Method and system for constructing anal fistula model based on endoscope and application

Abstract

The invention provides an anal fistula model construction method and system based on an endoscope and application. The method comprises the steps that data are acquired and adjusted to obtain an operation guide of an electronic colonoscope; checking according to an operation guide; judging whether the patient has anal fistula or not through the analytical model; if the anal fistula exists, analyzing first anal fistula data, searching a related anal fistula case, and further predicting to obtain second anal fistula data; and constructing a plurality of complete anal fistula three-dimensional models according to the first anal fistula data and the second anal fistula data. According to the method, comprehensive and multi-dimensional data support is provided for model construction by integrating multiple data types, the anal fistula related data is intelligently analyzed by means of the analytical model, the modeling process is simplified, the diagnosis efficiency is improved, and the method is of great significance to the accuracy and rapidity of clinical diagnosis.

Description

Technical Field

[0001] The present application relates to the field of role simulation, and specifically, to a method, system and application for constructing an anal fistula model based on an endoscope. Background Art

[0002] Anal fistula is a common disease in anorectal medicine. Its pathogenesis is complex and its prevalence is on the rise, which seriously affects the health and quality of life of patients. The construction of anal fistula model plays a vital role in the accurate diagnosis and effective treatment of anal fistula. The diagnosis of early anal fistula relies on digital examination and probe examination. Such methods require extremely high experience of doctors and are seriously affected by subjective factors. In the face of complex anal fistulas with many fistula branches and deep locations, the accuracy of diagnosis is difficult to guarantee. With the development of medical imaging technology, detection methods such as CT, MRI and endoscopy have gradually been applied to the diagnosis of anal fistula, providing more data support for the construction of anal fistula model. However, the existing anal fistula model construction is overly dependent on test data. Taking CT and MRI as examples, although they can provide detailed anatomical information of anal fistula, the detection process is cumbersome, time-consuming, and the examination cost is high, which brings a great burden to patients. In addition, model construction often requires a lot of manpower to process and analyze these test data, resulting in low efficiency in model construction. Even with the introduction of computer-assisted technology, it is still impossible to efficiently construct an accurate anal fistula model due to the lack of integration of comprehensive information such as patient symptoms and medical history. In actual clinical applications, low diagnostic efficiency not only delays the treatment of patients, but also causes a waste of medical resources to a certain extent. Summary of the invention

[0003] Based on the problems existing in the prior art, this application provides a method, system and application for constructing an anal fistula model based on endoscope. The specific scheme is as follows: In the first part, this application proposes a method for constructing an anal fistula model based on endoscope, including the following: Acquire the patient's physical data and symptom data related to anal fistula, select a corresponding operation template according to the physical data, and adjust the operation template according to the symptom data to obtain an operation guide for the electronic colonoscope for the patient; Controlling the electronic colonoscope to examine the patient according to the operation guide to obtain detection image data; judging whether the patient has an anal fistula according to the detection image data and the symptom manifestation data through a preset analytical model; If anal fistula exists, first anal fistula data involving the internal opening and fistula of the anal fistula is parsed from the detection image data in combination with the preset non-anal fistula pathological data, and related anal fistula cases are searched in an anal fistula medical record database based on the first anal fistula data and the symptom manifestation data; Predicting one or more groups of second anal fistula data according to preset medical test data, symptom data, anal fistula cases and first anal fistula data by the analytical model; A primary three-dimensional model of anal fistula is constructed based on the first anal fistula data, and multiple complete three-dimensional models of anal fistula are constructed based on the primary three-dimensional model and combined with the second anal fistula data.

[0004] In some specific embodiments, each operation template corresponds to non-anal fistula pathological data obtained by performing electronic colonoscopy according to the operation template, and the anal fistula case data is pre-labeled with pathological features and pathological areas; According to the difference between the operation template and the operation guide, the non-anal fistula pathology data corresponding to the operation template is adjusted to obtain reference image data; Extracting local features of the corresponding pathological area in the detection image data through the analytical model, and comparing them with the corresponding pathological features, so as to quickly analyze the difference between the detection image data and the reference image data; The difference analysis results and the symptom data are combined to determine whether the patient has anal fistula.

[0005] In some specific embodiments, the non-anal fistula pathology data is divided into health data and non-anal fistula disease data; The health data includes healthy intestinal mucosal morphology data and histological pathology data detected in sequence according to the operation template, and the non-anal fistula disease data includes inflammatory bowel disease image data, intestinal polyp image data and intestinal tumor image data detected in sequence according to the operation template.

[0006] In some specific embodiments, the first difference area is obtained by analyzing the difference between the detection image data and the health data through the analytical model; Analyze the difference between the detection image data and the non-anal fistula disease data in the first difference area; if a first difference area in the detection image data is different from any non-anal fistula disease data, identify the first difference area as a lesion area; The image features of the lesion area are extracted and analyzed to see whether there is a preset internal opening feature. If so, it is determined that the patient has an anal fistula, and relevant information related to the internal opening and fistula of the anal fistula in the lesion area is extracted to obtain the first anal fistula data.

[0007] In some specific embodiments, the first anal fistula data includes the internal opening position, the internal opening shape and the fistula starting direction; The second anal fistula data includes the length of the fistula, the branches of the fistula, and the relationship between the tissues around the fistula.

[0008] In some specific embodiments, in the anal fistula case database, all cases whose similarity with the patient in terms of the internal opening position, internal opening shape, and fistula starting direction is higher than a preset value are searched to obtain anal fistula cases; Determine the extension direction and extension length of all fistulas in the anal fistula case, and count the number of various extension directions and extension lengths, so as to predict the length and extension direction of the fistula; Analyze the mass characteristics in the symptom manifestation data, set a data statistical method according to the mass characteristics, and use the data statistical method to count the fistula branches and the relationship between the tissues around the fistula in the anal fistula case, so as to predict the relationship between the fistula branches and the tissues around the fistula of the patient.

[0009] In some specific embodiments, if in the first anal fistula data, the starting direction of the fistula extends in multiple directions, the internal opening is far away from the normal position of the anal crypt, or the internal opening is irregular in shape, the patient's anal fistula is determined to be a complex anal fistula, and medical examination items are added to obtain the medical examination data.

