Method and system for identifying the cause of ischemic stroke based on neuroimaging

Abstract

The present invention relates to the field of neuroscience, and proposes a method and system for realizing the cause identification of ischemic stroke based on neuroimaging, wherein the method comprises: obtaining a patient's brain nerve image, including CT and MRI scan images, diagnosing and analyzing the stroke patient, extracting the vascular load value by identifying the vascular tissue and the high-density sign of the cerebral artery in the CT scan image, identifying the thrombus pore through the MRI scan image, extracting the thrombus morphological characteristics, calculating the thrombus shape coefficient, combining the vascular load value and the thrombus shape coefficient, determining the abnormal area in the brain nerve image, identifying the ischemic area, and detecting the damaged nerve tissue, and finally, using the cause identification network to classify the lesions of the damaged nerve tissue, correlating the cause of stroke with the clinical index, and generating a cause report. The present invention can improve the accuracy of the cause identification of ischemic stroke.

Description

Technical Field

[0001] The present invention relates to the field of neuroscience, and in particular to a method and system for identifying the cause of ischemic stroke based on neuroimaging. Background Art

[0002] Ischemic stroke is an acute brain dysfunction caused by obstruction of blood supply to the brain. Timely and accurate identification of the cause of stroke is crucial to the treatment and rehabilitation of patients.

[0003] At present, the common methods for identifying the cause of stroke are mainly based on the size and location of the brain lesion area, lacking an in-depth understanding of the direct impact of the cause. In addition, due to the complexity and diversity of stroke lesions, the judgment method based on the size and location of the brain lesion area often cannot fully reveal the etiology mechanism, thereby affecting the accuracy of cause identification. Therefore, a method for identifying the cause of ischemic stroke based on neuroimaging is needed to improve the accuracy of identifying the cause of ischemic stroke. Summary of the invention

[0004] The present invention provides a method and system for identifying the cause of ischemic stroke based on neuroimaging, the main purpose of which is to improve the accuracy of identifying the cause of ischemic stroke.

[0005] Acquiring a brain nerve image of a stroke patient, wherein the brain nerve image includes a CT scan image and an MRI scan image;

[0006] Determine a vascular tissue area in the CT scan image, identify a cerebral artery high-density sign in the vascular tissue area, extract characteristic parameters of the cerebral artery high-density sign, the characteristic parameters including the number of blood vessels, blood vessel diameter, and blood vessel length, and calculate a vascular load value corresponding to the CT scan image based on the characteristic parameters;

[0007] identifying thrombus pores in the MRI scan image, determining thrombus morphological features in the MRI scan image based on the thrombus pores, extracting thrombus penetration parameters corresponding to the thrombus morphological features, and calculating thrombus shape coefficients corresponding to the MRI scan image based on the thrombus penetration parameters;

[0008] Based on the vascular load value and the thrombus shape coefficient, determining the abnormal brain tissue area in the brain nerve image, performing ischemic area recognition on the abnormal brain tissue area to obtain a brain ischemic area, and performing damage detection on the brain ischemic area to obtain damaged nerve tissue;

[0009] The damaged nerve tissue is classified by using a preset cause identification network to obtain the cause of stroke corresponding to the stroke patient, the cause of stroke and the clinical indicators corresponding to the stroke patient are correlated to generate a cause report corresponding to the stroke patient.

[0010] Optionally, the identifying the cerebral artery high density sign in the vascular tissue region comprises:

[0011] separating the vascular tissue in the vascular tissue area, and detecting high-density signals corresponding to the vascular tissue;

[0012] Performing signal screening on the high-density signal to obtain a density screening signal;

[0013] The cerebral artery corresponding to the density screening signal is queried, and the cerebral artery high-density sign corresponding to the cerebral artery is identified.

[0014] Optionally, the calculating, based on the characteristic parameter, a vascular load value corresponding to the CT scan image includes:

[0015] The vascular load value corresponding to the CT scan image is calculated using the following formula:

[0016]

[0017] Wherein, Vz represents the vascular load value corresponding to the CT scan image, N represents the number of features corresponding to the feature parameter, i represents the feature index corresponding to the feature parameter, Sm i represents the number of blood vessels in the i-th characteristic parameter, Dz i represents the blood vessel diameter in the i-th characteristic parameter, Lc i represents the length of the blood vessel in the i-th characteristic parameter.

[0018] Optionally, determining the thrombus morphological features in the MRI scan image based on the thrombus pores includes:

[0019] Locating the thrombus area and the pore area corresponding to the thrombus pore;

[0020] Performing correlation analysis on the thrombus region and the pore region to obtain correlation parameters corresponding to the thrombus pore;

[0021] Extracting the morphological factors corresponding to the associated parameters;

[0022] Based on the morphological factors, thrombus morphological features in the MRI scan image are determined.

[0023] Optionally, the calculating the thrombus shape coefficient corresponding to the MRI scan image based on the thrombus penetration parameter includes:

[0024] The thrombus shape coefficient corresponding to the MRI scan image is calculated using the following formula:

[0025]

[0026] Among them, XSI represents the thrombus shape coefficient corresponding to the MRI scan image, M represents the total number of thrombus pixels in the MRI scan image, j represents the index of the thrombus pixel in the MRI scan image, T represents the total time of the MRI scanning process, ρt refers to the thrombus permeability parameter, which represents the thrombus permeability at time t, and A represents the area of ​​thrombus pixels.

[0027] Optionally, determining the abnormal brain tissue area in the brain nerve image based on the vascular load value and the thrombus shape coefficient includes:

[0028] Querying characteristic indicators corresponding to the vascular load value and the thrombus shape coefficient;

[0029] Based on the characteristic index, a reference mark is performed on the brain tissue region in the brain nerve image to obtain a reference label;

[0030] Querying anomaly index in the benchmark tag;

[0031] Based on the abnormality index, an abnormal brain tissue region in the brain nerve image is determined.

