Neurosurgery medical image analysis method and system based on image processing

Abstract

The present invention provides a neurosurgery medical image analysis method and system based on image processing, which relates to the technical field of medical image processing. The method includes: collecting medical images to be processed through medical imaging equipment; performing image preprocessing on the medical images to be processed to obtain processed medical images; determining vascular parameters according to the processed medical images, wherein the vascular parameters include: vascular diameter and vascular wall thickness; determining vascular safety detection scores according to the vascular parameters; determining lesion feature discrimination coefficients according to the processed medical images; and generating medical image analysis reports according to the vascular safety detection scores and lesion feature discrimination coefficients. According to the present invention, medical images can be comprehensively analyzed from two aspects: vascular safety status and lesion features, thereby improving the comprehensiveness and accuracy of neurosurgery medical image analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a neurosurgery medical image analysis method and system based on image processing. Background Art

[0002] In the related technology, CN116823701A discloses an AI-based medical image analysis and processing method, which marks the medical image and stores it in the corresponding preset image processing channel, and performs image restoration processing according to the image quality of the medical image; divides the medical image into several image sub-screens, and determines all target areas in the image sub-screens according to the image elements in the image sub-screens; compares and processes different biological tissue feature information to determine the target area where abnormal biological tissue exists, obtains the location information of the lesion source in the corresponding medical image, and then determines the location information corresponding to the lesion source in the patient's body; the above method performs restoration processing and segmentation processing on the medical image, which can effectively reduce the noise component of the medical image and facilitate the subsequent synchronous processing of different sub-screen parts of the medical image. It also performs identification and comparison processing on the biological tissue features of the medical image, accurately locates the lesion source, and improves the accuracy of medical image processing.

[0003] CN118365610B discloses a multimodal medical image data analysis method based on machine learning, which specifically includes: MRI data collection and preprocessing, MRI feature extraction, clinical diagnosis model of trigeminal neuralgia, and model result analysis. The method described in this scheme uses machine learning to extract features from medical images and establish a clinical diagnosis model, demonstrating its potential application value in assisting the diagnosis of complex diseases.

[0004] Based on the above related technologies, the accuracy of medical image processing can be improved. However, the related technologies do not take into account the importance of vascular safety status in medical images, that is, it is impossible to comprehensively analyze medical images based on the two aspects of vascular safety status and lesion characteristics.

[0005] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The present invention provides a neurosurgery medical image analysis method and system based on image processing, which can solve the technical problem that related technologies cannot comprehensively analyze medical images based on the two aspects of vascular safety status and lesion characteristics.

[0007] According to a first aspect of the present invention, a neurosurgery medical image analysis method based on image processing is provided, comprising:

[0008] Collecting medical images to be processed through medical imaging equipment;

[0009] Performing image preprocessing on the medical image to be processed to obtain a processed medical image;

[0010] Determining blood vessel parameters according to the processed medical image, wherein the blood vessel parameters include: blood vessel diameter and blood vessel wall thickness;

[0011] Determining a vascular safety detection score according to the vascular parameters;

[0012] Determining a lesion feature discrimination coefficient based on the processed medical image;

[0013] A medical image analysis report is generated based on the vascular safety detection score and the lesion characteristic discrimination coefficient.

[0014] According to a second aspect of the present invention, there is provided a neurosurgery medical image analysis system based on image processing, comprising:

[0015] An image acquisition module, used for acquiring medical images to be processed through medical imaging equipment;

[0016] An image processing module, used for performing image preprocessing on the medical image to be processed to obtain a processed medical image;

[0017] A vascular parameter module, used to determine vascular parameters according to the processed medical image, wherein the vascular parameters include: vascular diameter and vascular wall thickness;

[0018] A blood vessel detection module, used to determine a blood vessel safety detection score according to the blood vessel parameters;

[0019] A lesion feature module, used to determine a lesion feature discrimination coefficient based on the processed medical image;

[0020] The analysis report module is used to generate a medical image analysis report based on the vascular safety detection score and the lesion characteristic discrimination coefficient.