[0010] In the second part, this application proposes an endoscope-based anal fistula model construction system, including the following: An operation unit, used to obtain the patient's physical data and symptom data related to anal fistula, select a corresponding operation template according to the physical data, and adjust the operation template according to the symptom data to obtain an operation guide for the electronic colonoscope for the patient; An acquisition unit is used to control the electronic colonoscope to examine the patient according to the operation guide to obtain detection image data; and to determine whether the patient has an anal fistula according to the detection image data and the symptom manifestation data through a preset analytical model; The parsing unit, if an anal fistula exists, parses the detected image data to obtain first anal fistula data related to the anal fistula internal opening and fistula tract in combination with the preset non-anal fistula pathological data, and searches for related anal fistula cases in an anal fistula medical record database based on the first anal fistula data and the symptom manifestation data; A prediction unit, configured to predict one or more groups of second anal fistula data according to preset medical test data, symptom data, anal fistula cases and first anal fistula data through the analytical model; The model unit is used to construct a primary three-dimensional model of anal fistula around the first anal fistula data, and to construct multiple complete three-dimensional models of anal fistula based on the primary three-dimensional model in combination with the second anal fistula data.

[0011] In the third part, the present application proposes a computer device, the computer device comprising: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the endoscope-based anal fistula model construction method as described in any one of the first parts.

[0012] In the fourth part, the present application proposes a computer program product, including executable instructions, for implementing the endoscope-based anal fistula model construction method as described in any one of the first parts when executed by a processor.

[0013] Beneficial effects: The present application proposes a method, system and application for constructing an anal fistula model based on endoscope, which provides comprehensive and multi-dimensional data support for model construction by integrating multiple data types, simplifies the modeling process, overcomes the problem of low efficiency of existing model construction, and meets the needs of clinical rapid diagnosis. After integrating multiple data types, the preset analytical model and computer system are used for data processing, and the whole process from data acquisition to model construction can be completed quickly. Compared with traditional manual analysis and modeling methods, the time required for modeling is greatly shortened. By obtaining the patient's physical data and symptom data to select and adjust the operation template, a personalized electronic colonoscopy operation guide is obtained, which helps to conduct more targeted examinations and improve the quality and efficiency of obtaining detection image data. At the same time, the preset analytical model can be used to more accurately identify anal fistulas and reduce misdiagnosis and missed diagnosis. Not only can the first anal fistula data involving the internal opening and fistula of the anal fistula be analyzed from the detection image data, but also the second anal fistula data including the length of the fistula, the branches of the fistula and the relationship between the tissues around the fistula can be obtained by combining multiple data predictions, so as to comprehensively obtain the relevant information of the anal fistula, and provide a richer and more accurate basis for subsequent model construction and treatment. Based on the first anal fistula data and symptom data, relevant cases are searched in the anal fistula medical record database, and the second anal fistula data is predicted accordingly, which makes full use of previous case experience and improves the accuracy and reliability of the prediction of complex anal fistula situations.

[0014] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a schematic diagram of the process of constructing anal fistula model of the present application; Figure 2It is a schematic diagram of the principle of the anal fistula model construction method of the present application; Figure 3 It is a schematic diagram of the anal fistula determination process of the present application; Figure 4 It is a schematic diagram of the anal fistula model construction system module of the present application.

[0017] Figure numerals: 1 - operation unit; 2 - acquisition unit; 3 - parsing unit; 4 - prediction unit; 5 - model unit. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] This application proposes a method for constructing an anal fistula model based on endoscope, which breaks the reliance of traditional anal fistula three-dimensional model construction on single test data, and integrates patient physical data, symptom data, test image data and anal fistula medical record database information. Figure 1 The principle is shown in the attached Figure 2 The specific plan is as follows: A method for constructing an anal fistula model based on endoscope, comprising the following steps: 101. Obtaining the patient's physical data and symptom data related to anal fistula, selecting a corresponding operation template according to the physical data, and adjusting the operation template according to the symptom data to obtain an operation guide for the electronic colonoscopy for the patient; 102. Control the electronic colonoscope to examine the patient according to the operation guide to obtain detection image data; determine whether the patient has an anal fistula according to the detection image data and symptom data through a preset analytical model; 103. If anal fistula exists, first anal fistula data involving the internal opening and fistula of the anal fistula is parsed from the detection image data in combination with the non-anal fistula pathological data, and related anal fistula cases are searched in an anal fistula medical record database based on the first anal fistula data and the symptom data; 104. Predicting one or more groups of second anal fistula data according to preset medical test data, symptom data, anal fistula cases and first anal fistula data through an analytical model; 105. Construct a primary three-dimensional model of anal fistula based on the first anal fistula data, and construct multiple complete three-dimensional models of anal fistula based on the primary three-dimensional model and in combination with the second anal fistula data.

[0020] Step 101 mainly involves tailoring the electronic colonoscopy operation guide for the patient. The patient's physical data includes multiple aspects of information, such as age, gender, basic physical condition, and past medical history. These data are very critical because different physical conditions will affect the operation mode and risk of electronic colonoscopy. For example, the intestines of elderly patients may be more fragile, so they need to be more cautious during the operation; for patients with a history of abdominal surgery, the structure of the intestine may change, and special attention needs to be paid to avoid damage during the operation.

[0021] Symptom data related to anal fistula is an important basis for judging the patient's condition and adjusting the operation template. Common symptoms include pain, swelling, pus discharge, itching around the anus, etc., as well as specific information such as the frequency, duration, and severity of the symptoms. For example, patients who frequently experience pus discharge around the anus and severe pain may have more serious anal fistulas and need to adopt different strategies during operations. The sources for obtaining the symptom data include patient self-report, medical history data, doctor observation, palpation examination, and auxiliary examinations; the physical data include age, basic vital signs, body shape, and nutritional status.

[0022] According to the acquired patient's physical data, a matching operation template is selected from a series of pre-set operation templates. Each operation template is formulated according to people with different physical characteristics, covering the operation process, parameter settings, precautions and other contents. For example, for patients with relatively healthy bodies and no obvious history of intestinal diseases, a more conventional operation template may be selected; while for patients with certain chronic diseases, such as diabetes and hypertension, special operation templates that take these disease factors into consideration will be selected to ensure the safety and effectiveness of the operation.

[0023] After selecting the operation template, it is also necessary to adjust it according to the patient's symptom data. Because even patients with similar physical conditions may have different symptoms of anal fistula, the operation template needs to be personalized. For example, if the patient's symptoms show that the anal fistula may be in a more complicated state, it is necessary to optimize the operation steps in the operation template, which may add some examination details or adjust the order of examinations, so as to obtain an electronic colonoscopy operation guide specifically for the patient.

