Remote management method and system based on video image processing

Abstract

A remote management method and system based on video image processing includes: constructing a dynamic disease tracking tree based on a disease course template tree and interactive information; in the dynamic disease tracking tree, nodes are used to describe treatment sub-steps, where sub-steps are image data or staged diagnostic results, and edges are used to describe the associations between sub-steps; determining the black, white, and gray attributes of each node based on preliminary identification results; and determining the traceability path of each node based on these attributes in combination with a depth-first search algorithm; wherein gray nodes are stored in a cache module; and black nodes are stored in an associative database with referential integrity constraints. This invention creatively utilizes a depth-first search algorithm in combination with a dynamic disease tracking tree to address the issues of expert identification delays and the rapid manipulation of stored data.

Description

Technical Field

[0001] The present invention belongs to the technical field of scalp hair loss treatment, and in particular relates to a remote management method and system based on video image processing. Background Art

[0002] With the increasing popularity of telemedicine and the development of 5G networks and high-definition cameras, remote management systems based on video image processing are becoming an increasingly important supplement to medical services. This system uses cameras to capture videos of patients' scalps and, through real-time image processing and data transmission, enables doctors to quickly diagnose hair loss and dynamically adjust treatment plans. The system not only overcomes geographical limitations and enables cross-regional sharing of medical resources, but also uses algorithms such as image enhancement and feature extraction to conduct detailed analysis of scalp conditions, providing patients with personalized, fully monitored treatment services. Traditional treatments relied on physical visits, making continuous monitoring difficult. The widespread application of this remote scalp and hair loss treatment management system is expected to significantly improve diagnostic accuracy and treatment efficiency, providing strong support for the in-depth expansion of telemedicine in areas such as skin diseases and hair loss.

[0003] Currently, remote scalp hair loss management primarily uses webcams to capture video images and then uses preprocessing (such as noise filtering and histogram equalization) to improve image quality (image data clarity accounts for approximately 10% of cases). For example, adaptive thresholding algorithms (with a threshold T set to dynamically adjust between 50 and 150) are commonly used for image processing, but these algorithms are still affected by ambient lighting and equipment variations, resulting in loss of image detail. A more prominent issue is data backtracking. During treatment monitoring, doctors rely on historical video data for comparison and assessment of patients' hair loss progression. However, data storage standards are currently inconsistent across remote platforms, and the collected time-series data lacks a unified index, resulting in data fragmentation or discontinuous timestamps, making it difficult to fully track the patient's condition. Some solutions attempt to use time series models (such as tracking algorithms based on state-space models) to compensate for data loss, but due to missing information between video frames and algorithmic errors, accurate backtracking is difficult to achieve. Overall, while certain data preprocessing and backtracking mechanisms exist, a mature and reliable closed-loop management system for the storage, standardization, and time-series analysis of historical image data has yet to be established, directly impacting doctors' ability to continuously track and evaluate the patient's treatment progress. Summary of the Invention

[0004] In order to address the deficiencies in the prior art, the present invention aims to solve the above-mentioned defects and further propose a remote management method and system based on video image processing.

[0005] The present invention adopts the following technical solutions.

[0006] The first aspect of the present invention discloses a remote management method based on video image processing, comprising: constructing a dynamic disease course tracking tree based on a disease course template tree and interactive information; in the dynamic disease course tracking tree: nodes are used to describe sub-links of treatment, sub-links are image data or stage-by-stage diagnostic results, and edges are used to describe the association between sub-links; based on preliminary identification results, the black, white and gray attributes of each node are determined, and based on the black, white and gray attributes of each node, a depth-first search algorithm is combined to determine the tracing path of each node; wherein, gray nodes are stored in a cache module; black nodes are stored in an associative database in a reference integrity constraint manner.

[0007] Specifically, gray nodes are not stored in the relational database in a referential integrity constraint manner.

[0008] Specifically, if the final identification result of the gray node is consistent with the preliminary identification result, the gray node is set to black and the gray node is cleared from the cache module.

[0009] Specifically, if the first node corresponds to image data and the preliminary identification result is unreliable, a second node is generated and sent; if the preliminary identification result of the second node is credible, the second node is stored in the associative database in the form of a reference integrity constraint; based on the depth-first search algorithm, the gray node is traced back and the cache corresponding to the gray node is cleared.