[0021] Technical effect: According to the present invention, the collected medical images can be preprocessed to improve the accuracy of subsequent image analysis results, the vascular safety status can be evaluated according to the vascular parameters of the blood vessels in the processed medical images, and the lesion characteristics in the processed medical images can be identified through the image detection model to determine the lesion characteristic discrimination coefficient, thereby improving the objectivity of the neurosurgery medical image analysis results. When determining the vascular wall thickness relationship function, the vascular wall thickness relationship function can be determined according to the historical patient gender recognition results, the historical patient age and the historical patient vascular wall thickness, accurately describing the relationship between the historical patient gender, the historical patient age and the historical patient vascular wall thickness, and fully considering the overall trend of the vascular wall thickness increasing with age and the influence of gender factors on the vascular wall thickness in the calculation process, thereby improving the accuracy and objectivity of the vascular wall thickness relationship function. When determining the vascular safety detection score, the vascular safety detection score can be determined according to the normal vascular wall thickness, vascular diameter and vascular wall thickness. In the calculation process, the vascular safety detection score can be determined according to the overall vascular stenosis degree and the overall vascular wall thickness safety status in the medical image, thereby improving the accuracy and comprehensiveness of the vascular safety detection score. When determining the lesion feature discrimination coefficient, the image detection model can be used to detect various elements in the medical image, so as to judge the lesion features shown in the medical image from multiple aspects, thereby improving the accuracy and objectivity of the lesion feature discrimination coefficient.

[0022] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and not limiting of the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work.

[0024] Figure 1 A schematic diagram of a process flow of a neurosurgery medical image analysis method based on image processing according to an embodiment of the present invention is exemplarily shown;

[0025] Figure 2 A block diagram of a neurosurgery medical image analysis system based on image processing according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0028] Figure 1 The following is a flow chart of a method for analyzing a neurosurgery medical image based on image processing according to an embodiment of the present invention, wherein the method comprises:

[0029] Step S101, collecting medical images to be processed by medical imaging equipment;

[0030] Step S102, performing image preprocessing on the medical image to be processed to obtain a processed medical image;

[0031] Step S103, determining blood vessel parameters according to the processed medical image, wherein the blood vessel parameters include: blood vessel diameter and blood vessel wall thickness;

[0032] Step S104, determining a blood vessel safety detection score according to the blood vessel parameters;

[0033] Step S105, determining a lesion feature discrimination coefficient according to the processed medical image;

[0034] Step S106, generating a medical image analysis report according to the vascular safety detection score and the lesion characteristic discrimination coefficient.

[0035] According to the neurosurgery medical image analysis method based on image processing in the embodiment of the present invention, the collected medical images can be pre-processed to improve the accuracy of subsequent image analysis results, the vascular safety status can be evaluated based on the vascular parameters of the blood vessels in the processed medical images, and the lesion features in the processed medical images can be identified through the image detection model to determine the lesion feature discrimination coefficient, thereby improving the objectivity of the neurosurgery medical image analysis results.

[0036] According to an embodiment of the present invention, in step S101, a medical image to be processed is acquired by a medical imaging device.

[0037] For example, medical images of the inside of the human body (e.g., brain blood vessels, skull, and spinal cord) are collected using DSA technology or through medical imaging equipment such as X-ray machines, CT scanners, and MR scanners.

[0038] According to an embodiment of the present invention, in step S102, image preprocessing is performed on the medical image to be processed to obtain a processed medical image.

[0039] For example, since the medical images to be processed often contain a large amount of noise and artifacts, preprocessing is required. The preprocessing mainly uses filtering and other techniques to remove random noise in the image, perform contrast stretching, edge sharpening and other means to improve the visual effect of the image, and correct the distortion in the image (such as collective distortion and grayscale distortion).

[0040] According to an embodiment of the present invention, in step S103, blood vessel parameters are determined according to the processed medical image, wherein the blood vessel parameters include: blood vessel diameter and blood vessel wall thickness.