[0024] Regarding the construction of the operation template: Collect a large number of detailed information on patients of different ages, genders, physical conditions, and whether or not there are underlying diseases, and organize detailed records of previous electronic colonoscopy operations. Analyze the symptoms, signs, pathological characteristics, etc. of different types of anal fistulas and related intestinal diseases to understand the impact of diseases on electronic colonoscopy operations. According to factors such as the patient's age, gender, and underlying diseases, divide the patients into different categories to summarize the electronic colonoscopy operation procedures of different categories of patients and find out the commonalities and differences. Associate different types of anal fistulas and related intestinal diseases with the corresponding operation procedures. According to the results of the classification and summary, formulate detailed electronic colonoscopy operation steps for each type of patient. Clarify the range of various parameters during the operation, such as the speed of the scope, the amount of inflation, the amount of suction, etc. These parameters should be reasonably set according to the physical condition and disease characteristics of different patients. For example, for patients with more sensitive intestines, the speed of the scope should be slow and the amount of inflation should be appropriately reduced to avoid causing discomfort to the patient or intestinal damage. Formulate corresponding countermeasures for various risks that may arise during the operation.

[0025] The electronic colonoscope is equipped with a high-definition imaging system that can clearly display the internal conditions of the intestine on the screen. Doctors can directly observe the color, texture, blood vessel distribution and other fine structures of the intestinal mucosa. They can also clearly see whether there are lesions in the intestine, such as ulcers, polyps, tumors, inflammation, etc., as well as the location, size, morphology and other characteristics of the lesions.

[0026] In step 101, a personalized electronic colonoscopy operation guide has been customized based on the patient's physical data and symptom data. In this step, the operator will strictly follow the operation guide to control the electronic colonoscopy to examine the patient. The operation guide covers details such as the speed, direction, depth of the scope, as well as the focus of observation and operation techniques in different parts of the intestine. For example, if the patient's intestine is narrow or curved, the operation guide will instruct the operator to slowly advance the scope and adjust the angle to avoid damage to the intestine. During the electronic colonoscopy, the device will capture images of the inside of the intestine in real time. These images contain rich information, such as the morphology, color, and presence of lesions of the intestinal mucosa. Through advanced imaging technology, the subtle structure in the intestine can be clearly captured, providing an intuitive and accurate data source for subsequent analysis.

[0027] In this application, a parsing model is pre-established, which integrates a lot of medical knowledge and clinical experience. The models that can be selected for image analysis and classification include convolutional neural networks (CNN), recurrent neural networks (RNN) and their variants, long short-term memory networks (LSTM), etc. Through learning and training of many anal fistula cases and normal intestinal image data, it has the ability to identify the characteristics of anal fistula. The parsing model will analyze the detection image data and symptom data at the same time. The symptom data includes the patient's previous symptoms of pain, swelling, pus discharge around the anus, etc. The parsing model will look for features related to anal fistula from the detection image, such as signs of the internal opening of the anal fistula, the morphology of the fistula, etc., and make a comprehensive judgment based on the symptom manifestations. For example, if a small hole suspected to be the internal opening of an anal fistula is found in the detection image, and the patient has symptoms of long-term pus discharge around the anus, then the model will be more inclined to judge that the patient has an anal fistula.

[0028] In some specific embodiments, each operation template corresponds to non-anal fistula pathological data obtained by performing an electronic colonoscopy operation according to the operation template, and the anal fistula case data is pre-marked with pathological features and pathological areas; the non-anal fistula pathological data corresponding to the operation template is adjusted according to the difference between the operation template and the operation guide to obtain reference image data; the local features of the corresponding pathological area in the detection image data are extracted through the analytical model, and compared with the corresponding pathological features to quickly analyze the differences between the detection image data and the reference image data; the difference analysis results and symptom manifestation data are combined to determine whether the patient has an anal fistula.

[0029] Each pre-set electronic colonoscopy operation template is associated with a set of non-anal fistula pathological data obtained when operating according to this template. These data represent the characteristic information of the intestine when it is normal or suffering from non-anal fistula diseases under a specific operation process, which serves as a reference for subsequent analysis. This correspondence provides a standardized reference basis for the analysis of the detection image, so that doctors and analytical models can compare and analyze the detection results of different patients based on unified specifications. Since the operation guide is obtained by adjusting the operation template based on the patient's individual physical data and symptom data. Therefore, it is necessary to adjust the non-anal fistula pathological data corresponding to the operation template according to the difference between the two, so as to obtain reference image data that is more in line with the actual situation of the patient. Through this personalized adjustment, it is ensured that the reference image data matches the individual patient's situation and improves the accuracy of subsequent detection image analysis. For example, a patient with a history of abdominal surgery needs to pay special attention to the intestinal adhesion site in the operation guide. At this time, on the basis of the non-anal fistula pathological data of the corresponding operation template, the intestinal morphological changes that may be caused by intestinal adhesions, such as intestinal torsion and local stenosis, are added to obtain reference image data.

[0030] Using the pre-trained parsing model, the local features of the specific pathological area in the test image data are extracted, and compared with the pathological features of the corresponding pathological area in the reference image data, so as to quickly analyze the difference between the two. With the powerful feature extraction and comparison capabilities of the parsing model, efficient analysis of the test image is achieved, providing objective data support for determining whether the patient has anal fistula. For example, in the anorectal area of ​​the test image, the parsing model extracts the texture, color and other features of the area. If the normal mucosal texture of the area in the reference image is clear and the color is uniform, but the local mucosal roughness and dark red color appear in the test image, this provides clues for further judgment of anal fistula.

[0031] The difference analysis results are combined with the patient's symptom data to comprehensively judge whether the patient has anal fistula from multiple dimensions, avoiding the one-sidedness caused by single data judgment. This comprehensive analysis method makes full use of the detection image information and the patient's clinical symptom information, significantly improving the accuracy of anal fistula diagnosis.

[0032] When analyzing new detection image data, since the pathological features and pathological areas are known, there is no need to conduct a comprehensive and indiscriminate analysis of the entire image data, but instead focus directly on specific areas that may be related to anal fistulas. For example, in a large number of intestinal images taken by electronic colonoscopes, the area around the anus and possible fistulas can be quickly located to avoid wasting time in irrelevant areas, thereby greatly improving data processing speed. For the analytical model, the pre-labeled information is equivalent to providing it with prior knowledge, which can directly compare the extracted detection image features with the labeled pathological features, make judgments quickly, and reduce the calculation steps and time consumption of the model.