[0010] Specifically, if the sub-link is image data, the preliminary identification result is determined based on the ResNet-50 network, which specifically includes: image preprocessing and LED parameter extraction; based on the preprocessed image, the ResNet-50 network extracts deep image features; the deep image features are integrated with the LED parameters, and classified through the fully connected layer of the ResNet-50 network to obtain the final detection result; if all indicators of the detection result are greater than the preset confidence threshold, the preliminary identification result is judged to be credible.

[0011] Specifically, if the gray node corresponds to the stage diagnosis result and the final identification result is inconsistent with the preliminary identification result, the gray node is set to black and the gray node is cleared from the cache module; the final identification result is used as the pathology classification standard, and whether the two identification results are consistent is used as the sample data classification standard. The cache and index of the gray node are respectively stored in the pathology classification table and the sample classification table in the associative database in the form of reference integrity constraints.

[0012] Specifically, the stage-by-stage diagnosis result is used as the main node of the disease course dynamic tracking tree, and the image data is used as the slave node of the disease course dynamic tracking tree. Specifically, when the main node corresponding to the stage-by-stage diagnosis result is set as a black node, the image data associated with the stage-by-stage diagnosis result is used as the slave node mounted on the main node.

[0013] The second aspect of the present invention discloses a remote management system based on video image processing, which is used to execute the method described in the first aspect. The system includes multiple remote terminals and multiple clients; wherein the remote terminals adopt a star-shaped hierarchical structure, including sub-remote terminals;

[0014] The remote terminal is used to construct a dynamic disease tracking tree based on the disease template tree and interactive information. In the dynamic disease tracking tree, nodes are used to describe sub-links of treatment, which are image data or staged diagnostic results, and edges are used to describe the associations between sub-links.

[0015] In response to the child nodes generated by the client, the sub-remote terminal determines the black, white and gray attributes of each node based on the preliminary identification results, and based on the black, white and gray attributes of each node, combined with the depth-first search algorithm, determines the traceability path of each node; wherein, the sub-remote terminal stores the gray nodes in the cache module; the remote terminal stores the black nodes in the associative database in the form of reference integrity constraints.

[0016] A third aspect of the present invention discloses a terminal, comprising a processor and a storage medium; the terminal is characterized in that:

[0017] The storage medium is used to store instructions;

[0018] The processor is configured to operate according to the instructions to execute the steps of the method of the first aspect.

[0019] A fourth aspect of the present invention discloses a computer-readable storage medium having a computer program stored thereon, wherein the program implements the steps of the method described in the first aspect when executed by a processor.

[0020] Beneficial effects of the present invention:

[0021] Compared to existing technologies, this invention first discloses a remote management concept based on video image processing. Unlike traditional disease-tracking chains, this invention first generates preliminary identification results based on an algorithm. If the confidence level does not meet the required level, a final identification result is then provided by an expert. The latency of this process is addressed by creatively utilizing a depth-first search algorithm combined with a dynamic disease-tracking tree. Furthermore, since the image data in this application scenario is obtained by a multispectral LED light source array, the tree nodes are either image data or periodic diagnostic results subject to integrity constraints. It is important to note that the dynamic disease-tracking tree in this invention is not solely used for dynamic disease-tracking; it actually serves as a node mapping function. Furthermore, the disease-tracking tree integrates image data and periodic diagnostic results, making it easier for medical staff to analyze the patient's progress. Furthermore, this invention creatively utilizes the concept of a depth-first search algorithm and integrates the mapping concept into the dynamic disease-tracking tree, eliminating the need for any additional costly information (such as a mapping table). Whether backtracking to clear erroneous image data or reclassifying defective sample data, it enables rapid manipulation of stored data. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1A This is a flowchart of the traditional treatment process.

[0023] Figure 1B 4 is a flowchart of the treatment process of an embodiment of the present invention.

[0024] Figure 2A Schematic diagram of branch nodes of a disease course dynamic tracking tree according to an embodiment of the present invention.

[0025] Figure 2B 2 is another schematic diagram of the disease course dynamic tracking tree according to an embodiment of the present invention.

[0026] Figure 3 4 is a schematic diagram of a remote management system based on video image processing according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0028] The remote management system of the present invention is used to treat scalp hair loss. Prior art remote management systems based on video image processing, as shown in Figure 1, can include multiple remote terminals and multiple clients. The remote terminals should at least include a multimedia data storage module for recording and storing image data and a clinical process monitoring module for real-time tracking and recording of a single patient's entire onset and treatment process, forming a disease tracking chain that facilitates remote analysis and guidance of the entire case.