[0041] According to one embodiment of the present invention, step S103 includes:

[0042] In the processed medical image, a plurality of first sampling points are set;

[0043] In the processed medical image, obtaining blood vessel diameters at a plurality of first sampling points by using a first image detection model;

[0044] In the processed medical image, a plurality of second sampling points are set;

[0045] In the processed medical image, the blood vessel wall thickness at a plurality of second sampling points is obtained through a first image detection model.

[0046] For example, DSA (digital subtraction angiography) is mainly used to observe the structure and pathological conditions of cerebral blood vessels. For the pure blood vessel image generated by DSA after preprocessing, multiple first sampling points are set in the pure blood vessel image, wherein the first sampling point can represent the diameter characteristics of the entire blood vessel or a specific blood vessel segment (e.g., in the anterior cerebral artery and middle cerebral artery images, the first sampling point should be set at the horizontal segment of the anterior cerebral artery and the horizontal segment of the middle cerebral artery), and the selection of locations that may lead to inaccurate diameter measurement, such as vascular bends, bifurcations or severe lesions, should be avoided. The first image detection model is a deep learning neural network model, which can be used to detect the first An image detection model is trained so that the first image detection model can identify the vessel diameter and vessel wall thickness of the blood vessels in the medical image. The vessel diameter at the first sampling point is identified through the first image detection model. A plurality of second sampling points are set in the pure vessel image, wherein the second sampling point should be able to represent the wall thickness characteristics of the entire blood vessel or a specific blood vessel segment (e.g., in the common carotid artery image, the second sampling point should be set at the posterior wall at the distal end of the common carotid artery), and selection should be avoided in areas where obvious lesions (e.g., plaques and calcifications) exist in the vessel wall. The vessel wall thickness at the second sampling point is identified through the first image detection model.

[0047] According to an embodiment of the present invention, in step S104, a blood vessel safety detection score is determined according to the blood vessel parameters.

[0048] According to one embodiment of the present invention, step S104 includes:

[0049] Acquiring historical personal information of historical patients, wherein the historical personal information includes: the thickness of the blood vessel wall of the historical patients, the gender of the historical patients, and the age of the historical patients;

[0050] Determine the historical patient gender identification result according to the historical patient gender;

[0051] Determine a blood vessel wall thickness relationship function of the historical patient gender identification result, the historical patient age and the blood vessel wall thickness of the historical patient;

[0052] Obtaining personal information of the patient to whom the processed medical images belong, wherein the personal information includes: patient gender and patient age;

[0053] Determining normal vascular wall thickness according to the patient's gender, the patient's age and the vascular wall thickness relationship function;

[0054] A blood vessel safety detection score is determined according to the normal blood vessel wall thickness, the blood vessel diameter and the blood vessel wall thickness.

[0055] For example, historical personal information of historical patients whose historical medical images are of the same part as that of the current medical image is obtained in the historical database, the historical personal information including the historical patient gender, historical patient age and historical patient vascular wall thickness (the average value of the vascular wall thickness at each sampling point in the historical medical image of the historical patient); according to the historical patient gender of the historical patient, the historical patient gender recognition result is determined, for example, when the historical patient gender is male, the historical patient gender recognition result is 1, and when the historical patient gender is female, the historical patient gender recognition result is 0; vascular wall thickness is related to age and gender to a certain extent, for example, males have thicker vascular walls, and older people have thicker vascular walls. Based on the correlation of the above data, a vascular wall thickness relationship function between the historical patient gender recognition result, the historical patient age and the historical patient vascular wall thickness can be determined; the gender and age of the patient to whom the current medical image belongs are obtained; the gender and age of the patient to whom the current medical image belongs are substituted into the vascular wall thickness relationship function to determine the normal vascular wall thickness under normal circumstances; based on the normal vascular wall thickness, vascular diameter and vascular wall thickness, the safety status of the blood vessels shown in the medical image is evaluated to determine the vascular safety detection score.