[0032] Optionally, the identifying the ischemic region of the abnormal brain tissue region to obtain the brain ischemic region includes:

[0033] Converting the abnormal brain tissue region into a binary image, and determining a seed point in the binary image;

[0034] Using a preset growth rule, the seed point is regularly expanded to obtain an expanded pixel area;

[0035] Extracting ischemic pixels in the extended pixel area;

[0036] Based on the ischemic pixels, the ischemic region of the abnormal brain tissue area is identified to obtain the brain ischemic region.

[0037] Optionally, the performing damage detection on the ischemic area of ​​the brain to obtain damaged neural tissue comprises:

[0038] Analyzing the regional dimension corresponding to the brain ischemic area;

[0039] Normalizing the regional dimensions to obtain a unified dimension corresponding to the brain ischemic region;

[0040] Extracting dimensional features in the unified dimension;

[0041] Performing linear dimensionality reduction on the dimensional features to obtain dimensionality reduction feature parameters;

[0042] Based on the dimension reduction feature parameters, querying the neural region in the brain ischemic region;

[0043] The nerve region is subjected to damage detection to obtain damaged nerve tissue.

[0044] Optionally, the using a preset etiology identification network to classify the damaged nerve tissue to obtain the stroke etiology corresponding to the stroke patient includes:

[0045] Inputting the image data corresponding to the damaged nerve tissue into an input layer of a preset etiology identification network to obtain input image data;

[0046] Extracting key features from the input image data using a feature extraction layer in the preset etiology identification network;

[0047] Using the feature selection layer in the preset etiology identification network to screen the key features to obtain screening features;

[0048] The lesion classification layer in the preset cause identification network is used to classify the lesions of the screening features to obtain the cause of stroke corresponding to the stroke patient.

[0049] In order to solve the above problems, the present invention also provides a system for identifying the cause of ischemic stroke based on neuroimaging, the system comprising:

[0050] An image acquisition module is used to acquire brain nerve images of stroke patients, wherein the brain nerve images include CT scan images and MRI scan images;

[0051] a load calculation module, configured to determine a vascular tissue region in the CT scan image, identify a cerebral artery high-density sign in the vascular tissue region, extract characteristic parameters from the cerebral artery high-density sign, wherein the characteristic parameters include the number of blood vessels, the diameter of the blood vessels, and the length of the blood vessels, and calculate a vascular load value corresponding to the CT scan image based on the characteristic parameters;

[0052] a coefficient calculation module, configured to identify thrombus pores in the MRI scan image, determine thrombus morphological features in the MRI scan image based on the thrombus pores, extract thrombus penetration parameters corresponding to the thrombus morphological features, and calculate a thrombus shape coefficient corresponding to the MRI scan image based on the thrombus penetration parameters;

[0053] a damaged detection module, configured to determine the abnormal brain tissue region in the brain nerve image based on the vascular load value and the thrombus shape coefficient, perform ischemic region identification on the abnormal brain tissue region to obtain a brain ischemic region, and perform damage detection on the brain ischemic region to obtain damaged nerve tissue;

[0054] The report generation module is used to use a preset cause identification network to classify the lesions of the damaged neural tissue, obtain the cause of stroke corresponding to the stroke patient, associate the information of the cause of stroke with the clinical indicators corresponding to the stroke patient, and generate a cause report corresponding to the stroke patient.

[0055] First, by acquiring brain nerve images of stroke patients, the present invention can more accurately locate abnormal areas, detect ischemic lesions, and evaluate the degree of damage to nerve tissue, and provide patients with accurate medical advice and treatment plans, thereby helping to improve the subsequent treatment effect. At the same time, by determining the vascular tissue area in the CT scan image, the present invention can evaluate the patency of the blood vessels, abnormal expansion or stenosis, and provide important references for surgical planning, interventional treatment and drug selection, thereby helping patients receive effective treatment in a timely manner, reducing the risk of complications, and improving the success rate of treatment and patient survival rate. By identifying the thrombus pores in the MRI scan image, the present invention can help doctors accurately and quickly locate and diagnose thrombi in patients. The present invention can identify abnormal brain tissue areas in the brain nerve image based on the vascular load value and the thrombus shape coefficient, and accurately locate abnormal brain tissue areas in the brain nerve image, thereby helping to quickly diagnose and treat abnormal brain conditions in patients, and help to gain an in-depth understanding of the occurrence mechanism and pathological characteristics of abnormal brain phenomena. Furthermore, the present invention classifies lesions of the damaged nerve tissue by utilizing a preset etiology recognition network, and obtains the corresponding cause of stroke for the stroke patient, which can provide an individualized etiology diagnosis and treatment plan for ischemic stroke patients, so as to improve efficacy and reduce risks. Therefore, the method and system for etiology recognition of ischemic stroke based on neuroimaging proposed in the present invention can improve the accuracy of etiology recognition of ischemic stroke. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of a process for identifying the cause of ischemic stroke based on neuroimaging provided in one embodiment of the present invention;

[0057] Figure 2 A schematic diagram of a module of a system for identifying the cause of ischemic stroke based on neuroimaging provided in one embodiment of the present invention;

[0058] Figure 3A schematic diagram of the internal structure of an electronic device for implementing a method for identifying the cause of ischemic stroke based on neuroimaging provided in one embodiment of the present invention.