[0033] In practical applications, experienced pathologists, anorectal doctors or professionally trained medical technicians are responsible for the annotation work. Collect high-quality anal fistula case image data, including electronic colonoscopy images, MRI images, ultrasound images, etc. If there are different levels or structures in the pathological area, such as the wall and lumen of the fistula, and the surrounding inflammatory tissue, different colors or annotation symbols can be used to distinguish them so as to record the lesion information in more detail. Based on medical knowledge and clinical experience, identify various pathological features of anal fistula in the image. Common pathological features include the shape of the fistula (such as bending, straightness, uneven thickness), the thickness of the wall (whether it is thickened), the signal strength (in MRI or ultrasound images), the inflammatory response of the surrounding tissue (such as edema, exudation, etc.), the presence or absence of calcification, etc.

[0034] In some specific embodiments, non-anal fistula pathology data is divided into health data and non-anal fistula disease data; the health data includes healthy intestinal mucosal morphology data and histological pathology data detected in sequence according to the operation template, and the non-anal fistula disease data includes inflammatory bowel disease image data, intestinal polyp image data and intestinal tumor image data detected in sequence according to the operation template.

[0035] Health data: Healthy intestinal mucosal morphology data: refers to the data on the appearance and morphology of normal intestinal mucosa obtained when testing according to a specific operation template. For example, normal intestinal mucosa should be smooth, uniform in color, and have clear texture. These characteristics will be converted into specific data for recording, such as the roughness value of the mucosa, the RGB value of the color, etc.

[0036] Histological pathology data: This is the data obtained from the histological examination of healthy intestinal tissue. After slicing and staining the intestinal tissue, the morphology, structure, arrangement and other characteristics of the cells are observed under a microscope, and these observations are quantified into data, such as cell size, nuclear-cytoplasmic ratio, density of tissue structure, etc. These data can reflect the normal physiological state of intestinal tissue at the microscopic level.

[0037] Data on non-anal fistula diseases: Inflammatory bowel disease image data: Inflammatory bowel disease includes ulcerative colitis, Crohn's disease, etc. When detected according to the operation template, the image data of the intestine in an inflammatory state will be recorded. These images can show the pathological characteristics of the intestinal mucosa, such as congestion, edema, ulcers, erosions, etc. The image data can be two-dimensional or three-dimensional, and contain information such as the location, range, and degree of the lesions. For example, quantitative data such as the area of ​​ulcers and the degree of mucosal congestion can be obtained through image analysis. Under the electronic colonoscope, inflammatory bowel disease has specific image characteristics, such as continuous or segmental distribution of lesions, blurred mucosal vascular texture, brittle and easy bleeding, etc. These characteristics can be used as the basis for dividing inflammatory bowel disease image data.

[0038] Intestinal polyp image data: Intestinal polyps are raised lesions on the surface of the intestinal mucosa. During the detection process, image data of polyps will be obtained. These data describe the morphology (such as flat, spherical, pedunculated or sessile, etc.), size, color, surface characteristics and other information of the polyps. Intestinal polyps generally appear as round or oval protrusions under electronic colonoscopy, with a smooth or lobed surface, and with or without pedicles. Different types of polyps have different characteristics in imaging and pathology, and the intestinal polyp image data can be divided accordingly.

[0039] Intestinal tumor image data: For intestinal tumors, their image data will also be recorded. These data can reflect the growth pattern of the tumor (such as invasive growth, expansive growth, etc.), whether the boundary is clear, whether there are necrotic foci, and other characteristics. Under electronic colonoscopy, intestinal tumors have unique image features, such as cauliflower-like tumors, ulcerative tumors, intestinal stenosis, etc., combined with pathological examination to find cancer cells or dysplastic cells, which can be used as the basis for classifying intestinal tumor image data.

[0040] These diseases may have similarities with anal fistula in symptoms and colonoscopy manifestations. Including their image data in the test can help doctors make differential diagnoses when diagnosing anal fistulas and avoid misdiagnosing anal fistulas as other diseases. For example, inflammatory bowel disease may cause ulcers and inflammation of the intestinal mucosa, which has certain similarities with the local inflammatory manifestations of anal fistulas, but by analyzing the details such as the range and distribution characteristics of the lesions in the image data, it can help doctors accurately judge the condition. Intestinal polyps and tumors may also cause abnormal changes in the local intestine, which may be confused with some manifestations of anal fistulas. By comparing and analyzing the image data features of these diseases, the image features unique to anal fistulas can be more accurately identified. In addition, patients may have anal fistulas and other intestinal diseases at the same time, such as inflammatory bowel disease, intestinal polyps or tumors. Including the image data of these diseases in the test can help doctors fully understand the intestinal condition of patients, promptly detect possible combined diseases, and provide a basis for formulating comprehensive treatment plans.

[0041] Step 103 is to further mine and use existing data to find relevant cases based on the determination that the patient has an anal fistula, so as to provide support for subsequent anal fistula disease prediction and model building. Non-anal fistula pathological data includes health data and non-anal fistula disease data. Using these data as a reference helps to more accurately distinguish the features belonging to the anal fistula internal opening and fistula from the detection image data. The analytical model will compare the detection image data with the non-anal fistula pathological data, exclude those features belonging to normal tissue or other non-anal fistula diseases, and thus extract key information related to the anal fistula internal opening and fistula, that is, the first anal fistula data. The first anal fistula data usually covers the location of the internal opening, the shape of the internal opening, and the starting direction of the fistula. The information such as the location of the internal opening, the shape of the internal opening, the starting direction of the fistula in the first anal fistula data, and the symptom data (such as the degree of pain around the anus, purulent discharge, etc.) are used as search conditions to match and search in the anal fistula medical record database. Those cases in the database that are highly similar to the current patient in these aspects will be screened out as related anal fistula cases.

[0042] The purpose of searching for relevant cases in the anal fistula medical record database is to draw on the experience of similar cases in the past and provide a reference for the prediction of the current patient's condition and the formulation of a treatment plan. Similar cases may have certain commonalities in the development process of anal fistula and the treatment effect. By analyzing these cases, we can better understand the current patient's condition.

