter configurations, providing standardized and regulated input for the four-dimensional physiotherapy prescription generation module 106. Its dynamic configuration logic ensures that parameters can be adjusted in real time according to changes in the intensity of the user's symptoms, achieving dynamic adaptation throughout the entire physiotherapy cycle. This upgrades personalized physiotherapy from a "static plan" to "dynamic optimization," further solidifying the scientific closed loop of the entire system from diagnosis to treatment.

[0105] In this embodiment, when the four-dimensional physiotherapy prescription generation module 106 generates an executable personalized four-dimensional physiotherapy prescription based on the configured output intensity ratio, action time ratio, and application time sequence, it is specifically used for: Based on the output intensity ratio, the relative intensity relationship between the dominant physical field and each auxiliary physical field is transformed into the independent intensity parameters corresponding to each physical field; Based on the proportion of the action time, the proportion of the dominant physical field and each auxiliary physical field in the total duration is transformed into the independent action time of each physical field. Based on the application time sequence, the intervention order and time point of the dominant physical field and each auxiliary physical field are transformed into the independent start time and duration of each physical field. By combining independent intensity parameters, independent duration of action, and independent start time and duration of action, a personalized four-dimensional physiotherapy prescription containing all physical field execution instructions is compiled.

[0106] In practice, the relative intensity relationship between the dominant physical field and each auxiliary physical field is transformed into an independent intensity parameter for each physical field based on the output intensity ratio. First, the intensity correlation logic between the dominant and auxiliary physical fields, as reflected in the output intensity ratio, is clarified. Using the intensity of the dominant physical field as a benchmark, the intensity values ​​of each auxiliary physical field relative to the dominant physical field are derived according to the proportion of each physical field in the output intensity ratio.

[0107] By transforming this relative intensity relationship into a specific and executable intensity representation, each physical field has a clear and independent intensity standard, ultimately yielding the independent intensity parameters corresponding to each physical field.

[0108] Furthermore, based on the proportion of action time, the proportion of the dominant physical field and each auxiliary physical field in the total duration is transformed into the independent action time of each physical field, thus clarifying the range of the total duration of a single physiotherapy session.

[0109] Furthermore, the time share that the dominant physical field should occupy in the total duration is determined based on the proportion of the action time, and the independent action duration of the dominant physical field is derived by the correlation between the total duration and this share.

[0110] Then, based on the corresponding share of each auxiliary physical field in the duration of action, the independent duration of each auxiliary physical field is derived with reference to the total duration, ensuring that the sum of the independent durations of each physical field matches the total duration.

[0111] Furthermore, based on the application time sequence, the intervention order and time point of the dominant physical field and each auxiliary physical field are transformed into the independent start time and duration of each physical field, and the clear temporal logic in the application time sequence is sorted out first.

[0112] Furthermore, taking the start time of physiotherapy as the starting point, the intervention time of the dominant physical field is determined as its independent activation time, and the continuous action period corresponding to this activation time is its duration. Then, according to the intervention order of each auxiliary physical field relative to the dominant physical field in the application time sequence, the intervention time of each auxiliary physical field is determined as its independent activation time, and the continuous action period of each auxiliary physical field from activation to termination is defined as its duration.

[0113] Furthermore, the independent intensity parameters, independent duration of action, independent start time, and duration of duration are combined to form a personalized four-dimensional physiotherapy prescription containing all physical field execution instructions. First, the independent intensity parameters, independent duration of action, independent start time, and duration of duration corresponding to each physical field are associated and matched to ensure that the parameters of the same physical field correspond to each other.

[0114] Then, according to the time sequence of the physiotherapy process, all the associated parameters of the physical field are organized into ordered execution instructions, with each execution instruction clearly corresponding to the start time, duration and intensity parameters of a physical field.

[0115] Finally, these orderly execution instruction systems are integrated to form a complete and logically clear personalized four-dimensional physiotherapy prescription, which can be directly used as the basis for physiotherapy execution.

[0116] In summary, the four-dimensional physiotherapy prescription generation module 106 transforms the output intensity ratio, action time ratio, and application time sequence output by the physiotherapy parameter dynamic planning module 105 into specific executable instructions such as independent intensity parameters, action duration, and start time for each physical field. By clearly defining the quantitative execution standards for each physical field, it avoids equipment operation deviations caused by parameter ambiguity, enabling the physiotherapy plan to be directly and accurately implemented with the physiotherapy equipment. This solves the core pain point of "disconnect between plan design and actual execution," truly transforming the results of intelligent diagnosis and parameter planning into implementable physiotherapy actions.