[0059] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0060] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0061] The embodiment of the present application provides a method for identifying the cause of ischemic stroke based on neuroimaging. The execution subject of the method for identifying the cause of ischemic stroke based on neuroimaging includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for identifying the cause of ischemic stroke based on neuroimaging can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0062] Reference Figure 1 FIG. 1 is a flow chart of a method for identifying the cause of ischemic stroke based on neuroimaging provided by an embodiment of the present invention. In this embodiment, the method for identifying the cause of ischemic stroke based on neuroimaging includes:

[0063] S1. Acquire brain nerve images of stroke patients, wherein the brain nerve images include CT scan images and MRI scan images.

[0064] By acquiring brain nerve images of stroke patients, the present invention can more accurately locate abnormal areas, detect ischemic lesions, assess the degree of damage to neural tissue, and provide patients with precise medical advice and treatment plans, thereby helping to improve subsequent treatment outcomes.

[0065] Among them, the brain nerve images usually include cross-sectional or stereoscopic images obtained through medical imaging technologies such as CT scanning images (computed tomography) and MRI scanning images (magnetic resonance imaging). These images can show the details of the brain structure, tissue and vascular system. In detail, the brain nerve images of stroke patients can be obtained through medical imaging equipment, such as: CT scanners, MRI machines, ultrasound equipment and other equipment.

[0066] S2. Determine the vascular tissue area in the CT scan image, identify the cerebral artery high density sign in the vascular tissue area, extract characteristic parameters of the cerebral artery high density sign, the characteristic parameters include the number of blood vessels, blood vessel diameter and blood vessel length, and calculate the vascular load value corresponding to the CT scan image based on the characteristic parameters.

[0067] By determining the vascular tissue area in the CT scan image, the present invention can evaluate the patency, abnormal expansion or stenosis of the blood vessels, and provide important reference for surgical planning, interventional treatment and drug selection, thereby helping patients receive effective treatment in a timely manner, reducing the risk of complications, and improving the treatment success rate and patient survival rate.

[0068] The vascular tissue area refers to the arterial and venous structures clearly displayed in the CT scan image. By analyzing these vascular areas, the patency, abnormal expansion or stenosis of the blood vessels can be accurately evaluated. Optionally, the determination of the vascular tissue area in the CT scan image can be achieved through a deep learning model, such as: U-Net, SegNet and other models.

[0069] Furthermore, the present invention can effectively evaluate the patency and abnormality of cerebral blood vessels by identifying the high-density sign of cerebral arteries in the vascular tissue area, which is helpful for the early detection of cerebrovascular diseases such as aneurysms, cerebral infarction, etc., and provides an important reference for surgical planning, drug treatment and interventional treatment.

[0070] The cerebral artery high density sign refers to the abnormal density or high density sign displayed in the cerebral artery area.

[0071] As an embodiment of the present invention, the identifying of the cerebral artery high-density sign in the vascular tissue area includes: separating the vascular tissue in the vascular tissue area; detecting the high-density signal corresponding to the vascular tissue; performing signal screening on the high-density signal to obtain a density screening signal; querying the cerebral artery corresponding to the density screening signal; and identifying the cerebral artery high-density sign corresponding to the cerebral artery.

[0072] Among them, the vascular tissue refers to the vascular structure located in a specific area; the high-density signal refers to the signal with a higher density shown by the vascular tissue; the density screening signal refers to the density-related signal obtained after screening; and the cerebral artery refers to the arterial structure in the brain area.

[0073] Furthermore, the separation of vascular tissue in the vascular tissue area can be achieved through a regional separation model, such as: U-Net, MaskR-CNN and other models; the detection of the high-density signal corresponding to the vascular tissue can be achieved through an edge detection algorithm; the signal screening of the high-density signal can be achieved through a filtering algorithm, such as: mean filtering, median filtering, Gaussian filtering and other algorithms; the query of the cerebral artery corresponding to the density screening signal can be achieved through feature matching technology, such as: SIFT, SURF, ORB and other methods; the identification of the cerebral artery high-density sign corresponding to the cerebral artery can be achieved through a deep learning model, such as: CNN, RNN, LSTM and other models.

[0074] By extracting characteristic parameters from the cerebral artery hyperdensity sign, the present invention can understand the morphology, size, position and other information of the hyperdensity sign, accurately identify the patient's lesion condition, and thus provide more timely treatment and intervention measures.

[0075] Among them, the characteristic parameters refer to representative and distinguishing characteristic values ​​extracted from the cerebral artery hyperdensity sign data, and the characteristic parameters include the number of blood vessels, blood vessel diameter and blood vessel length. Optionally, the characteristic parameters extracted from the cerebral artery hyperdensity sign can be obtained through a feature extraction algorithm, such as: gray level co-occurrence matrix, local binary pattern and other algorithms.

[0076] As an embodiment of the present invention, the calculating the vascular load value corresponding to the CT scan image based on the characteristic parameter includes:

[0077] The vascular load value corresponding to the CT scan image is calculated using the following formula:

[0078]

[0079] Wherein, Vz represents the vascular load value corresponding to the CT scan image, N represents the number of features corresponding to the feature parameter, i represents the feature index corresponding to the feature parameter, Sm i represents the number of blood vessels in the i-th characteristic parameter, Dz i represents the blood vessel diameter in the i-th characteristic parameter, Lc i represents the length of the blood vessel in the i-th characteristic parameter.

[0080] S3. Identify the thrombus pores in the MRI scan image, determine the thrombus morphological characteristics in the MRI scan image based on the thrombus pores, extract the thrombus penetration parameters corresponding to the thrombus morphological characteristics, and calculate the thrombus shape coefficient corresponding to the MRI scan image based on the thrombus penetration parameters.

[0081] By identifying the thrombus pores in the MRI scan image, the present invention can help doctors accurately and quickly locate and diagnose the thrombus condition in the patient's body, improve the accuracy and effectiveness of the treatment plan, and help to take necessary medical intervention measures in a timely manner.