[0029] In the use case of the present invention, the remote terminals adopt a star-shaped hierarchical structure, with a central remote terminal providing unified coordination, and the remaining remote terminals participating in data exchange as slave nodes. The specific network topology is determined by the actual deployment environment and will not be further described. As can be understood, due to the star-shaped hierarchical structure of the remote terminals, the disease tracking chain must ensure data consistency across each remote terminal.

[0030] The client can be a mobile phone app or a multi-spectral light source acquisition terminal. Unlike the mobile phone app, the multi-spectral light source acquisition terminal includes: a multi-spectral LED light source array and a communication module.

[0031] Multispectral LED light source arrays can include white, red, and blue LEDs, among others, arranged in a matrix. Each LED unit can be independently controlled to optimize imaging detail across different wavelengths. LED parameters include intensity, PWM frequency, and duty cycle. The intensity of the LED is set based on specific testing requirements. For example, white LEDs use the D65 standard (daylight) to simulate natural sunlight at a color temperature of 6500K, with a brightness range of 500–1500 lm / m². To ensure radiation safety, the output power of ultraviolet LEDs (365–405nm) must be strictly controlled. Current industry standards recommend that ultraviolet radiation should not exceed 15 µW / cm² to ensure human safety.

[0032] White light LEDs simulate standard daylight to ensure the overall brightness and color restoration of the image, making it easier for subsequent algorithms to distinguish the color boundary between the scalp and the hair. Red light LEDs (620-750 nm) are used to compare the bleeding phenomenon between the scalp and the base of the hair. Blue light LEDs (450-495 nm) can enhance surface details and are suitable for distinguishing between exposed scalp parts and covered hair. Green light LEDs (500-570 nm) are extremely sensitive to the fiber structure, color and diameter of hair and are used to accurately detect the boundary between hair and scalp and the density of hair. Ultraviolet LEDs (365-405 nm) are used to enhance the imaging of the hair surface, cuticle and fine cracks (such as hair scales) under low power conditions. Near-infrared LEDs are used for monitoring hair follicles and microcirculation under the scalp. Near-infrared light has strong penetrating ability and is sensitive to the spectral response of the internal tissue of the hair. In the embodiment of the present invention, a dual-band near-infrared design can be used: (1) short-wave band (750-850 nm) (1) Long-wavelength band (850–950 nm): suitable for detecting subscalp structures; (2) Long-wavelength band (850–950 nm): for penetrating imaging of the interior of hair fibers and distinguishing the connection between the hair shaft and the hair follicle.

[0033] The communication module supports wired or wireless transmission protocols and can transmit the collected video or image data to the remote terminal in real time, realizing remote monitoring and instant diagnosis.

[0034] Traditional treatment processes, such as Figure 1A As shown, the exemplary method may include steps 1 to 4.

[0035] Step 1: The client generates and sends first image data as a first node;

[0036] The image data may include first image data and second image data described below. The image data is typically video information of the patient's scalp; the image data may be a single image, a sequence of images, or a video, and may record both the initial state of the scalp and changes before and after medication during treatment.

[0037] Step 2: Store the first image data in a relational database in a referential integrity constraint manner.

[0038] It should be understood that the treatment process for scalp hair loss itself is not complicated. The complexity lies in generating high-quality image data and ensuring interactivity between the remote terminal and the client.

[0039] Among them, the referential integrity constraints include: constraints between nodes in the disease tracing chain (if the nodes correspond to image data) and image data, and constraints between nodes in the disease tracing chain (if the nodes correspond to stage diagnosis results) and pathological classification.

[0040] In practice, the remote terminal provides a visual interface for medical staff to conduct analysis. When a patient's window is opened, the patient's corresponding medical history chain is first displayed. Then, for the target node (usually the last node, representing the most up-to-date status), guidance and feedback are provided based on the patient's questions or by directly analyzing the image data. For example, the constraints between nodes in the medical history chain and image data are used. In the interface, medical staff can access corresponding image data or interactive information by clicking on the target node or edge. This means that referential integrity constraints must be observed between nodes in the medical history chain and image data. By analogy, the sample data corresponding to pathological classifications (essentially, part or all of the data in the medical history chain) should also be mapped to real nodes to facilitate statistical analysis of patients with similar conditions. Referential integrity constraints must also be observed between nodes in the medical history chain and pathological classifications. In other words, each node representing a patient should be considered a whole. When medical staff analyze a node (corresponding data), they should be able to access all of the patient's data and information based on the medical history chain, allowing them to conduct a holistic analysis. That is, the nodes and their corresponding data information must be constrained.