[0117] In summary, the prescriptions generated by the 4D physiotherapy prescription generation module 106 are entirely based on the user's unique syndrome type and intensity, integrating the synergistic configuration logic of the "dominant + auxiliary" physical fields. The intensity parameters, duration of action, and activation sequence of each prescription are tailored to the individual pathogenesis characteristics of the user, ensuring both the targeted effect of the dominant physical field on the core pathogenesis and the synergistic supplementary effect of the auxiliary physical field. This achieves deep personalization, avoiding the problems of poor adaptability and ineffective treatment of generic prescriptions.

[0118] In summary, the 4D physiotherapy prescription generation module 106, as the terminal output of the entire system, inherits the core achievements of all preceding modules. It condenses the information from the entire process of multi-source data acquisition, feature extraction, syndrome differentiation, physical field decision-making, and parameter planning into a structured and standardized physiotherapy prescription. Its clear output instructions not only provide a clear basis for a single physiotherapy session but also reserve traceable and comparable basic data for dynamic adjustments in subsequent physiotherapy cycles. It constructs a complete intelligent physiotherapy closed loop of "data acquisition - syndrome differentiation decision-making - plan execution - effect feedback," promoting a comprehensive upgrade of eye physiotherapy from "experience-based operation" to "standardized intelligent treatment."

[0119] Example 2, refer to Figure 2 The diagram shown is a flowchart illustrating an intelligent four-dimensional eye therapy method according to an embodiment of the present invention. In this embodiment, the intelligent four-dimensional eye therapy method includes: S1. Acquire the user's visible light image and periocular infrared thermal image, and collect the user's structured TCM symptom information; S2. Extract the blood vessel quantification features that characterize the morphology and distribution density of scleral blood vessels from visible light images of the eye, and extract the thermal imaging quantification features that characterize the temperature gradient of specific acupoint areas around the eye socket from infrared thermal imaging images of the periorbital area. S3. The structured TCM symptom information, blood vessel quantitative features and thermal imaging quantitative features are fused from multiple sources, and the judgment is made based on the preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value. S4. Based on the current dominant TCM syndrome type, determine one dominant physical field and at least one auxiliary physical field from the thermophysical field, low-frequency pulsed magnetic field, specific spectral light field and micro-vibration sound field. S5. Based on the quantitative syndrome intensity value, dynamically configure the output intensity ratio, action time ratio and application time sequence of the dominant physical field and each auxiliary physical field within the physiotherapy cycle. S6. Based on the configured output intensity ratio, action time ratio, and application time sequence, generate an executable personalized four-dimensional physiotherapy prescription.

[0120] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0121] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent four-dimensional eye therapy system, characterized in that, The system includes a multi-source data acquisition module, a dual-modal feature extraction module, an intelligent syndrome differentiation module, a physical field scheme decision-making module, a physiotherapy parameter dynamic planning module, and a four-dimensional physiotherapy prescription generation module, wherein: The multi-source data acquisition module is used to acquire visible light images and periocular infrared thermal images of the user's eyes, and to collect the user's structured TCM symptom information; The dual-modal feature extraction module is used to extract blood vessel quantification features that characterize the morphology and distribution density of scleral blood vessels from visible light images of the eye, and to extract thermal imaging quantification features that characterize the temperature gradient of specific acupoint areas around the eye socket from infrared thermal imaging images of the periorbital eye. The intelligent syndrome differentiation module is used to fuse structured TCM symptom information, blood vessel quantitative features and thermal imaging quantitative features into multi-source information, and make a judgment based on the preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value. The physical field scheme decision module is used to determine a dominant physical field and at least one auxiliary physical field from thermophysical fields, low-frequency pulsed magnetic fields, specific spectral light fields, and micro-vibration sound fields based on the current dominant TCM syndrome type. The dynamic planning module for physiotherapy parameters is used to dynamically configure the output intensity ratio, action time ratio, and application time sequence of the dominant physical field and each auxiliary physical field according to the quantitative syndrome intensity value. The four-dimensional physiotherapy prescription generation module is used to generate executable, personalized four-dimensional physiotherapy prescriptions based on the configured output intensity ratio, action time ratio, and application time sequence.