[0082] The thrombus pore refers to a tiny cavity or structure formed in a blood vessel or around a blood vessel, which may cause blood flow obstruction or other serious health problems. Optionally, the identification of the thrombus pore in the MRI scan image can be achieved through a genetic algorithm, such as selection, crossover, mutation and other algorithms.

[0083] Furthermore, the present invention determines the thrombus morphological characteristics in the MRI scan image based on the thrombus pores, which can help predict the development trend of the patient's condition, take intervention measures in advance, reduce the risk of complications, and help improve the automatic analysis and decision support system of medical images, thereby improving the accuracy and efficiency of medical diagnosis.

[0084] The thrombus morphological characteristics refer to the use of the extracted morphological factors to describe and characterize the overall morphological characteristics of the thrombus in the MRI scan image.

[0085] As an embodiment of the present invention, determining the thrombus morphological characteristics in the MRI scan image based on the thrombus pore includes: locating the thrombus area and the pore area corresponding to the thrombus pore; performing correlation analysis on the thrombus area and the pore area to obtain correlation parameters corresponding to the thrombus pore; extracting the morphological factors corresponding to the correlation parameters; and determining the thrombus morphological characteristics in the MRI scan image based on the morphological factors.

[0086] Among them, the thrombus area refers to the specific area where thrombus formation is shown in the MRI scan image, such as: blood clots, coagulants, etc.; the pore area refers to the cavity or void part inside the thrombus area shown in the MRI scan image; the association parameter refers to the numerical parameter extracted from the relationship between the thrombus area and the pore area, which is used to describe the interaction or connection between them, such as distance, shape matching, etc.; the morphological factor refers to the parameter that can quantify the thrombus pore characteristics calculated based on the association parameter, which is used to describe the morphological characteristics of the thrombus, such as size, shape, density, structure, etc.

[0087] Furthermore, the thrombus area and the pore area corresponding to the positioning of the thrombus pore can be obtained by image segmentation or object detection algorithm; the association analysis of the thrombus area and the pore area can be obtained by Pandas and Scikit-learn libraries in Python; the extraction of the morphological factors corresponding to the association parameters can be obtained by morphological methods, such as morphological feature extraction, morphological analysis and other methods; the determination of the thrombus morphological features in the MRI scan image can be obtained by feature extraction algorithms, such as corner detection, edge detection and other algorithms.

[0088] The present invention helps to accurately describe the morphological characteristics and tissue structure of the thrombus by extracting the thrombus permeation parameters corresponding to the thrombus morphological characteristics, and further reveals the permeability properties inside the thrombus, thereby effectively optimizing the patient's treatment plan, improving the treatment effect, and promoting the precise diagnosis and treatment of thrombus-related diseases.

[0089] Among them, the thrombus permeation parameter refers to the numerical representation of the internal permeability properties of the thrombus, usually including specific indicators of the thrombus permeation rate, permeation degree, permeation depth, etc. Optionally, the thrombus permeation parameter corresponding to the thrombus morphological characteristics can be extracted by a semantic segmentation algorithm, such as: U-Net, MaskR-CNN and other algorithms.

[0090] As an embodiment of the present invention, the calculating of the thrombus shape coefficient corresponding to the MRI scan image based on the thrombus penetration parameter includes:

[0091] The thrombus shape coefficient corresponding to the MRI scan image is calculated using the following formula:

[0092]

[0093] Among them, XSI represents the thrombus shape coefficient corresponding to the MRI scan image, M represents the total number of thrombus pixels in the MRI scan image, j represents the index of the thrombus pixel in the MRI scan image, T represents the total time of the MRI scanning process, ρt refers to the thrombus permeability parameter, which represents the thrombus permeability at time t, and A represents the area of ​​thrombus pixels.

[0094] S4. Based on the vascular load value and the thrombus shape coefficient, determine the abnormal brain tissue area in the brain nerve image, perform ischemic area identification on the abnormal brain tissue area to obtain the brain ischemic area, perform damage detection on the brain ischemic area to obtain damaged nerve tissue.

[0095] The present invention determines the abnormal brain tissue area in the brain nerve image based on the vascular load value and the thrombus shape coefficient, and can accurately locate the abnormal brain tissue area in the brain nerve image, thereby helping to quickly diagnose and treat the patient's brain abnormality and helping to gain a deep understanding of the occurrence mechanism and pathological characteristics of brain abnormalities.

[0096] The abnormal brain tissue area refers to the area in the brain image or scan that is different from the normal brain tissue structure or has problems, such as tumors, injuries, hemorrhages, inflammation, degeneration, etc.

[0097] As an embodiment of the present invention, the method of determining the abnormal brain tissue area in the brain nerve image based on the vascular load value and the thrombus shape coefficient includes: querying characteristic indicators corresponding to the vascular load value and the thrombus shape coefficient; based on the characteristic indicators, performing a reference mark on the brain tissue area in the brain nerve image to obtain a reference label; querying the abnormal index in the reference label; and determining the abnormal brain tissue area in the brain nerve image based on the abnormal index.

[0098] Among them, the characteristic index refers to an indicator used to describe, quantify or measure the characteristics or attributes of an object or phenomenon; the benchmark label refers to a label or classification used as a reference in a certain field or task; the anomaly index refers to an indicator used to measure or judge the degree of abnormality of a sample, object or phenomenon.

[0099] Furthermore, the query of the characteristic index corresponding to the vascular load value and the thrombus shape coefficient can be obtained by a feature extraction algorithm, such as GLCM, LBP, HOG and other algorithms; the benchmark marking of the brain tissue area in the brain nerve image can be obtained by a deep learning model, such as U-Net, SegNet, Mask R-CNN and other models; the query of the abnormal index in the benchmark label can be obtained by a statistical method; the determination of the abnormal brain tissue area in the brain nerve image can be obtained by a medical image processing tool, such as ITK-SNAP, MRIcron and other tools.