[0056] According to a first embodiment of the present invention, determining a vascular wall thickness relationship function of the historical patient gender recognition result, the historical patient age and the historical patient vascular wall thickness includes: determining an undetermined coefficient equation of the vascular wall thickness relationship function according to formula (1),

[0057] (1),

[0058] in, is the historical patient vascular wall thickness of the kth historical patient, is the historical patient age of the kth historical patient, is the historical patient gender recognition result of the kth historical patient, , , , and is the unknown coefficient of the unknown coefficient equation, K is the number of historical patients, k≤K, k and K are both positive integers;

[0059] Solving the undetermined coefficients according to the historical patient gender identification results, the historical patient ages, and the historical patient vascular wall thicknesses to obtain solution values ​​of the undetermined coefficients;

[0060] The blood vessel wall thickness relationship function is determined according to the solved value of the undetermined coefficient and the undetermined coefficient equation.

[0061] According to one embodiment of the present invention, is the average vascular wall thickness of K historical patients, represents the vascular wall thickness of the benchmark historical patients, The thickness of blood vessel walls changes with age. It represents the linear variation of vascular wall thickness with age, thereby expressing the overall trend that vascular wall thickness increases with age. For example, the structure and function of the vascular wall gradually degenerate with age, and the vascular wall may gradually thicken. It indicates that the historical patient vascular wall thickness of the k-th historical patient and the historical patient gender identification result of the k-th historical patient are positively correlated. When the historical patient gender identification result of the k-th historical patient is 1 (the historical patient gender of the k-th historical patient is male), the larger the historical patient vascular wall thickness of the k-th historical patient is (for example, males may be more likely to have vascular wall thickening than females). According to the overall trend of vascular wall thickness increasing with age and the influence of gender factors on vascular wall thickness, the equation of undetermined coefficients is obtained to express the relationship between the historical patient gender identification result, the historical patient age and the historical patient vascular wall thickness.

[0062] According to one embodiment of the present invention, fitting can be performed based on multiple parameters involved in the above undetermined coefficient equation, that is, fitting can be performed based on the historical patient gender identification results, historical patient age and historical patient vascular wall thickness of at least 4 historical patients to solve the above multiple undetermined coefficients. There are 4 undetermined coefficients, namely, , , and According to the historical patient gender identification results, historical patient age and historical patient vascular wall thickness, the above four undetermined coefficients are solved to obtain the solution values ​​of the above four undetermined coefficients, and the solution values ​​of the above four undetermined coefficients are substituted into the undetermined coefficient equation to determine the vascular wall thickness relationship function.

[0063] In this way, the vascular wall thickness relationship function can be determined based on the historical patient gender identification results, historical patient age and historical patient vascular wall thickness, accurately describing the relationship between historical patient gender, historical patient age and historical patient vascular wall thickness. In the calculation process, the overall trend of vascular wall thickness increasing with age and the influence of gender factors on vascular wall thickness are fully considered, thereby improving the accuracy and objectivity of the vascular wall thickness relationship function.

[0064] According to an embodiment of the present invention, determining a blood vessel safety detection score according to the normal blood vessel wall thickness, the blood vessel diameter and the blood vessel wall thickness includes: determining the blood vessel safety detection score according to formula (2): ,

[0065] ,

[0066] Among them, max is the maximum value function, if is the conditional function, and is the preset weight, is the blood vessel diameter at the i-th first sampling point, is the preset blood vessel diameter threshold, is the preset ratio threshold, is the vascular wall thickness at the rth second sampling point, is the normal blood vessel wall thickness, n is the number of the first sampling points, i≤n, R is the number of the second sampling points, r≤R, i, n, r and R are all positive integers.

