Wilson disease damage degree assessment system and method based on artificial intelligence

Abstract

The invention relates to the technical field of data processing, and discloses an artificial intelligence-based Wilson disease damage degree assessment system and method thereof, comprising: obtaining geometric features about the basal ganglia from a first MRI image, and preliminarily segmenting a first segmented image about the basal ganglia part from a second MRI image according to the geometric features; performing a second round of segmentation on the first segmented image according to texture features to obtain a second segmented image; dividing a copper deposition area in the basal ganglia according to the first segmented image and the second segmented image, and calculating a volume ratio of the copper deposition volume to the basal ganglia volume; inputting physiological feature data and the volume ratio into a pre-trained convolutional neural network model to obtain a quantitative score, and assessing the damage degree of a patient affected by Wilson disease according to the size of the quantitative score; the invention can accurately assess the damage degree of the copper deposition to the patient.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to an artificial intelligence-based Wilson disease damage degree assessment system and method. Background Art

[0002] Wilson disease is a rare inherited metabolic disease characterized by abnormal accumulation of copper in the body, especially in the basal ganglia region of the brain. This copper deposition can lead to degenerative changes in the nervous system, which in turn affects the patient's movement, cognition, and mental state. Accurately assessing the extent of damage caused by copper deposition to patients is crucial for early diagnosis, disease monitoring, and formulation of treatment plans for Wilson disease.

[0003] However, existing technologies still face multiple challenges in the diagnosis and damage assessment of Wilson's disease. Although imaging technologies such as MRI can provide high-resolution brain structure images, copper deposition usually occurs in the basal ganglia region, and the anatomical structure of this region is relatively complex. Existing image segmentation methods have difficulty accurately segmenting the basal ganglia region. Furthermore, they lack sufficient accuracy and sensitivity to reveal the specific impact of copper deposition on patients. In addition, based on the identification and quantification of copper deposition areas, the complex correlation between data has not been fully explored, resulting in a lack of effective methods for evaluating copper deposition and the neurological damage it causes. This makes it difficult for doctors to accurately assess the condition and formulate treatment plans in clinical practice, and also restricts the early detection and intervention of Wilson's disease. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an artificial intelligence-based Wilson disease damage degree assessment system and method.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] In a first aspect, a method for assessing the degree of damage of Wilson's disease based on artificial intelligence is provided, the method comprising:

[0007] Acquire a first MRI image and a second MRI image of the patient, acquire geometric features about the basal ganglia from the first MRI image, and preliminarily segment a first segmented image about the basal ganglia from the second MRI image according to the geometric features;

[0008] Acquire texture features of the basal ganglia in the first MRI image, and perform a second round of segmentation on the first segmented image according to the texture features to obtain a second segmented image, wherein the texture features are gray-level co-occurrence matrix features;

[0009] The copper deposition area in the basal ganglia is divided according to the first segmented image and the second segmented image, and after the copper deposition volume is determined according to the copper deposition area, the volume ratio of the copper deposition volume to the basal ganglia volume is calculated;

[0010] The patient's physiological characteristic data is obtained, and the physiological characteristic data and volume ratio are input into a pre-trained convolutional neural network model to obtain a quantitative score of cognitive or neurological function, and the degree of damage to the patient affected by Wilson's disease is evaluated according to the size of the quantitative score of cognitive or neurological function; the degree of damage includes mild, moderate and severe.

[0011] Further, the preliminarily segmenting a first segmented image of the basal ganglia from the second MRI image includes:

[0012] Using a K-means clustering algorithm to distinguish pixels of the second MRI image, and clustering the pixels to form regions as candidate regions, thereby obtaining a plurality of candidate regions;

[0013] Obtaining a first feature parameter set about geometric features of each candidate region, wherein the first feature parameter set includes a plurality of geometric feature vectors, including but not limited to a position feature vector of the basal ganglia relative to the brain region, a shape feature vector of the basal ganglia, an area feature vector of the basal ganglia, and a perimeter feature vector of the basal ganglia;

[0014] Retrieving a second feature parameter set related to geometric features in the first MRI image;

[0015] Calculate the similarity index between the first feature parameter and the second feature parameter of each candidate region by using a preconfigured similarity measurement model to obtain multiple similarity indexes;

[0016] ;

[0017] Where: represents the similarity index, represents the kth geometric eigenvector in the first eigenparameter set, represents the kth geometric eigenvector in the second eigenparameter set, represents the number of geometric eigenvectors;

[0018] The multiple similarity indexes are sorted from large to small according to the numerical values, and the candidate area corresponding to the first sorted similarity index is used as the first segmented image.