[0043] In some specific embodiments, the difference between the detection image data and the health data is analyzed by an analytical model to obtain a first difference area; the difference between the detection image data and the non-anal fistula disease data in the first difference area is analyzed; if there is a first difference area in the detection image data that is different from any non-anal fistula disease data, the first difference area is identified as a lesion area; the image features of the lesion area are extracted and analyzed to see whether there is a preset internal opening feature, if so, the patient is identified to have an anal fistula, and the relevant information about the internal opening and fistula of the anal fistula in the lesion area is extracted to obtain the first anal fistula data. The principle is as shown in the attached figure. Figure 3 As shown. The parsing model is a tool built based on machine learning or other data analysis algorithms, which has the ability to identify and analyze the features of image data. By comparison, the model will find out the areas in the test image data that do not match the healthy data, and these areas are defined as the first difference area. For example, the healthy intestinal mucosa should be smooth and uniform in color. If a part of the mucosa in the test image appears rough or changes in color, then this part of the area will be identified as the first difference area. The first difference area reflects the parts of the test image that may be abnormal. After determining the first difference area, the parsing model will continue to analyze these areas in depth. At this time, the features of the test image data in the first difference area are compared with non-anal fistula disease data (such as inflammatory bowel disease image data, intestinal polyp image data, intestinal tumor image data, etc.). Through this comparison, the model attempts to find the differences between the features of the first difference area and the features of various non-anal fistula diseases. For example, inflammatory bowel disease usually has the characteristics of extensive congestion and edema of the mucosa. If the features of the first difference area do not match the typical features of inflammatory bowel disease, then further analysis of other differences is required.

[0044] If there is a first difference area in the detected image data, its features are different from the features of all non-anal fistula disease data, which means that the abnormality in this area is not caused by common non-anal fistula diseases. Based on this situation, the first difference area is identified as a lesion area. The lesion area is defined in this way because it is relatively likely to have anal fistula-related lesions. By excluding the possibility of common non-anal fistula diseases, the cause of the abnormal area is narrowed, making it more likely to point to anal fistula. Once the lesion area is determined, the analytical model will extract detailed image features from the area, including features in texture, color, shape, etc. These features can describe the situation of the lesion area in more detail. Then, the model will compare and analyze the extracted features with the preset anal fistula internal opening features. The preset internal opening features are typical features of the anal fistula internal opening summarized based on a large number of anal fistula cases, such as the internal opening may appear as a small hole, and the surrounding mucosa may have an inflammatory reaction. If a situation that meets the preset internal opening features is found in the lesion area, then there is a greater possibility that the patient has an anal fistula. When the preset internal opening features are detected in the lesion area, it can be determined that the patient has an anal fistula. This is a diagnostic conclusion based on data analysis and feature comparison. At the same time, relevant information about the internal opening of the anal fistula and the fistula tract is extracted from the lesion area, such as the location and shape of the internal opening, the starting direction of the fistula tract, etc. This information constitutes the first anal fistula data. The first anal fistula data is of great reference value for further understanding of the specific situation of anal fistula and formulating treatment plans.

[0045] In some specific embodiments, the first anal fistula data includes the position of the internal opening, the shape of the internal opening, and the starting direction of the fistula; the second anal fistula data includes the length of the fistula, the branches of the fistula, and the relationship of the tissues around the fistula. The internal opening is the primary site of anal fistula infection, and its position and shape are crucial for determining the type of anal fistula, the source of infection, and formulating a surgical plan. The starting direction of the fistula determines the extension path of the fistula in the body, which helps doctors understand the direction and range of infection spread. The length of the fistula, the branches of the fistula, and the relationship of the tissues around the fistula further supplement the detailed information of the anal fistula. The length of the fistula affects the scope of fistula resection or treatment during surgery; the situation of the fistula branches is related to whether the infection focus can be completely removed during surgery to avoid recurrence; the relationship of the tissues around the fistula involves the difficulty and risk of the operation, as well as the impact on the function of the surrounding tissues. By clarifying these data, doctors can more accurately judge the complexity and severity of anal fistulas. For example, the internal opening of a high anal fistula is higher, the fistula may be longer and have more branches, and it is closely related to surrounding important tissues such as the sphincter, and the difficulty and risk of surgery are relatively large. The internal opening of a low anal fistula is located at a lower position, the fistula is relatively simple, and the choice of treatment plan may be relatively direct. These data can also provide precise guidance for the formulation of surgical plans. Doctors can choose the appropriate surgical approach and method based on the location of the internal opening, the direction of the fistula, and the relationship between the surrounding tissues, so as to minimize damage to normal tissues, improve the success rate of surgery, and reduce the recurrence rate.

[0046] The first anal fistula data mainly involves some basic and intuitive information about the internal opening and fistula of the anal fistula, such as the location and shape of the internal opening and the starting direction of the fistula. This information is relatively clear and can be directly observed through detection methods such as electronic colonoscopy, and has relatively clear features on the image. It can be accurately parsed directly from the detection image data using the preset parsing model. The second anal fistula data contains information such as the length of the fistula, the branching situation, and the relationship with the surrounding tissues. This information not only involves the complex structure of the anal fistula itself, but is also closely related to the surrounding physiological environment. The length of the fistula may vary due to individual differences and different stages of disease development. The branches of the fistula may be relatively hidden and may not be completely clearly displayed in the detection image. The relationship between the fistula and the surrounding tissues requires comprehensive consideration of multiple factors, which is highly complex and uncertain, and is difficult to obtain accurately directly from the detection image. Due to its complexity and uncertainty, more information needs to be combined for prediction. The anal fistula case database contains a large amount of complete anal fistula information of previous patients. By searching for cases similar to the current patient, we can learn from the second anal fistula data in these cases, and use the analytical model to comprehensively consider multiple factors such as preset medical test data, symptom data, anal fistula cases and first anal fistula data to make a more comprehensive and reasonable prediction of the second anal fistula data of the current patient, thereby constructing a complete anal fistula three-dimensional model that is more in line with the actual situation of the patient. In this way, we can make full use of the experience information in the historical data, make up for the shortcomings of the current test data, and improve the accuracy and reliability of the model.

[0047] Step 104 is to further predict the patient's anal fistula condition based on the data obtained in the previous steps with the aid of an analytical model to obtain more comprehensive anal fistula-related information. By comprehensively analyzing the preset medical test data, symptom data, anal fistula cases and first anal fistula data, other characteristics of the patient's anal fistula are predicted, thereby obtaining one or more sets of second anal fistula data, providing more detailed information for the subsequent construction of a complete anal fistula three-dimensional model. The patient's own symptom data is an important basis for reflecting his or her condition. Symptom data include the patient's subjective feelings, such as the degree and frequency of pain around the anus, whether it is accompanied by itching, etc., as well as objective manifestations, such as whether there is pus discharge, the size and location of the mass, etc. These symptoms are closely related to the pathological characteristics of anal fistula. The analytical model can further understand the severity and development stage of the patient's anal fistula by analyzing these symptoms, so as to make more accurate predictions. For example, frequent pus discharge may indicate that the fistula has more branches or the internal opening is deeper.