[0067] According to one embodiment of the present invention, is the relative difference between the preset blood vessel diameter threshold and the blood vessel diameter at the i-th first sampling point. The larger the ratio, the smaller the blood vessel diameter at the i-th first sampling point, the narrower the blood vessel at the i-th first sampling point, and the lower the blood vessel safety. In formula (2), the conditional function The value of includes the following two cases: When the condition is not met, it means that the medical image shows severe stenosis or complete occlusion of the blood vessels, the blood vessel safety is low, and the value of the condition function is 0. , it means that there is no severe stenosis or complete occlusion of blood vessels in the medical image, and the value of the conditional function is , is the preset blood vessel diameter threshold and the average relative difference of the blood vessel diameters at the multiple first sampling points, It is the overall degree of vascular stenosis in medical images. The larger the value, the more serious the overall degree of vascular stenosis. This means that as the average relative difference increases, the impact on the overall degree of vascular stenosis in medical images increases at an accelerated rate rather than at a uniform rate.

[0068] According to one embodiment of the present invention, is the relative difference between the normal blood vessel wall thickness and the blood vessel wall thickness at the rth second sampling point. The larger the ratio, the greater the difference between the blood vessel wall thickness at the rth second sampling point and the normal blood vessel wall thickness. Severe thickening of the blood vessel wall can lead to stenosis of the lumen, which hinders the passage of blood and affects the normal blood supply to the organs. Thinning of the blood vessel wall can affect the regulation of blood pressure, leading to unstable blood pressure, possible hypertension or hypotension, and low blood vessel safety. represents the safety status of the vascular wall thickness at the rth second sampling point, Indicates the overall blood vessel wall thickness safety status in medical images.

[0069] According to one embodiment of the present invention, To determine the vascular safety detection score based on the overall vascular stenosis degree and overall vascular wall thickness safety status in medical images.

[0070] In this way, the vascular safety detection score can be determined based on the normal vascular wall thickness, vascular diameter and vascular wall thickness. During the calculation process, the vascular safety detection score can be determined based on the overall vascular stenosis degree and the overall vascular wall thickness safety status in the medical image, thereby improving the accuracy and comprehensiveness of the vascular safety detection score.

[0071] According to an embodiment of the present invention, in step S105, a lesion feature discrimination coefficient is determined based on the processed medical image.

[0072] According to one embodiment of the present invention, step S105 includes:

[0073] Determining a lesion feature discrimination coefficient based on the processed medical image includes:

[0074] Determine tumor identification results in processed medical images using image detection models , hematoma identification results , hydrocephalus identification results and brain atrophy identification results , and then determine the lesion characteristic discrimination coefficient.

[0075] For example, the image detection model is a deep learning neural network model. The image detection model can be trained through features related to tumors, hematomas, hydrocephalus and brain atrophy in historical medical images (such as size features, texture features and grayscale features), so that the image detection model can identify whether there are tumors, hematomas, hydrocephalus and brain atrophy in the images. In the preprocessed medical images, the image detection model is used to identify whether there is a tumor in the processed medical images, and the tumor identification result is determined; the image detection model is used to identify whether there is a hematoma in the processed medical images, and the hematoma identification result is determined; the image detection model is used to identify whether there is hydrocephalus in the processed medical images, and the hydrocephalus identification result is determined; the image detection model is used to identify whether there is brain atrophy in the processed medical images, and the brain atrophy identification result is determined; the tumor identification result, hematoma identification result, hydrocephalus identification result and brain atrophy identification result are summed up to determine the lesion feature discrimination coefficient.

[0076] According to one embodiment of the present invention, in the processed medical image, the tumor recognition result is determined by the image detection model. , hematoma identification results , hydrocephalus identification results and brain atrophy identification results , and then determine the lesion characteristic discrimination coefficient, including: determining the lesion characteristic discrimination coefficient according to formula (3) ,

[0077] (3)

[0078] in,

[0079] , , , .