[0019] Furthermore, the performing a second round of segmentation on the first segmented image comprises:

[0020] Using an edge detection algorithm to extract boundaries between different levels in the first segmented image, a plurality of boundaries are obtained;

[0021] Marking the hierarchical area surrounded by every two boundaries as a candidate area, and obtaining multiple candidate areas;

[0022] Extracting gray-level co-occurrence matrix features of the basal ganglia in the first MRI image and gray-level co-occurrence matrix features of each candidate region;

[0023] Marking the gray level co-occurrence matrix features of the basal ganglia in the first MRI image as first gray level co-occurrence matrix features, and marking the gray level co-occurrence matrix features of the candidate region as second gray level co-occurrence matrix features;

[0024] Calculate the cosine similarity coefficient between the first gray level co-occurrence matrix feature and the second gray level co-occurrence matrix feature to obtain a plurality of cosine similarity coefficients;

[0025] The multiple cosine similarity coefficients are sorted from large to small in value, and the candidate area corresponding to the first sorted cosine similarity coefficient is marked as the second segmented image.

[0026] Furthermore, dividing the copper deposition area in the basal ganglia includes:

[0027] After rigid registration is performed on the first segmented image and the second segmented image, an AND operation is performed on the first segmented image and the second segmented image to obtain an overlapping area and a non-overlapping area;

[0028] The non-overlapping areas are marked as copper deposition areas.

[0029] Further, determining the copper deposition volume according to the copper deposition area includes:

[0030] Acquiring pixel parameters of the copper deposition area, wherein the pixel parameters include the number of pixels and the pixel volume;

[0031] The copper deposition volume is obtained by calculating the product of the number of pixels and the pixel volume.

[0032] Furthermore, the physiological characteristic data include but are not limited to age, gender, years of education and body mass index (BMI);

[0033] Among them, the training method of the convolutional neural network model is as follows:

[0034] Collecting historical loss degree training data, and dividing the historical loss degree training data into a loss degree training set and a loss degree test set; the historical loss degree training data includes loss degree reflection feature data and its corresponding quantitative score;

[0035] Wherein, the loss degree reflecting characteristic data includes physiological characteristic data and volume ratio;

[0036] Construct a convolutional neural network for performing regression tasks, use the loss degree reflection feature data in the loss degree training set as the input of the convolutional neural network, and use the label of the loss degree reflection feature data as the output of the convolutional neural network, train the convolutional neural network, and obtain an initial convolutional neural network;

[0037] The initial convolutional neural network model is verified using the loss degree test set, and an initial convolutional neural network with a preset test accuracy is output as the trained convolutional neural network model.

[0038] Further, the assessment of the degree of damage to the patient affected by Wilson's disease includes:

[0039] Setting a quantitative scoring threshold, wherein the quantitative scoring threshold includes Td1 and Td2, Td1>Td2>0;

[0040] comparing the quantitative scores of cognitive or neurological function to quantitative score thresholds;

[0041] If the quantitative score of cognitive or neurological function is greater than the quantitative score threshold Td1, the patient is judged to have severe damage due to Wilson's disease;

[0042] If the quantitative score of cognitive or neurological function is less than or equal to the quantitative score threshold Td1, and greater than or equal to the quantitative score threshold Td2, the degree of damage to the patient affected by Wilson's disease is judged to be moderate;

[0043] If the quantitative score of cognitive or neurological function is less than the quantitative score threshold Td2, the degree of impairment of the patient affected by Wilson's disease is judged to be mild.

[0044] In a second aspect, a Wilson disease damage degree assessment system based on artificial intelligence is provided, which is implemented based on the above-mentioned Wilson disease damage degree assessment method based on artificial intelligence, and the system comprises:

[0045] A data acquisition module, for the first MRI image and the second MRI image of the patient, acquires geometric features about the basal ganglia from the first MRI image, and preliminarily segments a first segmented image about the basal ganglia from the second MRI image according to the geometric features;

[0046] An image segmentation module is used to obtain texture features of the basal ganglia in the first MRI image, and perform a second round of segmentation on the first segmented image according to the texture features to obtain a second segmented image, wherein the texture features are gray-level co-occurrence matrix features;

[0047] A region identification module is used to divide the copper deposition region in the basal ganglia according to the first segmented image and the second segmented image, and after determining the copper deposition volume according to the copper deposition region, calculate the volume ratio of the copper deposition volume to the basal ganglia volume;

[0048] The damage assessment module is used to obtain the patient's physiological characteristic data, input the physiological characteristic data and volume ratio into a pre-trained convolutional neural network model to obtain a quantitative score of cognitive or neural function, and evaluate the degree of damage to the patient affected by Wilson's disease based on the size of the quantitative score of cognitive or neural function; the degree of damage includes mild, moderate and severe.