[0100] The present invention obtains the brain ischemic area by identifying the ischemic area of ​​the abnormal brain tissue area, thereby reducing treatment errors, improving treatment effects, and protecting the patient's brain function to the greatest extent.

[0101] The brain ischemic region refers to the ischemic region obtained by identifying the abnormal brain tissue region based on the extracted ischemic pixels.

[0102] As an embodiment of the present invention, the ischemic region identification of the abnormal brain tissue region to obtain the brain ischemic region includes: converting the abnormal brain tissue region into a binary image; determining a seed point in the binary image; using a preset growth rule to regularly expand the seed point to obtain an extended pixel region; extracting ischemic pixel points in the extended pixel region; and based on the ischemic pixel points, the ischemic region identification of the abnormal brain tissue region to obtain the brain ischemic region.

[0103] Among them, the binary image refers to an image in which the abnormal area of ​​brain tissue is converted into an image containing only two pixel values ​​(usually 0 and 1), which is used to indicate the location of the abnormal area; the seed point refers to an initial point specified in the binary image, which serves as the starting point of the growth algorithm; the preset growth rule refers to a set of rules that define how to expand the target area according to the selected seed point in the growth algorithm; the extended pixel area refers to a series of pixels connected to the seed point obtained by recursively expanding the seed point by applying the growth rule; the ischemic pixel point refers to a pixel point representing ischemia extracted from the extended pixel area.

[0104] Furthermore, the conversion of the abnormal brain tissue area into a binary image can be achieved through image processing tools, such as OpenCV, Scikit-image and other tools; the determination of the seed points in the binary image can be achieved through a threshold segmentation algorithm, such as Otsu algorithm, adaptive threshold algorithm, etc.; the rule expansion of the seed points can be achieved by utilizing a preset growth rule; the extraction of ischemic pixels in the expanded pixel area can be achieved through a pixel classifier, such as support vector machine, random forest, convolutional neural network, etc.; the identification of ischemic areas in the abnormal brain tissue area can be achieved through a semantic segmentation model, such as FCN, UNet, DeepLab and other models.

[0105] The present invention obtains damaged nerve tissue by performing damage detection on the ischemic brain area, and can timely and accurately identify the ischemic brain area and locate the damaged nerve tissue, thereby providing important auxiliary information to help formulate a more effective treatment plan.

[0106] The damaged neural tissue refers to the neural tissue judged to be in a damaged state according to the damage detection result.

[0107] As an embodiment of the present invention, the damage detection of the brain ischemic area to obtain damaged neural tissue includes: analyzing the regional dimension corresponding to the brain ischemic area; normalizing the regional dimension to obtain a unified dimension corresponding to the brain ischemic area; extracting dimensional features in the unified dimension; performing linear dimensionality reduction on the dimensional features to obtain dimensionality reduction feature parameters; based on the dimensionality reduction feature parameters, querying the neural area in the brain ischemic area; and performing damage detection on the neural area to obtain damaged neural tissue.

[0108] Among them, the regional dimension refers to the size information such as width, height and depth of the brain ischemic area; the unified dimension refers to the dimensional information with the same scale range obtained after normalizing the regional dimension; the dimensional feature refers to the feature representing the shape and size of the brain ischemic area extracted from the unified dimension; the dimensionality reduction feature parameter refers to the parameter that converts high-dimensional dimensional features into low-dimensional dimensionality reduction features through linear dimensionality reduction technology; the neural area refers to the regional information associated with the dimensionality reduction feature parameter in the database.

[0109] Furthermore, the analysis of the regional dimension corresponding to the brain ischemic area can be obtained through a target detection model, such as: YOLO, Faster R-CNN and other models; the normalization of the regional dimension can be obtained through a normalization method, such as: Min-Max normalization, Z-score normalization and other methods; the extraction of dimensional features in the unified dimension can be obtained through a shape descriptor, such as: Hu moment, Zernike moment, Fourier descriptor, etc.; the linear dimensionality reduction of the dimensional features can be obtained through a dimensional analysis method, such as: PCA, LDA and other methods; the query of the neural area in the brain ischemic area can be obtained through segmentation based on energy functions, such as: Graph Cut, GrabCut and other methods; the damage detection of the neural area can be obtained through a machine learning classification model, such as: support vector machine, decision tree, random forest and other models.

[0110] S5. Use a preset etiology identification network to classify the damaged neural tissue, obtain the stroke cause corresponding to the stroke patient, associate the stroke cause with the clinical indicators corresponding to the stroke patient, and generate a cause report corresponding to the stroke patient.

[0111] The present invention utilizes a preset cause identification network to classify the lesions of the damaged neural tissue and obtain the cause of stroke corresponding to the stroke patient, thereby being able to provide individualized cause diagnosis and treatment plans for ischemic stroke patients to improve efficacy and reduce risks.

[0112] Among them, the preset cause identification network refers to a neural network model designed and trained in advance, which is used to identify and classify the causes of stroke. This network model uses a large amount of neuroimaging data for learning during the training phase to capture the features and patterns related to different causes of stroke. The model is usually composed of multiple layers, such as input layer, feature extraction layer, feature selection layer and lesion classification layer.

[0113] As an embodiment of the present invention, the use of a preset cause identification network to classify lesions of the damaged neural tissue to obtain the cause of stroke corresponding to the stroke patient includes: inputting image data corresponding to the damaged neural tissue into an input layer in a preset cause identification network to obtain input image data; using a feature extraction layer in the preset cause identification network to extract key features in the input image data; using a feature selection layer in the preset cause identification network to filter the key features to obtain filtered features; using a lesion classification layer in the preset cause identification network to classify lesions on the filtered features to obtain the cause of stroke corresponding to the stroke patient.