[0080] According to one embodiment of the present invention, is the tumor recognition result. If a tumor appears in the medical image, then The value of is 1, otherwise The value of is 0, is the hematoma identification result. If a hematoma appears in the medical image, The value of is 3, otherwise The value of is 0, is the hydrocephalus identification result. If hydrocephalus appears in the medical image, The value of is 5, otherwise The value of is 0, is the result of brain atrophy identification. If brain atrophy occurs in medical images, The value of is 7, otherwise The value of is 0, , , and The values ​​of are added together to obtain the lesion characteristic discrimination coefficient , the discriminant coefficient can be determined based on the lesion characteristics The value of determines the type of lesion features that appear, such as when When the value of is 1, a tumor appears in the medical image. When the value is 4, hematomas and tumors appear in medical images.

[0081] In this way, multiple elements in medical images can be detected based on the image detection model, so that the lesion characteristics shown in the medical images can be judged from multiple aspects, thereby improving the accuracy and objectivity of the lesion characteristic discrimination coefficient.

[0082] According to an embodiment of the present invention, in step S106, a medical image analysis report is generated according to the vascular safety detection score and the lesion characteristic discrimination coefficient.

[0083] For example, first, the type of lesion is determined based on the lesion feature discrimination coefficient. For example, when the lesion feature discrimination coefficient is 1, it is determined that a tumor exists in the area. Further, the harmfulness caused by the type of lesion is determined based on the vascular safety detection score. For example, when the vascular safety detection score is less than the set vascular safety detection score threshold, it indicates that the harm caused by the lesion feature is greater.

[0084] According to the neurosurgery medical image analysis method based on image processing of the embodiment of the present invention, the collected medical images can be preprocessed to improve the accuracy of the subsequent image analysis results, the vascular safety status can be evaluated according to the vascular parameters of the blood vessels in the processed medical images, and the lesion characteristics in the processed medical images can be identified through the image detection model to determine the lesion characteristic discrimination coefficient, thereby improving the objectivity of the neurosurgery medical image analysis results. When determining the vascular wall thickness relationship function, the vascular wall thickness relationship function can be determined according to the historical patient gender recognition results, the historical patient age and the historical patient vascular wall thickness, accurately describing the relationship between the historical patient gender, the historical patient age and the historical patient vascular wall thickness, and fully considering the overall trend of the vascular wall thickness increasing with age and the influence of gender factors on the vascular wall thickness in the calculation process, thereby improving the accuracy and objectivity of the vascular wall thickness relationship function. When determining the vascular safety detection score, the vascular safety detection score can be determined according to the normal vascular wall thickness, vascular diameter and vascular wall thickness. In the calculation process, the vascular safety detection score can be determined according to the overall vascular stenosis degree and the overall vascular wall thickness safety status in the medical image, thereby improving the accuracy and comprehensiveness of the vascular safety detection score. When determining the lesion feature discrimination coefficient, the image detection model can be used to detect various elements in the medical image, so as to judge the lesion features shown in the medical image from multiple aspects, thereby improving the accuracy and objectivity of the lesion feature discrimination coefficient.

[0085] Figure 2 A block diagram of a neurosurgery medical image analysis system based on image processing according to an embodiment of the present invention is exemplarily shown, wherein the system comprises:

[0086] An image acquisition module, used for acquiring medical images to be processed through medical imaging equipment;

[0087] An image processing module, used for performing image preprocessing on the medical image to be processed to obtain a processed medical image;

[0088] A vascular parameter module, used to determine vascular parameters according to the processed medical image, wherein the vascular parameters include: vascular diameter and vascular wall thickness;

[0089] A blood vessel detection module, used to determine a blood vessel safety detection score according to the blood vessel parameters;

[0090] A lesion feature module, used to determine a lesion feature discrimination coefficient based on the processed medical image;

[0091] The analysis report module is used to generate a medical image analysis report based on the vascular safety detection score and the lesion characteristic discrimination coefficient.