[0041] Taking the constraints between nodes in the disease tracking chain and image data as an example, the code for the referential integrity constraint can be as follows:

[0042] CREATE TABLEMediaInfo(

[0043] NodeID INT,

[0044] MediaInfomationBLOB );

[0046] CREATE TABLE Node (

[0047] NodeID INT,

[0048] FaNodeID INT,

[0049] ChildNodeID STRING,

[0050] CONSTRAINT FOREIGN KEY (NodeID)

[0051] REFERENCESMediaInfo

[0052] ON DELETE CASCADE

[0053] ON UPDATE CASCADE );

[0055] Among them, the nodes in the disease tracking chain represent NodeID, and the image data is MediaInfomation.

[0056] It's easy to understand that it's impossible to store the disease tracking chain, image data, and interactive information in the same database table. This is because image data can be large, and may even be video data. If mixed together, it would not conform to table design principles. Furthermore, the refresh process and consistency constraints of the disease tracking chain would be difficult to guarantee. Interactive information can include the guidance feedback information described below, as well as patient questions.

[0057] Step 3: Generate guidance feedback information based on the first node.

[0058] Guidance feedback can be generated by the scalp hair loss treatment management system's algorithmic processing of video images, or it can be guidance provided by professional medical staff after viewing the scalp video. This guidance feedback may include suggestions for re-shooting a specific area using LEDs with different frequency bands to obtain more accurate raw data. For example, for cases of alopecia areata, it may be recommended to locally increase blue light output (adjusting blue light brightness to approximately 350 lm / m²) to enhance local scalp texture and hair follicle detail. For cases of diffuse hair loss, it may be recommended to appropriately reduce UV LED output and increase red LED output (adjusting red light output to a medium brightness level) to avoid localized high reflection interference. Furthermore, the feedback should include detailed information on the specific parameter settings for each LED. For example, for areas with sparse hair, it is recommended to increase the intensity of the green LED by 10%-20% to more clearly capture hair details.

[0059] Step 4: Generate second image data based on the guidance feedback information as the second node.

[0060] It is understood that in some embodiments, the second image data can be used as the first image data, tracing back to step 1 and iterating again to generate more accurate original scalp information. The motivation for iteration can be to enhance the original first image data or to directly replace invalid first image data. This iterative mechanism can not only enhance the initial imaging effect, but also effectively eliminate image data that is poorly captured due to environmental interference.

[0061] Since there may be a long delay between the final identification result and the preliminary identification result, it is actually quite difficult to remove the first image data after step 4. It is understandable that the first image data may have more than one node, and may very well include multiple nodes in parallel, multiple nodes in series, or a combination of the two.

[0062] Based on this, the present invention discloses a remote management method based on video image processing, including: constructing a dynamic disease course tracking tree based on a disease course template tree and interactive information; in the dynamic disease course tracking tree: nodes are used to describe sub-links of treatment, sub-links are image data or stage-by-stage diagnostic results, and edges are used to describe the association between sub-links; based on preliminary identification results, the black, white and gray attributes of each node are determined, and based on the black, white and gray attributes of each node, the traceability path of each node is determined in combination with a depth-first search algorithm; wherein, the image data corresponding to the gray node is stored in a cache module; the image data of the black node is stored in an associative database in a reference integrity constraint manner.

[0063] It should be noted that the significance of determining the gray nodes in the upper end is to facilitate the use of the depth-first search algorithm for tracing. Since the dynamic tracing tree of the disease course is usually stored in a cache system (Etcd) with high consistency requirements, on the one hand, a mapping table will not be established in practice to associate the tracing path of the dynamic tracing tree of the disease course; on the other hand, the cache system with high consistency requirements usually does not support this storage method similar to the integrity constraint. It should be understood that the present invention uses the depth-first search algorithm, not to find the shortest path, but to obtain its tracing path under the premise that the black, white and gray nodes are determined, that is, in the depth-first search algorithm, the path constructed when returning (secondary) traversal. The specific process of the depth-first search algorithm will not be repeated in this invention.