2. The intelligent four-dimensional eye therapy system as described in claim 1, characterized in that, When the multi-source data acquisition module acquires the user's visible light image and periocular infrared thermal image, and collects the user's structured TCM symptom information, it is specifically used for: Segmenting the sclera region in a visible light image of the eye yields a sclera region image; Identify the vascular network in the scleral region image, analyze the morphological bifurcation features of the vascular network, and obtain the quantitative features of blood vessels; The inner and outer canthal regions in the periocular infrared thermal imaging image were calibrated, and the temperature field distribution pattern between the inner and outer canthal regions was extracted to obtain the thermal imaging quantitative features. The symptom descriptions in structured TCM symptom information are analyzed, and the symptom descriptions are associated with TCM organ differentiation items to form a standard symptom feature vector.

3. The intelligent four-dimensional eye therapy system as described in claim 1, characterized in that, The dual-modal feature extraction module, when performing the extraction of vascular quantification features characterizing the morphology and distribution density of scleral vascular morphology from visible light images of the eye, is specifically used for: Identify the vascular network in the sclera region of the eye in visible light images to obtain a vascular network structure map; By analyzing the bifurcation and terminal points of blood vessels in the vascular network structure diagram and statistically analyzing the density of bifurcation points per unit area, the characteristics of blood vessel distribution density are obtained. Assess the tortuosity and straightness characteristics of the main vessel segments in the vascular network structure diagram, quantify their tortuosity and consistency of direction, and form the morphological characteristics of blood vessels; Combining the distribution density characteristics and morphological characteristics of blood vessels, we can construct the quantitative characteristics of blood vessels.

4. The intelligent four-dimensional eye therapy system as described in claim 1, characterized in that, The dual-modal feature extraction module, when performing thermal imaging quantization features characterizing the temperature gradient of specific acupoint areas around the eye socket from periorbital infrared thermal imaging images, is specifically used for: In the periocular infrared thermal imaging image, the Jingming acupoint area and the Tongziliao acupoint area were marked respectively to obtain the first acupoint area and the second acupoint area; Extract the temperature distribution in the first acupoint region and the second acupoint region, calculate the average temperature decay rate between the highest temperature point and the contour boundary in each region, and obtain the local temperature gradient characteristics of the acupoint. By comparing the temperature distribution patterns of the first and second acupoint areas, the relative positional relationship between the high-temperature core and the surrounding area is analyzed to form the temperature field pattern characteristics between acupoints. The local temperature gradient characteristics of acupoints and the temperature field pattern characteristics between acupoints constitute the quantitative features of thermal imaging.

5. The intelligent four-dimensional eye therapy system as described in claim 1, characterized in that, When the intelligent syndrome differentiation module performs a judgment based on a preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value, it is specifically used for: Based on the correspondence between the internal organs and the eyes in traditional Chinese medicine, the symptom items in the structured TCM symptom information are mapped with the quantitative features of blood vessels and the quantitative features of thermal imaging to form a multi-source feature set with correlation weights. Based on the association weights of each feature item in the multi-source feature set, the single organ syndrome type pointed to by the feature combination with the highest weight is identified as the current dominant TCM syndrome type. The degree to which each characteristic item constituting the current dominant TCM syndrome deviates from its theoretical health benchmark is analyzed, and a weighted synthesis is performed based on their correlation weights to obtain a quantitative syndrome intensity value.

6. The intelligent four-dimensional eye therapy system as described in claim 5, characterized in that, The intelligent syndrome differentiation module, when analyzing the degree to which each feature item constituting the current dominant TCM syndrome deviates from its theoretical health benchmark, and performing weighted synthesis based on their correlation weights to obtain a quantitative syndrome intensity value, is specifically used for: Based on the quantitative characteristics of blood vessels, the dispersion of their blood vessel distribution density and morphological straightness relative to the baseline range was determined; Based on the quantitative characteristics of thermal imaging, the magnitude of the change in temperature gradient in the acupoint area relative to the baseline gradient was determined. Based on the severity level of each symptom item in the structured TCM symptom information, the intensity range of the symptom is determined. The discrete amplitude, the variation amplitude, and the symptom intensity amplitude are multiplied by their respective association weights in the multi-source feature set and then summed to obtain the quantitative syndrome intensity value.