[0092] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0093] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

Claims

1. A neurosurgery medical image analysis method based on image processing, characterized in that: include: Collecting medical images to be processed through medical imaging equipment; Performing image preprocessing on the medical image to be processed to obtain a processed medical image; Determining blood vessel parameters according to the processed medical image, wherein the blood vessel parameters include: blood vessel diameter and blood vessel wall thickness; Determining a vascular safety detection score according to the vascular parameters; Determining a lesion feature discrimination coefficient based on the processed medical image; generating a medical imaging analysis report according to the vascular safety detection score and the lesion characteristic discrimination coefficient; Determining a lesion feature discrimination coefficient based on the processed medical image includes: Determine tumor identification results in processed medical images using image detection models , hematoma identification results , hydrocephalus identification results and brain atrophy identification results , and then determine the lesion feature discrimination coefficient, including: According to the formula ， Determine the lesion characteristics discrimination coefficient , ， ， ， ； In the processed medical image, blood vessel parameters are obtained through an image detection model, including: In the processed medical image, a plurality of first sampling points are set; In the processed medical image, obtaining blood vessel diameters at a plurality of first sampling points by using a first image detection model; In the processed medical image, a plurality of second sampling points are set; In the processed medical image, obtaining the blood vessel wall thickness at a plurality of second sampling points by using a first image detection model; According to the vascular parameters, a vascular safety detection score is determined, including: Acquiring historical personal information of historical patients, wherein the historical personal information includes: the thickness of the blood vessel wall of the historical patients, the gender of the historical patients, and the age of the historical patients; Determine the historical patient gender identification result according to the historical patient gender; Determine a blood vessel wall thickness relationship function of the historical patient gender identification result, the historical patient age and the blood vessel wall thickness of the historical patient; Obtaining personal information of the patient to whom the processed medical images belong, wherein the personal information includes: patient gender and patient age; Determining normal vascular wall thickness according to the patient's gender, the patient's age and the vascular wall thickness relationship function; Determining a blood vessel safety detection score according to the normal blood vessel wall thickness, the blood vessel diameter, and the blood vessel wall thickness; Determining a blood vessel wall thickness relationship function of the historical patient gender recognition result, the historical patient age, and the blood vessel wall thickness of the historical patient includes: According to the formula , Determine the equation of unknown coefficients of the relationship function of the vascular wall thickness, where: is the historical patient vascular wall thickness of the kth historical patient, is the historical patient age of the kth historical patient, is the historical patient gender recognition result of the kth historical patient, , , , and is the unknown coefficient of the unknown coefficient equation, K is the number of historical patients, k≤K, k and K are both positive integers; Solving the undetermined coefficients according to the historical patient gender identification results, the historical patient ages, and the historical patient vascular wall thicknesses to obtain solution values ​​of the undetermined coefficients; The blood vessel wall thickness relationship function is determined according to the solved value of the undetermined coefficient and the undetermined coefficient equation.

2. The neurosurgery medical image analysis method based on image processing according to claim 1, characterized in that: Determining a blood vessel safety detection score according to the normal blood vessel wall thickness, the blood vessel diameter and the blood vessel wall thickness includes: According to the formula ， Determining the Vascular Safety Test Score , where max is the maximum value function and if is the conditional function. and is the preset weight, is the blood vessel diameter at the i-th first sampling point, is the preset blood vessel diameter threshold, is the preset ratio threshold, is the vascular wall thickness at the rth second sampling point, is the normal blood vessel wall thickness, n is the number of the first sampling points, i≤n, R is the number of the second sampling points, r≤R, i, n, r and R are all positive integers.

3. A neurosurgery medical image analysis system based on image processing for executing the method according to any one of claims 1 to 2, characterized in that: include: An image acquisition module, used for acquiring medical images to be processed through medical imaging equipment; An image processing module, used for performing image preprocessing on the medical image to be processed to obtain a processed medical image; A vascular parameter module, used to determine vascular parameters according to the processed medical image, wherein the vascular parameters include: vascular diameter and vascular wall thickness; A blood vessel detection module, used to determine a blood vessel safety detection score according to the blood vessel parameters; A lesion feature module, used to determine a lesion feature discrimination coefficient based on the processed medical image; The analysis report module is used to generate a medical image analysis report based on the vascular safety detection score and the lesion characteristic discrimination coefficient.

Images