[0049] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the artificial intelligence-based Wilson disease damage degree assessment method described in any one of the above items is implemented.

[0050] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed, any of the above-mentioned artificial intelligence-based Wilson disease damage degree assessment methods is implemented.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present application discloses an artificial intelligence-based Wilson disease damage assessment system and method, comprising: obtaining geometric features of the basal ganglia from a first MRI image, and preliminarily segmenting a first segmented image of the basal ganglia from a second MRI image according to the geometric features; performing a second round of segmentation on the first segmented image according to texture features to obtain a second segmented image; dividing the copper deposition area in the basal ganglia according to the first segmented image and the second segmented image, and calculating the volume ratio of the copper deposition volume to the basal ganglia volume; inputting the physiological feature data and the volume ratio into a pre-trained convolutional neural network model to obtain a quantitative score, and assessing the degree of damage to the patient affected by Wilson disease according to the size of the quantitative score; based on the above features, The present invention not only helps doctors accurately identify the basal ganglia region through a precise segmentation method, but also provides a reliable basis for the quantification of the copper deposition area. Furthermore, based on the identification and quantification of the copper deposition area, the present invention further analyzes and mines the complex data relationship between copper deposition and cognitive or neural function, thereby providing a more accurate and comprehensive tool for assessing the degree of damage to patients affected by Wilson's disease. Compared with traditional methods, the present invention can more comprehensively reveal the actual impact of copper deposition, help doctors assess the patient's condition from multiple dimensions, and provide a more personalized treatment plan. It enables doctors to make more accurate judgments on the condition in clinical practice, helps doctors achieve earlier and more accurate diagnosis in clinical practice, and adjust treatment plans in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flowchart of a method for assessing the degree of damage of Wilson's disease based on artificial intelligence provided by the present invention;

[0054] Figure 2 A schematic diagram of the module structure of an artificial intelligence-based Wilson disease damage degree assessment system provided by the present invention;

[0055] Figure 3 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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.

[0057] Example 1: Please refer to Figure 1 As shown, this embodiment discloses a method for assessing the degree of damage of Wilson's disease based on artificial intelligence, the method comprising:

[0058] S101: acquiring a first MRI image and a second MRI image of a patient, acquiring geometric features about the basal ganglia from the first MRI image, and preliminarily segmenting a first segmented image about the basal ganglia from the second MRI image according to the geometric features;

[0059] It should be understood that: the first MRI image is a standard reference image of the cerebral hemisphere, which is derived from a standard brain template (such as a standard brain in the MNI space). Unlike the second MRI image, the second MRI image is derived from a nuclear magnetic resonance imaging of the patient's brain area; the MRI image is formed by performing magnetic resonance imaging of the human brain, and a high-resolution T1-weighted or T2-weighted image is usually selected because it can better display the brain structure. The image resolution is usually 1mm×1mm×1mm (or higher) to better display the details of the brain, especially the morphological structure of the basal ganglia;

[0060] Wherein, the geometric features of the basal ganglia are obtained from the first MRI image using a medical image processing tool (such as ITK, SimpleITK, 3D Slicer, MATLAB, etc.);

[0061] In implementation, the step of preliminarily segmenting a first segmented image of the basal ganglia from the second MRI image includes:

[0062] Using a K-means clustering algorithm to distinguish pixels of the second MRI image, and clustering the pixels to form regions as candidate regions, thereby obtaining a plurality of candidate regions;

[0063] Obtaining a first feature parameter set about geometric features of each candidate region, wherein the first feature parameter set includes a plurality of geometric feature vectors, including but not limited to a position feature vector of the basal ganglia relative to the brain region, a shape feature vector of the basal ganglia, an area feature vector of the basal ganglia, and a perimeter feature vector of the basal ganglia;

[0064] Retrieving a second feature parameter set related to geometric features in the first MRI image;

[0065] The second feature parameter set is pre-stored in the system database, and the type and quantity of geometric features in the second feature parameter set are completely consistent with those in the first feature parameter set. For details, refer to the first feature parameter set, and no further details are given here.