[0114] Among them, the input image data refers to the neural tissue image data of stroke patients; the key features refer to the important features related to the lesions extracted from the input image data by the feature extraction layer, such as the number of blood vessels, blood vessel diameter, blood vessel length, thrombus morphological characteristics, etc.; the screening features refer to the most discriminative and relevant features screened and retained by the feature selection layer; the cause of stroke refers to the root cause of stroke, such as hemorrhage, infarction or ischemia.

[0115] Furthermore, the present invention generates a cause report corresponding to the stroke patient by associating information between the cause of the stroke and the clinical indicators corresponding to the stroke patient, and can provide personalized diagnosis and treatment recommendations, provide a comprehensive condition assessment, promote research and knowledge accumulation, and support decision-making and resource allocation, thereby helping to improve the quality and outcome of care for stroke patients.

[0116] Among them, the clinical indicators refer to quantitative indicators or parameters used to evaluate the condition and physiological state of stroke patients, including blood pressure, heart rate, body temperature, blood test results (such as hemoglobin level, diabetes indicators, etc.), neuroimaging features (such as the size and location of brain damage, etc.) and other clinical measurements related to stroke; the cause report refers to a comprehensive report generated based on the patient's clinical indicators and lesion classification results, which includes content describing and explaining the cause of stroke. Optionally, the information association between the cause of stroke and the clinical indicators corresponding to the stroke patient can be obtained through data mining methods, such as: association rule mining, cluster analysis, principal component analysis and other methods.

[0117] First, by acquiring brain nerve images of stroke patients, the present invention can more accurately locate abnormal areas, detect ischemic lesions, and evaluate the degree of damage to nerve tissue, and provide patients with accurate medical advice and treatment plans, thereby helping to improve the subsequent treatment effect. At the same time, by determining the vascular tissue area in the CT scan image, the present invention can evaluate the patency of the blood vessels, abnormal expansion or stenosis, and provide important references for surgical planning, interventional treatment and drug selection, thereby helping patients receive effective treatment in a timely manner, reducing the risk of complications, and improving the success rate of treatment and patient survival rate. By identifying the thrombus pores in the MRI scan image, the present invention can help doctors accurately and quickly locate and diagnose thrombi in patients. The present invention can identify abnormal brain tissue areas in the brain nerve image based on the vascular load value and the thrombus shape coefficient, and accurately locate abnormal brain tissue areas in the brain nerve image, thereby helping to quickly diagnose and treat abnormal brain conditions in patients, and help to gain an in-depth understanding of the occurrence mechanism and pathological characteristics of abnormal brain phenomena. Furthermore, the present invention classifies lesions of the damaged nerve tissue by utilizing a preset etiology recognition network, and obtains the corresponding cause of stroke for the stroke patient, which can provide an individualized etiology diagnosis and treatment plan for ischemic stroke patients, so as to improve efficacy and reduce risks. Therefore, the method and system for etiology recognition of ischemic stroke based on neuroimaging proposed in the present invention can improve the accuracy of etiology recognition of ischemic stroke.

[0118] like Figure 2 , which is a functional module diagram of a system for identifying the cause of ischemic stroke based on neuroimaging provided by an embodiment of the present invention.

[0119] The system 200 for identifying the cause of ischemic stroke based on neuroimaging of the present invention can be installed in an electronic device. According to the functions implemented, the system 200 for identifying the cause of ischemic stroke based on neuroimaging can include an image acquisition module 201, a load calculation module 202, a coefficient calculation module 203, a damage detection module 204 and a report generation module 205. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0120] In this embodiment, the functions of each module / unit are as follows:

[0121] The image acquisition module 201 is used to acquire brain nerve images of stroke patients, and the brain nerve images include CT scan images and MRI scan images;

[0122] The load calculation module 202 is used to determine the vascular tissue area in the CT scan image, identify the cerebral artery high density sign in the vascular tissue area, extract characteristic parameters of the cerebral artery high density sign, the characteristic parameters include the number of blood vessels, blood vessel diameter and blood vessel length, and calculate the vascular load value corresponding to the CT scan image based on the characteristic parameters;

[0123] The coefficient calculation module 203 is used to identify the thrombus pores in the MRI scan image, determine the thrombus morphological features in the MRI scan image based on the thrombus pores, extract the thrombus penetration parameters corresponding to the thrombus morphological features, and calculate the thrombus shape coefficient corresponding to the MRI scan image based on the thrombus penetration parameters;

[0124] The damage detection module 204 is used to determine the abnormal brain tissue area in the brain nerve image based on the vascular load value and the thrombus shape coefficient, perform ischemic area recognition on the abnormal brain tissue area to obtain the brain ischemic area, and perform damage detection on the brain ischemic area to obtain damaged nerve tissue;

[0125] The report generation module 205 is used to use a preset cause identification network to classify the lesions of the damaged neural tissue, obtain the cause of stroke corresponding to the stroke patient, associate the information of the cause of stroke with the clinical indicators corresponding to the stroke patient, and generate a cause report corresponding to the stroke patient.

[0126] In detail, each module described in the system 200 for realizing the cause identification of ischemic stroke based on neuroimaging in the embodiment of the present invention adopts the same technical means as the method for realizing the cause identification of ischemic stroke based on neuroimaging described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0127] like Figure 3 , which is a schematic diagram of the structure of an electronic device for realizing a method for identifying the cause of ischemic stroke based on neuroimaging according to the present invention.

[0128] The electronic device may include a processor 30, a memory 31, a communication bus 32 and a communication interface 33, and may also include a computer program stored in the memory 31 and executable on the processor 30, such as an engineering safety supervision program based on artificial intelligence.