[0064] It is understandable that the gray nodes are not stored in the relational database in a referential integrity constraint manner.

[0065] In some embodiments, the cache module can be either Redis or the relational database itself, but a new temporary table needs to be created as a distinction stored in the relational database in the form of referential integrity constraints.

[0066] Understandably, the gray nodes represent sub-links whose preliminary identification results are unreliable and have received treatment, the black nodes represent sub-links whose preliminary identification results are reliable and have received treatment; and the white nodes represent sub-links that have not yet received treatment.

[0067] For scalp hair loss treatment and management systems, preliminary identification results are typically provided by medical staff using intelligent algorithms, while final identification results can be provided by experts and professors. Each preliminary or final identification result corresponds to a separate node.

[0068] For image data, a convolutional neural network image quality detection algorithm can be used. For example, the preliminary identification result can be determined based on the ResNet-50 network. The specific process is as follows: image preprocessing and LED parameter extraction; based on the preprocessed image, the ResNet-50 network extracts deep image features; the deep image features are fused with the LED parameters and classified through the fully connected layer of the ResNet-50 network to obtain the final detection result. It can be understood that if all indicators of the detection result are greater than the preset confidence threshold, the preliminary identification result is determined to be credible. Since the ResNet-50 network is a known algorithm, the specific process will not be repeated.

[0069] For interim diagnostic results, preliminary identification is determined using the XGBoost algorithm. This involves preprocessing the positive and negative sample data in the sample classification table to extract features and generate mapping vectors. This mapping vector is trained using a cross-entropy loss function, and the XGBoost algorithm parameters are optimized using k-fold cross-validation. Since the XGBoost algorithm is well-known, the detailed process is omitted here.

[0070] In some embodiments, the patient ID and the disease progression dynamic tracking tree can be stored as key-value pairs in a key-value storage system. In some embodiments, the key-value storage system can be Etcd. Etcd is a distributed, reliable key-value storage system written in the Go language and is commonly used to store highly available and consistent data in distributed systems. Etcd can be thought of as a distributed notepad.

[0071] The medical history template tree represents a rough outline of the initial diagnostic process based on preliminary diagnostic information. The initial dynamic medical history tracking tree embodies this initial diagnostic process. It's understandable that the dynamic medical history tracking tree isn't pre-set but rather evolves in real time based on interactive information. For example, during treatment, if a patient experiences a treatment plan adjustment or complications (such as worsening local inflammation), the system can automatically create branches in the medical history record, recording both the initial treatment plan and the adjusted plan. These branches can be recorded using either a linked list or a tree-like structure. The linked list format is suitable for cases with a single, continuous treatment process, recording each image data and feedback. The tree-like structure is more suitable for cases with multiple divergent treatment plans, such as when a relapse or new pathological changes occur, branching into different sub-pathways. Furthermore, for difficult cases, researchers can add tags to the corresponding records to facilitate later specialized research, such as analyzing the cause of relapse or testing new treatment options.

[0072] like Figure 2AAs shown, a node is composed of a sub-step of the client's treatment. The sub-step here is not limited to a specific scope; that is, a node itself can include a main node and sub-nodes, where the main node is composed of multiple sub-nodes. In other words, because the disease course dynamic tracking tree changes in real time, in some specific examples, the initial disease course dynamic tracking tree can be described as {a->b; b->c}. However, during subsequent diagnosis and treatment, new conditions are discovered at nodes b or a, which may lead to divergent treatment plans. The subsequent disease course dynamic tracking tree eventually splits into {a->b1; b1->b2; b1->b3; b2->c; b3->d}. Here, b is the main node, while b1 and b2 are sub-nodes.

[0073] The edges in the disease course dynamic tracking tree mainly refer to guidance feedback information, which determines the association between the current node (for example, the first node) and the next node to be generated (for example, the second node).

[0074] In some embodiments, if the final identification result of the gray node is consistent with the preliminary identification result, the gray node is set to black, and the data corresponding to the gray node in the cache module is cleared at the same time, and transferred to the associative database in a referential integrity constraint manner.