7. The intelligent four-dimensional eye therapy system as described in claim 1, characterized in that, When the physical field scheme decision module determines a dominant physical field and at least one auxiliary physical field from the thermophysical field, low-frequency pulsed magnetic field, specific spectral light field, and micro-vibration sound field based on the current dominant TCM syndrome type, it is specifically used for: Analyze the core pathogenesis attributes indicated by the current dominant TCM syndrome types and classify them into either the attribute of excessive Yang or the attribute of deficiency. For the yang excess attribute, a wavelength band with a clearing and draining effect in a specific spectral light field is selected as the dominant physical field; for the deficiency attribute, a thermophysical field is selected as the dominant physical field. Based on the viscera and meridian pathways associated with the core pathogenesis attributes, at least one auxiliary physical field is selected from the micro-vibration sound field and low-frequency pulse magnetic field that complements the action site of the dominant physical field. Combine the dominant physical field with at least one auxiliary physical field to form a physical field combination scheme that is compatible with the core pathogenesis attributes and the internal organs and meridians.

8. The intelligent four-dimensional eye therapy system as described in claim 1, characterized in that, The dynamic programming module for physiotherapy parameters, when executing based on quantified syndrome intensity values ​​and dynamically configuring the output intensity ratio, duration of action, and application time sequence of the dominant and auxiliary physical fields within the physiotherapy cycle, is specifically used for: Based on the numerical range of quantitative symptom intensity values, the intensity levels of the fundamental effects corresponding to the dominant physical field are classified. Based on the basic intensity level, the duration of the dominant physical field's continuous action in the total action time is set to obtain its proportion of the action time. Based on the functional synergy between each auxiliary physical field and the dominant physical field, the intervention time point and duration of each auxiliary physical field are allocated within the duration of continuous action, forming an application time sequence; Based on the intensity level of the basic action and the synergistic effect requirements of each auxiliary physical field at the corresponding intervention time point, the relative intensity relationship between the dominant physical field and each auxiliary physical field at the same moment is determined to form the output intensity ratio.

9. The intelligent four-dimensional eye therapy system as described in claim 1, characterized in that, The four-dimensional physiotherapy prescription generation module, when generating an executable personalized four-dimensional physiotherapy prescription based on the configured output intensity ratio, action time ratio, and application time sequence, is specifically used for: Based on the output intensity ratio, the relative intensity relationship between the dominant physical field and each auxiliary physical field is transformed into the independent intensity parameters corresponding to each physical field; Based on the proportion of the action time, the proportion of the dominant physical field and each auxiliary physical field in the total duration is transformed into the independent action time of each physical field. Based on the application time sequence, the intervention order and time point of the dominant physical field and each auxiliary physical field are transformed into the independent start time and duration of each physical field. By combining independent intensity parameters, independent duration of action, and independent start time and duration of action, a personalized four-dimensional physiotherapy prescription containing all physical field execution instructions is compiled.

10. An intelligent four-dimensional eye therapy method, applied to the intelligent thinking eye therapy method described in any one of claims 1-9, characterized in that, The method includes: S1. Acquire the user's visible light image and periocular infrared thermal image, and collect the user's structured TCM symptom information; S2. Extract the blood vessel quantification features that characterize the morphology and distribution density of scleral blood vessels from visible light images of the eye, and extract the thermal imaging quantification features that characterize the temperature gradient of specific acupoint areas around the eye socket from infrared thermal imaging images of the periorbital area. S3. The structured TCM symptom information, blood vessel quantitative features and thermal imaging quantitative features are fused from multiple sources, and the judgment is made based on the preset syndrome classification relationship to obtain the current dominant TCM syndrome type and its corresponding quantitative syndrome intensity value. S4. Based on the current dominant TCM syndrome type, determine one dominant physical field and at least one auxiliary physical field from the thermophysical field, low-frequency pulsed magnetic field, specific spectral light field and micro-vibration sound field. S5. Based on the quantitative syndrome intensity value, dynamically configure the output intensity ratio, action time ratio and application time sequence of the dominant physical field and each auxiliary physical field within the physiotherapy cycle. S6. Based on the configured output intensity ratio, action time ratio, and application time sequence, generate an executable personalized four-dimensional physiotherapy prescription.

Images