[0129] In some embodiments, the processor 30 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor 30 is the control core (ControlUnit) of the electronic device, and uses various interfaces and lines to connect the various components of the entire electronic device, and executes or executes programs or modules stored in the memory 31 (for example, executing engineering safety supervision programs based on artificial intelligence, etc.), and calls the data stored in the memory 31 to execute various functions of the electronic device and process data.

[0130] The memory 31 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 31 may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory 31 may also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory 31 may also include both an internal storage unit of the electronic device and an external storage device. The memory 31 may not only be used to store application software and various types of data installed in the electronic device, such as the code of the database configuration connection program, but also be used to temporarily store data that has been output or is to be output.

[0131] The communication bus 32 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 31 and at least one processor 30, etc.

[0132] The communication interface 33 is used for communication between the above-mentioned electronic device 3 and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface may also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visual user interface.

[0133] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0134] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 30 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0135] It should be understood that the embodiment is for illustrative purposes only and the scope of the patent invention is not limited to this structure.

[0136] The database configuration connection program stored in the memory 31 in the electronic device is a combination of multiple computer programs. When running in the processor 30, it can achieve:

[0137] Acquiring a brain nerve image of a stroke patient, wherein the brain nerve image includes a CT scan image and an MRI scan image;

[0138] Determine a vascular tissue area in the CT scan image, identify a cerebral artery high-density sign in the vascular tissue area, extract characteristic parameters of the cerebral artery high-density sign, the characteristic parameters including the number of blood vessels, blood vessel diameter, and blood vessel length, and calculate a vascular load value corresponding to the CT scan image based on the characteristic parameters;

[0139] identifying thrombus pores in the MRI scan image, determining thrombus morphological features in the MRI scan image based on the thrombus pores, extracting thrombus penetration parameters corresponding to the thrombus morphological features, and calculating thrombus shape coefficients corresponding to the MRI scan image based on the thrombus penetration parameters;

[0140] Based on the vascular load value and the thrombus shape coefficient, determining the abnormal brain tissue area in the brain nerve image, performing ischemic area recognition on the abnormal brain tissue area to obtain a brain ischemic area, and performing damage detection on the brain ischemic area to obtain damaged nerve tissue;

[0141] The damaged nerve tissue is classified by using a preset cause identification network to obtain the cause of stroke corresponding to the stroke patient, the cause of stroke and the clinical indicators corresponding to the stroke patient are correlated to generate a cause report corresponding to the stroke patient.

[0142] Specifically, the specific implementation method of the processor 30 for the above computer program can refer to Figure 1 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0143] Furthermore, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium. The storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0144] The present invention further provides a storage medium, wherein the readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0145] Acquiring a brain nerve image of a stroke patient, wherein the brain nerve image includes a CT scan image and an MRI scan image;

[0146] Determine a vascular tissue area in the CT scan image, identify a cerebral artery high-density sign in the vascular tissue area, extract characteristic parameters of the cerebral artery high-density sign, the characteristic parameters including the number of blood vessels, blood vessel diameter, and blood vessel length, and calculate a vascular load value corresponding to the CT scan image based on the characteristic parameters;

[0147] identifying thrombus pores in the MRI scan image, determining thrombus morphological features in the MRI scan image based on the thrombus pores, extracting thrombus penetration parameters corresponding to the thrombus morphological features, and calculating thrombus shape coefficients corresponding to the MRI scan image based on the thrombus penetration parameters;

[0148] Based on the vascular load value and the thrombus shape coefficient, determining the abnormal brain tissue area in the brain nerve image, performing ischemic area recognition on the abnormal brain tissue area to obtain a brain ischemic area, and performing damage detection on the brain ischemic area to obtain damaged nerve tissue;

[0149] The damaged nerve tissue is classified by using a preset cause identification network to obtain the cause of stroke corresponding to the stroke patient, the cause of stroke and the clinical indicators corresponding to the stroke patient are correlated to generate a cause report corresponding to the stroke patient.

[0150] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0151] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0152] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0153] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0154] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0155] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0156] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for identifying the cause of ischemic stroke based on neuroimaging, characterized in that: The method comprises: Acquiring a brain nerve image of a stroke patient, wherein the brain nerve image includes a CT scan image and an MRI scan image; Determine a vascular tissue area in the CT scan image, identify a cerebral artery high-density sign in the vascular tissue area, extract characteristic parameters of the cerebral artery high-density sign, the characteristic parameters including the number of blood vessels, blood vessel diameter, and blood vessel length, and calculate a vascular load value corresponding to the CT scan image based on the characteristic parameters; identifying thrombus pores in the MRI scan image, determining thrombus morphological features in the MRI scan image based on the thrombus pores, extracting thrombus penetration parameters corresponding to the thrombus morphological features, and calculating thrombus shape coefficients corresponding to the MRI scan image based on the thrombus penetration parameters; The determining of the thrombus morphological features in the MRI scan image based on the thrombus pores includes: Locating the thrombus area and the pore area corresponding to the thrombus pore; Performing correlation analysis on the thrombus region and the pore region to obtain correlation parameters corresponding to the thrombus pore; Extracting the morphological factors corresponding to the associated parameters; Based on the morphological factors, determining thrombus morphological features in the MRI scan image; The calculating, based on the thrombus penetration parameter, a thrombus shape coefficient corresponding to the MRI scan image comprises: The thrombus shape coefficient corresponding to the MRI scan image is calculated using the following formula: Wherein, XSI represents the thrombus shape coefficient corresponding to the MRI scan image, M represents the total number of thrombus pixels in the MRI scan image, j represents the index of the thrombus pixel in the MRI scan image, T represents the total time of the MRI scan process, ρ(t) refers to the thrombus permeability parameter, which represents the thrombus permeability at time t, and A represents the area of ​​the thrombus pixel; Based on the vascular load value and the thrombus shape coefficient, determining the abnormal brain tissue area in the brain nerve image, performing ischemic area recognition on the abnormal brain tissue area to obtain a brain ischemic area, and performing damage detection on the brain ischemic area to obtain damaged nerve tissue; The damaged nerve tissue is classified by using a preset cause identification network to obtain the cause of stroke corresponding to the stroke patient, the cause of stroke and the clinical indicators corresponding to the stroke patient are correlated to generate a cause report corresponding to the stroke patient.