[0075] In other embodiments, if the first node corresponds to image data and the preliminary identification result is unreliable, the logic here can refer to Figure 1B As shown, in the dynamic tracing tree of the disease course, a subpath is re-branched (for example, if the first node is Figure 2A In b2, the second node is Figure 2A In b3, the second node is generated and sent. If the preliminary identification result of the second node is credible, the image data or interim diagnosis result corresponding to the second node is stored in the associative database using a referential integrity constraint. Based on a depth-first search algorithm, the gray node (i.e., the first node) is traced back, the cache corresponding to the gray node is cleared, and the gray node itself is cleared. It is understood that if the preliminary identification result of the second node is still unreliable, the backtracking iteration continues: a new subpath is forked to generate and send the third node.

[0076] In some other embodiments, if a gray node corresponds to a staged diagnosis result and the final identification result is inconsistent with the preliminary identification result, the gray node is still set to black and cleared from the cache module. The final identification result is used as the pathology classification standard, and the consistency of the two identification results is used as the sample data classification standard. The cache and index of the gray nodes (set to black) are respectively stored in the pathology classification table and the sample classification table in the relational database in a referential integrity constraint manner. It can be understood that each patient's corresponding dynamic disease course tracking tree should contribute to the establishment of the algorithm model. Therefore, it is necessary to perform pathology classification on the newly generated dynamic disease course tracking tree as the basis for subsequent algorithm updates and iterations. In the medical field, the algorithm model should at least be able to perform feedback learning. In other words, the algorithm model needs to learn and analyze negative sample data. Even if the new algorithm model cannot accurately judge, it should still perform special judgment on negative sample data. For example, if the corresponding conclusion cannot be reached based on the negative sample model, this can greatly reduce the false positive rate (it is better to be unable to judge, but it should be minimized as much as possible). Therefore, in the present invention, pathology classification actually mainly refers to whether the sample data is negative sample data. Understandably, negative sample data refers to sample data where the algorithm fails to classify (the initial identification result is inconsistent with the final identification result).

[0077] Note: The cache of a gray node may refer to the data stored in the cache system by the gray node, such as image data or interim diagnostic results. Since it is impossible to store two copies of data in the cache system in the associative database, the index of the gray node mentioned above should be understood as the index of the cache of the gray node.

[0078] In some more preferred embodiments, the disease course dynamic tracking tree can be as follows Figure 2B As shown, the stage diagnosis result is used as the main node of the disease course dynamic tracking tree, and the image data is used as the slave node of the disease course dynamic tracking tree; specifically, considering that when the expert professor gives the stage diagnosis result, he must also give the final identification result of the associated image data at the same time, therefore, the stage diagnosis result can be directly used as the main node of the disease course dynamic tracking tree, and when the main node corresponding to the stage diagnosis result is set to black, the image data associated with the stage diagnosis result is mounted on the main node (for example: Figure 2B On the slave node (for example: Figure 2B c4 in the .

[0079] It is understandable that in this embodiment, at any time, the gray slave node must be a slave node that is not mounted on the master node (for example: Figure 2Bc1~c3 in the tree), at this time, m2 has not yet formed the final identification result. Therefore, this approach can simplify the dynamic tracking logic of the disease course dynamic tracking tree without the need for additional labeling constraints; on the other hand, it also reduces the height of the tree, making the dynamic tracking process of the disease course dynamic tracking tree essentially only composed of nodes corresponding to the staged diagnostic results, thereby greatly reducing the burden on Etcd. Among them, c4 and c2 are the same node, and are described separately for the convenience of explanation. Figure 2B In the example, assume that c2 eventually transitions from a gray node to a black node, and that c1 and c3 do not. Therefore, c2 is retained, becoming c4, while c1 and c3 are discarded. The image data associated with each stage diagnosis (e.g., c1-c3) typically refers to the image data temporally (i.e., within the disease progression tree) between the previous stage diagnosis (e.g., m1) and the current stage diagnosis (e.g., m2).

[0080] Correspondingly, the present invention also discloses a remote management system based on video image processing, such as Figure 3 As shown, it includes multiple remote terminals and multiple clients; wherein the remote terminal adopts a star-shaped hierarchical structure, including sub-remote terminals;

[0081] The remote terminal is used to construct a dynamic disease tracking tree based on the disease template tree and interactive information. In the dynamic disease tracking tree, nodes are used to describe sub-links of treatment, which are image data or staged diagnostic results, and edges are used to describe the associations between sub-links.