[0048] The analytical model uses its built-in algorithms and rules to comprehensively analyze and process the above-mentioned data. The model will identify the associations and patterns between the data, and predict the second anal fistula data such as the patient's fistula length, fistula branches, and the relationship between the tissues around the fistula through comparison and reasoning. For example, the model may predict the patient's fistula length range based on the relationship between the internal opening position and the fistula length in similar cases, as well as the internal opening position of the current patient; by analyzing the characteristics of the mass in the symptom data and the starting direction of the fistula in the first anal fistula data, combined with the situation of similar cases, the model predicts the branching of the fistula.

[0049] In some specific embodiments, in an anal fistula case database, all cases whose similarity with the patient in terms of internal opening position, internal opening morphology, and fistula starting direction is higher than a preset value are searched to obtain anal fistula cases; the extension direction and extension length of all fistulas in the anal fistula cases are determined, and the number of various extension directions and extension lengths is counted to predict the fistula length and extension direction; the lump characteristics in the symptom manifestation data are analyzed, and a data statistical method is set according to the lump characteristics. The fistula branches and the relationship between the tissues around the fistula in the anal fistula cases are counted according to the data statistical method, and then the fistula branches and the relationship between the tissues around the fistula in the patient are predicted.

[0050] A large amount of previous anal fistula case information is stored in the anal fistula case database. Each case contains key data such as the position of the internal opening, the shape of the internal opening, and the starting direction of the fistula. When facing a new patient, by finding cases that are similar to the patient in these key aspects, the corresponding situation of the new patient is predicted using the information such as the fistula length, extension direction, branching, and the relationship of the tissues around the fistula in these similar cases. At the same time, combined with the new patient's own symptom data (such as mass characteristics), the statistical method is adjusted to make the prediction result more in line with the actual situation of the patient.

[0051] In the anal fistula case database, the data of the new patient's internal opening position, internal opening shape, fistula starting direction, etc. are compared with the existing cases in the database to calculate the similarity. A preset value is set to find all cases with a similarity higher than the preset value. These cases constitute the anal fistula case set for subsequent analysis. For example, if the preset value is 80%, then cases with a similarity of 80% or more with the new patient in the above key aspects will be screened out. For the screened anal fistula cases, the extension direction (such as horizontal, vertical, oblique, etc.) and extension length of all fistulas are determined. Then, the number of various extension directions and extension lengths is counted. For example, among these similar cases, the number of cases with horizontal extension direction of the fistula and the number of cases with a length in a certain range are counted. Through this statistical analysis, the most likely fistula length and extension direction are obtained as the prediction result for the new patient.

[0052] Analyze the characteristics of the masses in the patient's symptom data, such as the size, location, hardness, etc. of the mass. Set the corresponding data statistics method based on these characteristics. For example, if the mass is large and located close to the anal sphincter, you may pay more attention to the relationship between the fistula and the sphincter. When counting the relationship between the tissues around the fistula in anal fistula cases, this information will be focused on. According to the set data statistics method, statistics are taken on the fistula branches (such as the number of branches, the direction of branches, etc.) and the relationship between the tissues around the fistula (such as the adjacent relationship with muscles, blood vessels, nerves and other tissues) in anal fistula cases, and then the relationship between the fistula branches and the tissues around the fistula are predicted for new patients.

[0053] By referring to a large number of similar cases, we can have a more comprehensive understanding of the possible conditions of anal fistulas, avoid making one-sided judgments based on the limited data of new patients, and thus improve the accuracy of the diagnosis of fistula length, extension direction, branching, and the relationship between tissues around the fistula. By adjusting the statistical method in combination with the patient's own symptom data, the prediction results are more personalized and more in line with the patient's actual condition, reflecting the concept of personalized medicine and improving the quality and effectiveness of medical services.

[0054] In practical applications, the acquisition of the second anal fistula data by the model is a process of comprehensive analysis of multiple data. Medical test data, symptom data, anal fistula cases and first anal fistula data are collected in advance, and the data is extracted for feature extraction to convert various data into numerical features that the model can process. For example, the degree of pain in the symptom manifestation is quantified as 1, 2, and 3 according to light, moderate, and severe; the circular and elliptical shapes of the internal opening are represented by different codes; the signal intensity, edge clarity and other features of the fistula are extracted from the MRI image and quantified. Through these operations, various types of data are unified into digital feature vectors, which are convenient for the model to calculate and analyze. According to the characteristics of the data and the needs of the prediction task, select appropriate machine learning or deep learning models, such as decision trees, random forests, neural networks, etc. Take neural networks as an example, it has powerful nonlinear mapping capabilities and can handle complex data relationships and feature interactions. Use historical anal fistula case data as a training set, preset medical test data, symptom data, and first anal fistula data as input features, and the corresponding second anal fistula data (fistula length, fistula branches, tissue relationships around the fistula, etc.) as output labels to train the model. During the training process, the model continuously adjusts its own parameters and learns the mapping relationship between input features and output labels to improve the accuracy of predictions. In some specific embodiments, if in the first anal fistula data, the starting direction of the fistula extends in multiple directions, the internal opening is far away from the normal position of the anal crypt, or the internal opening is irregular in shape, the patient's anal fistula is determined to be a complex anal fistula, and medical examination items are added to obtain medical examination data.

[0055] The direction of the fistula of a normal anal fistula is relatively single, but the fistula's starting direction extends in multiple directions, which means that the infection has a complex diffusion path in the perianal tissue, and there may be multiple branches or complex connections with the surrounding tissues, which increases the difficulty of diagnosis and treatment. The anal crypt is a common site of the internal opening of anal fistula. The location of the internal opening is far away from the normal location of the anal crypt, indicating that the cause of the formation of the anal fistula may be more special, or the condition has developed to a more complicated degree, which may involve deeper tissue infection or abnormal channel formation. The normal internal opening has a relatively regular shape, while the irregular internal opening often indicates that the tissues around the internal opening have been severely damaged by inflammation, or there are multiple pathogenic factors, which have changed the structure and shape of the internal opening. This is also a manifestation of the complexity of the anal fistula. For complex anal fistulas, relying solely on routine tests may not be able to fully understand the condition, and other medical test items need to be added to obtain more detailed medical test data in order to accurately diagnose and formulate a reasonable treatment plan. Since electronic colonoscopy has been excluded, the following test items can be selected: Magnetic resonance imaging (MRI): The principle is to use the magnetic resonance phenomenon of hydrogen protons in the human body when they are stimulated by radio frequency pulses in a magnetic field, generate signals, and reconstruct images through computer processing. It can clearly show the anatomical structure of the perianal soft tissue, including the course and branching of the fistula, the relationship with the surrounding muscles and organs, and the exact location of the internal opening. It is of great value for the diagnosis of complex anal fistulas, and can help doctors fully understand the condition and formulate accurate surgical plans.