2. The method for identifying the cause of ischemic stroke based on neuroimaging as claimed in claim 1, characterized in that: The step of identifying the cerebral artery high density sign in the vascular tissue region comprises: separating the vascular tissue in the vascular tissue area, and detecting high-density signals corresponding to the vascular tissue; Performing signal screening on the high-density signal to obtain a density screening signal; The cerebral artery corresponding to the density screening signal is queried, and the cerebral artery high-density sign corresponding to the cerebral artery is identified.

3. The method for identifying the cause of ischemic stroke based on neuroimaging as claimed in claim 1, characterized in that: The calculating, based on the characteristic parameter, the blood vessel load value corresponding to the CT scan image comprises: The vascular load value corresponding to the CT scan image is calculated using the following formula: Wherein, Vz represents the vascular load value corresponding to the CT scan image, N represents the number of features corresponding to the feature parameter, i represents the feature index corresponding to the feature parameter, Sm i represents the number of blood vessels in the i-th characteristic parameter, Dz i represents the blood vessel diameter in the i-th characteristic parameter, Lc i represents the length of the blood vessel in the i-th characteristic parameter.

4. The method for identifying the cause of ischemic stroke based on neuroimaging as claimed in claim 1, characterized in that: The determining of the abnormal brain tissue area in the brain nerve image based on the vascular load value and the thrombus shape coefficient includes: Querying characteristic indicators corresponding to the vascular load value and the thrombus shape coefficient; Based on the characteristic index, a reference mark is performed on the brain tissue region in the brain nerve image to obtain a reference label; querying an abnormality index in the benchmark tag; Based on the abnormality index, an abnormal brain tissue region in the brain nerve image is determined.

5. The method for identifying the cause of ischemic stroke based on neuroimaging as claimed in claim 1, characterized in that: The step of identifying the ischemic region of the abnormal brain tissue region to obtain the brain ischemic region includes: Converting the abnormal brain tissue region into a binary image, and determining a seed point in the binary image; Using a preset growth rule, the seed point is regularly expanded to obtain an expanded pixel area; Extracting ischemic pixels in the extended pixel area; Based on the ischemic pixels, the ischemic region of the abnormal brain tissue area is identified to obtain the brain ischemic region.

6. The method for identifying the cause of ischemic stroke based on neuroimaging as claimed in claim 1, characterized in that: The step of performing damage detection on the ischemic area of ​​the brain to obtain damaged neural tissue comprises: Analyzing the regional dimension corresponding to the brain ischemic area; Normalizing the regional dimensions to obtain a unified dimension corresponding to the brain ischemic region; Extracting dimensional features in the unified dimension; Performing linear dimensionality reduction on the dimensional features to obtain dimensionality reduction feature parameters; Based on the dimension reduction feature parameters, querying the neural region in the brain ischemic region; The nerve region is subjected to damage detection to obtain damaged nerve tissue.

7. The method for identifying the cause of ischemic stroke based on neuroimaging as claimed in claim 1, characterized in that: The method of using a preset cause identification network to classify the damaged nerve tissue to obtain the cause of stroke corresponding to the stroke patient includes: Inputting the image data corresponding to the damaged nerve tissue into an input layer of a preset etiology identification network to obtain input image data; Extracting key features from the input image data using a feature extraction layer in the preset etiology identification network; Using the feature selection layer in the preset etiology identification network to screen the key features to obtain screening features; The lesion classification layer in the preset cause identification network is used to classify the lesions of the screening features to obtain the cause of stroke corresponding to the stroke patient.

8. A system for identifying the cause of ischemic stroke based on neuroimaging, characterized in that: The system is used to implement the method for identifying the cause of ischemic stroke based on neuroimaging as described in any one of claims 1 to 7, the system comprising: An image acquisition module is used to acquire brain nerve images of stroke patients, wherein the brain nerve images include CT scan images and MRI scan images; a load calculation module, configured to determine a vascular tissue region in the CT scan image, identify a cerebral artery high-density sign in the vascular tissue region, extract characteristic parameters from the cerebral artery high-density sign, wherein the characteristic parameters include the number of blood vessels, the diameter of the blood vessels, and the length of the blood vessels, and calculate a vascular load value corresponding to the CT scan image based on the characteristic parameters; a coefficient calculation module, configured to identify thrombus pores in the MRI scan image, determine thrombus morphological features in the MRI scan image based on the thrombus pores, extract thrombus penetration parameters corresponding to the thrombus morphological features, and calculate a thrombus shape coefficient corresponding to the MRI scan image based on the thrombus penetration parameters; A damaged detection module is used to determine the abnormal brain tissue area in the brain nerve image based on the vascular load value and the thrombus shape coefficient, perform ischemic area identification on the abnormal brain tissue area to obtain the brain ischemic area, and perform damage detection on the brain ischemic area to obtain damaged nerve tissue; a report generation module is used to use a preset cause identification network to perform lesion classification on the damaged nerve tissue to obtain the cause of stroke corresponding to the stroke patient, perform information association between the cause of stroke and the clinical indicators corresponding to the stroke patient, and generate a cause report corresponding to the stroke patient.

Images