[0082] In response to the child nodes generated by the client, the child remote terminal determines the black, white and gray attributes of each node based on the preliminary identification results, and determines the traceability path of each node based on the black, white and gray attributes of each node in combination with the depth-first search algorithm; wherein the child remote terminal identifies the gray nodes (for example: Figure 3 g1 in the cache module; the remote terminal stores the black node (for example: Figure 3 b1) in the above table is stored in the relational database as a referential integrity constraint.

[0083] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a particular embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined in any way. To keep the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0085] It will be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A remote management method based on video image processing, characterized in that: include: Based on the disease course template tree and interactive information, a dynamic disease course tracking tree is constructed; In the dynamic disease course tracking tree: nodes are used to describe sub-links of treatment, which are image data or stage-by-stage diagnostic results, and edges are used to describe the associations between sub-links; the black, white, and gray attributes of each node are determined based on the preliminary identification results, and the traceability path of each node is determined based on the black, white, and gray attributes of each node, combined with the depth-first search algorithm; among them, gray nodes are stored in the cache module; black nodes are stored in the associative database in the form of reference integrity constraints; gray nodes represent sub-links whose preliminary identification results are unreliable and have received treatment, and black nodes represent sub-links whose preliminary identification results are reliable and have received treatment; white nodes represent sub-links that have not yet received treatment.

2. The remote management method based on video image processing according to claim 1, characterized in that: Gray nodes are not stored in the relational database with referential integrity constraints.

3. The remote management method based on video image processing according to claim 1, characterized in that: If the final identification result of the gray node is consistent with the preliminary identification result, the gray node is set to black and the gray node is cleared from the cache module.

4. The remote management method based on video image processing according to claim 1, characterized in that: If the first node corresponds to the image data and the preliminary identification result is unreliable, generating and sending a second node; If the preliminary identification result of the second node is credible, the second node is stored in the relational database in a referential integrity constraint manner; Based on the depth-first search algorithm, trace back the gray nodes and clear the cache corresponding to the gray nodes.

5. The remote management method based on video image processing according to claim 1, characterized in that: If the sub-link is image data, the preliminary identification result is determined based on the ResNet-50 network, specifically including: image preprocessing and LED parameter extraction; based on the preprocessed image, the ResNet-50 network extracts deep image features; the deep image features are fused with the LED parameters, and classified through the fully connected layer of the ResNet-50 network to obtain the final detection result; if all indicators of the detection result are greater than the preset confidence threshold, the preliminary identification result is determined to be credible.

6. The remote management method based on video image processing according to claim 1, characterized in that: If the gray node corresponds to the staged diagnosis result, and the final identification result is inconsistent with the preliminary identification result, the gray node is set to black and the gray node is cleared from the cache module; The final identification result is used as the pathology classification standard, and the consistency of the two identification results is used as the sample data classification standard. The cache and index of the gray nodes are stored in the pathology classification table and sample classification table in the relational database in the form of referential integrity constraints.

7. The remote management method based on video image processing according to claim 1, characterized in that: The stage-by-stage diagnosis result is used as the main node of the dynamic disease course tracking tree, and the image data is used as the slave node of the dynamic disease course tracking tree. Specifically, when the main node corresponding to the stage-by-stage diagnosis result is set as a black node, the image data associated with the stage-by-stage diagnosis result is used as the slave node mounted on the main node.

8. A remote management system based on video image processing, used to execute the method according to any one of claims 1 to 7, characterized in that: The system includes multiple remote terminals and multiple clients; wherein the remote terminals adopt a star-shaped hierarchical structure, including sub-remote terminals; The remote terminal is used to construct a dynamic disease tracking tree based on the disease template tree and interactive information. In the dynamic disease tracking tree, nodes are used to describe sub-links of treatment, which are image data or staged diagnostic results, and edges are used to describe the associations between sub-links. In response to the child nodes generated by the client, the sub-remote terminal determines the black, white and gray attributes of each node based on the preliminary identification results, and based on the black, white and gray attributes of each node, combined with the depth-first search algorithm, determines the traceability path of each node; wherein, the sub-remote terminal stores the gray nodes in the cache module; the remote terminal stores the black nodes in the associative database in the form of reference integrity constraints.

9. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images