[0056] Intrarectal ultrasound examination: By placing the ultrasound probe into the rectal cavity, the perianal tissue is scanned and imaged using the reflection principle of ultrasound. The location, depth, shape of the fistula and its relationship with the anal sphincter can be accurately detected. It is also important for discovering some hidden fistula branches and determining the location of the internal opening, providing an important basis for the diagnosis and treatment of complex anal fistulas.

[0057] Fistula angiography: Contrast agent is injected into the fistula, and then examined by imaging methods such as X-ray or CT. The contrast agent can fill the fistula, clearly showing the direction, branches and communication between the fistula and surrounding tissues, which helps to find some complex fistula structures that are difficult to detect through routine examinations, and provides detailed anatomical information for surgical treatment.

[0058] Step 105 is to construct a three-dimensional model of anal fistula based on the first anal fistula data and the second anal fistula data obtained in the previous step, which is of great significance for doctors to intuitively understand the specific situation of anal fistula and formulate accurate treatment plans. First, a primary three-dimensional model is constructed around the first anal fistula data, and then multiple complete anal fistula three-dimensional models are constructed based on the primary three-dimensional model and combined with the second anal fistula data.

[0059] The first anal fistula data contains the key basic information of the internal opening and fistula of the anal fistula, such as the position of the internal opening, the shape of the internal opening and the starting direction of the fistula. This information provides the core elements for constructing the initial form of the anal fistula. Using the three-dimensional modeling technology, based on the patient's body structure as the basic framework, according to the position of the internal opening in the first anal fistula data, the specific coordinates of the internal opening are determined in the three-dimensional space; according to the shape of the internal opening, the three-dimensional shape of the internal opening is shaped; combined with the starting direction of the fistula, the extension direction of the starting part of the fistula in the three-dimensional space is preliminarily depicted. Through these operations, a preliminary primary three-dimensional model that can reflect the basic characteristics of the anal fistula is constructed. For example, if the position of the internal opening is at a certain height and circumference of the rectum, the internal opening is marked at the corresponding position in the three-dimensional model; if the shape of the internal opening is circular, a circular internal opening model is constructed with this position as the center; and then according to the starting direction of the fistula, an initial fistula model is extended from the internal opening.

[0060] The second anal fistula data further supplements the detailed information of the fistula, including the length of the fistula, the branches of the fistula, and the relationship between the tissues around the fistula. This information makes the morphology of the anal fistula more complete and specific. Since there may be certain uncertainties when predicting the second anal fistula data, different combinations of prediction results may produce different anal fistula morphologies. In order to fully consider various possible situations, multiple complete anal fistula three-dimensional models are constructed. On the basis of the primary three-dimensional model, according to the length of the fistula in the second anal fistula data, the fistula in the primary model is extended accordingly; according to the branching of the fistula, branches are added to the fistula to simulate the direction and connection method of different branches; combined with the relationship of the tissues around the fistula, the relative position and spatial relationship between the anal fistula model and the surrounding tissues (such as anal sphincter, blood vessels, nerves, etc.) are adjusted. By processing different combinations of the second anal fistula data, multiple complete anal fistula three-dimensional models are constructed to show the possible various morphologies of anal fistula.

[0061] Multiple complete three-dimensional models of anal fistulas can help doctors observe and understand the specific conditions of anal fistulas from different angles and possibilities, more accurately judge the complexity and severity of anal fistulas, and avoid misdiagnosis or missed diagnosis due to a single perspective or insufficient information. Doctors can intuitively plan surgical routes, assess surgical difficulty and risks, and formulate response strategies in advance based on these three-dimensional models. For example, by observing the relationship between the fistula and the surrounding important tissues, the appropriate surgical approach can be selected to reduce damage to surrounding tissues and improve the success rate of the operation. The three-dimensional model can show the patient's condition in an intuitive way, allowing the patient to better understand his or her disease condition and enhance the patient's understanding and cooperation with the treatment plan.

[0062] This application also proposes an anal fistula model construction system based on endoscope, as shown in the attached Figure 4 As shown, including the following: The operation unit 1 is used to obtain the patient's physical data and symptom data related to anal fistula, select a corresponding operation template according to the physical data, and adjust the operation template according to the symptom data to obtain an operation guide for the electronic colonoscope for the patient; The acquisition unit 2 is used to control the electronic colonoscope to examine the patient according to the operation guide to obtain the detection image data; and to determine whether the patient has an anal fistula according to the detection image data and the symptom data through a preset analytical model; The parsing unit 3, if anal fistula exists, parses the detection image data in combination with the non-anal fistula pathological data to obtain first anal fistula data involving the anal fistula internal opening and fistula tract, and searches for related anal fistula cases in the anal fistula medical record database based on the first anal fistula data and the symptom manifestation data; Prediction unit 4, used for predicting one or more groups of second anal fistula data according to preset medical test data, symptom data, anal fistula cases and first anal fistula data through an analytical model; The model unit 5 is used to construct a primary three-dimensional model of the anal fistula around the first anal fistula data, and to construct a plurality of complete three-dimensional models of the anal fistula based on the primary three-dimensional model in combination with the second anal fistula data.

[0063] The present application provides a computer program product, which includes computer instructions, which are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes a method for constructing an anal fistula model based on an endoscope. The method for constructing an anal fistula model based on an endoscope is applied to a computer program product for easy execution.

[0064] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for constructing an anal fistula model based on an endoscope as described above.

[0065] The computer storage medium of the present application may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. The present application applies a method for constructing an anal fistula model based on an endoscope to a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the clothing simulation method provided in the present application are implemented, which is simple and fast, easy to store, and not easy to lose.

[0066] The present application proposes a method, system and application for constructing an anal fistula model based on endoscope, which provides comprehensive and multi-dimensional data support for model construction by integrating multiple data types, simplifies the modeling process, overcomes the problem of low efficiency of existing model construction, and meets the needs of clinical rapid diagnosis. After integrating multiple data types, the whole process from data acquisition to model construction can be quickly completed by using a preset analytical model and a computer system for data processing. Compared with traditional manual analysis and modeling methods, the time required for modeling is greatly shortened. By obtaining the patient's physical data and symptom data to select and adjust the operation template, a personalized electronic colonoscopy operation guide is obtained, which helps to conduct more targeted examinations and improve the quality and efficiency of obtaining detection image data. At the same time, the preset analytical model can be used to more accurately identify anal fistulas and reduce misdiagnosis and missed diagnosis. Not only can the first anal fistula data involving the internal opening and fistula of the anal fistula be analyzed from the detection image data, but also the second anal fistula data including the length of the fistula, the branches of the fistula and the relationship between the tissues around the fistula can be obtained by combining multiple data predictions, so as to comprehensively obtain the relevant information of the anal fistula and provide a richer and more accurate basis for subsequent model construction and treatment. Based on the first anal fistula data and symptom data, relevant cases are searched in the anal fistula medical record database, and the second anal fistula data is predicted accordingly, which makes full use of previous case experience and improves the accuracy and reliability of the prediction of complex anal fistula situations.

[0067] Those skilled in the art should understand that the modules of the present application described above can be implemented by a general-purpose computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, and optionally, they can be implemented by program codes executable by a computer system, so that they can be stored in a storage system and executed by the computing system, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0068] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

[0069] The above disclosure only discloses several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by technicians in this field should fall within the scope of protection of the present application.

Claims

1. A method for constructing an anal fistula model based on endoscope, characterized in that: These include: Acquire the patient's physical data and symptom data related to anal fistula, select a corresponding operation template according to the physical data, and adjust the operation template according to the symptom data to obtain an operation guide for the electronic colonoscope for the patient; Controlling the electronic colonoscope to examine the patient according to the operation guide to obtain detection image data; judging whether the patient has an anal fistula according to the detection image data and the symptom manifestation data through a preset analytical model; If anal fistula exists, first anal fistula data involving the internal opening and fistula of the anal fistula is parsed from the detection image data in combination with the preset non-anal fistula pathological data, and related anal fistula cases are searched in an anal fistula medical record database based on the first anal fistula data and the symptom manifestation data; Predicting one or more groups of second anal fistula data according to preset medical test data, symptom data, anal fistula cases and first anal fistula data through the analytical model; A primary three-dimensional model of anal fistula is constructed based on the first anal fistula data, and multiple complete three-dimensional models of anal fistula are constructed based on the primary three-dimensional model and combined with the second anal fistula data.

2. The method for constructing anal fistula model according to claim 1, characterized in that: Each operation template corresponds to non-anal fistula pathological data obtained by performing electronic colonoscopy according to the operation template, and the anal fistula case data is pre-labeled with pathological features and pathological areas; According to the difference between the operation template and the operation guide, the non-anal fistula pathology data corresponding to the operation template is adjusted to obtain reference image data; Extracting local features of the corresponding pathological area in the detection image data by using the analytical model, and comparing them with the corresponding pathological features, so as to quickly analyze the difference between the detection image data and the reference image data; The difference analysis results and the symptom data are combined to determine whether the patient has anal fistula.

3. The method for constructing anal fistula model according to claim 2, characterized in that: The non-anal fistula pathology data is divided into health data and non-anal fistula disease data; The health data includes healthy intestinal mucosal morphology data and histological pathology data detected in sequence according to the operation template, and the non-anal fistula disease data includes inflammatory bowel disease image data, intestinal polyp image data and intestinal tumor image data detected in sequence according to the operation template.

4. The method for constructing anal fistula model according to claim 3, characterized in that: Analyzing the difference between the detection image data and the health data by the analytical model to obtain a first difference area; Analyze the difference between the detection image data and the non-anal fistula disease data in the first difference area; if a first difference area in the detection image data is different from any non-anal fistula disease data, identify the first difference area as a lesion area; The image features of the lesion area are extracted and analyzed to see whether there is a preset internal opening feature. If so, it is determined that the patient has an anal fistula, and relevant information related to the internal opening and fistula of the anal fistula in the lesion area is extracted to obtain the first anal fistula data.

5. The method for constructing anal fistula model according to claim 1, characterized in that: The first anal fistula data includes the position of the internal opening, the shape of the internal opening and the starting direction of the fistula; The second anal fistula data includes the length of the fistula, the branches of the fistula, and the relationship between the tissues around the fistula.

6. The method for constructing anal fistula model according to claim 5, characterized in that: In the anal fistula case database, all cases with a similarity with the patient in terms of the internal opening position, internal opening shape, and fistula starting direction higher than a preset value are searched to obtain anal fistula cases; Determine the extension direction and extension length of all fistulas in the anal fistula case, and count the number of various extension directions and extension lengths, so as to predict the length and extension direction of the fistula; Analyze the mass characteristics in the symptom manifestation data, set a data statistical method according to the mass characteristics, and use the data statistical method to count the fistula branches and the relationship between the tissues around the fistula in the anal fistula case, so as to predict the relationship between the fistula branches and the tissues around the fistula of the patient.

7. The method for constructing anal fistula model according to claim 5, characterized in that: If in the first anal fistula data, the starting direction of the fistula extends in multiple directions, the internal opening is far away from the normal position of the anal crypt, or the internal opening is irregular in shape, the patient's anal fistula is determined to be a complex anal fistula, and additional medical examination items are added to obtain the medical examination data.

8. An endoscope-based anal fistula model construction system, characterized in that: These include: An operation unit, used to obtain the patient's physical data and symptom data related to anal fistula, select a corresponding operation template according to the physical data, and adjust the operation template according to the symptom data to obtain an operation guide for the electronic colonoscope for the patient; An acquisition unit is used to control the electronic colonoscope to examine the patient according to the operation guide to obtain detection image data; and to determine whether the patient has an anal fistula according to the detection image data and the symptom manifestation data through a preset analytical model; The parsing unit, if an anal fistula exists, parses the detected image data to obtain first anal fistula data related to the anal fistula internal opening and fistula tract in combination with the preset non-anal fistula pathological data, and searches for related anal fistula cases in an anal fistula medical record database based on the first anal fistula data and the symptom manifestation data; A prediction unit, configured to predict one or more groups of second anal fistula data according to preset medical test data, symptom data, anal fistula cases and first anal fistula data through the analytical model; The model unit is used to construct a primary three-dimensional model of anal fistula around the first anal fistula data, and to construct multiple complete three-dimensional models of anal fistula based on the primary three-dimensional model in combination with the second anal fistula data.

9. A computer device, characterized in that: The computer device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the endoscope-based anal fistula model construction method as described in any one of claims 1-7.

10. A computer program product, characterized in that It includes executable instructions for implementing the endoscope-based anal fistula model construction method as described in any one of claims 1-7 when executed by a processor.

Images