[0066] Calculate the similarity index between the first feature parameter and the second feature parameter of each candidate region by using a preconfigured similarity measurement model to obtain multiple similarity indexes;

[0067] ;

[0068] Where: represents the similarity index, represents the kth geometric eigenvector in the first eigenparameter set, represents the kth geometric eigenvector in the second eigenparameter set, represents the number of geometric eigenvectors;

[0069] The multiple similarity indexes are sorted from large to small according to the numerical values, and the candidate area corresponding to the first sorted similarity index is used as the first segmented image.

[0070] S102: Acquire texture features of the basal ganglia in the first MRI image, and perform a second round of segmentation on the first segmented image according to the texture features to obtain a second segmented image, wherein the texture features are gray level co-occurrence matrix features;

[0071] It should be understood that: for the MRI image segmentation of Wilson disease patients, the existing technology usually relies on traditional image segmentation methods, such as threshold method and region growing method; however, although these methods can achieve good segmentation effects in some cases, there are still some key defects and limitations when applied to the MRI image segmentation of Wilson disease. This is because in the brain of Wilson disease patients, especially the basal ganglia, degenerative changes may occur due to the accumulation of copper. These changes lead to unclear or irregular tissue boundaries, especially in the early stages, making it difficult for traditional image segmentation methods to accurately identify them. In addition, as the disease progresses, the morphology of areas such as the basal ganglia may atrophy (the atrophied basal ganglia usually show reduced volume, blurred edges, irregular morphology, etc. on MRI images), which increases the difficulty of segmentation. It is difficult for existing methods to accurately capture these pathological changes;

[0072] Wherein, the texture features of the basal ganglia in the first MRI image are obtained by processing through an existing tool library, and the tool library includes but is not limited to SimpleITK, PyRadiomics, Scikit-image and Numpy+SciPy, etc.;

[0073] In implementation, the second round of segmentation of the first segmented image includes:

[0074] Using an edge detection algorithm to extract boundaries between different levels in the first segmented image, a plurality of boundaries are obtained;

[0075] Wherein, the edge detection algorithm includes but is not limited to Sobel operator and Canny operator, etc.;

[0076] Marking the hierarchical area surrounded by every two boundaries as a candidate area, and obtaining multiple candidate areas;

[0077] Extracting gray-level co-occurrence matrix features of the basal ganglia in the first MRI image and gray-level co-occurrence matrix features of each candidate region;

[0078] Marking the gray level co-occurrence matrix features of the basal ganglia in the first MRI image as first gray level co-occurrence matrix features, and marking the gray level co-occurrence matrix features of the candidate region as second gray level co-occurrence matrix features;

[0079] Calculate the cosine similarity coefficient between the first gray level co-occurrence matrix feature and the second gray level co-occurrence matrix feature to obtain a plurality of cosine similarity coefficients;

[0080] Wherein, the cosine similarity coefficient is calculated by a cosine similarity algorithm;

[0081] The multiple cosine similarity coefficients are sorted from large to small in value, and the candidate area corresponding to the first sorted cosine similarity coefficient is marked as the second segmented image.

[0082] S103: dividing the copper deposition area in the basal ganglia according to the first segmented image and the second segmented image, and after determining the copper deposition volume according to the copper deposition area, calculating the volume ratio of the copper deposition volume to the basal ganglia volume;

[0083] In implementation, the step of dividing the copper deposition area in the basal ganglia includes:

[0084] After rigid registration is performed on the first segmented image and the second segmented image, an AND operation is performed on the first segmented image and the second segmented image to obtain an overlapping area and a non-overlapping area;

[0085] It should be noted that: the rigid registration is implemented by a rigid registration algorithm, including but not limited to one of the methods such as the mutual information method or the least squares method; wherein, the AND operation is a common logical operation in computer image processing, and is particularly important in image segmentation, image fusion and image comparison, and its purpose is to find the area of ​​copper deposition. It is further noted that for two segmented images (binary images (each pixel value is 0 or 1)), if the pixel values ​​at the same position in the two images are both 1, then 1 is output (indicating that it is an overlapping area), and if any of the pixel values ​​is 0, then 0 is output (indicating that this part is not an overlapping area, i.e., a non-overlapping area);

[0086] Mark the non-overlapping areas as copper deposition areas;

[0087] Specifically, determining the copper deposition volume according to the copper deposition area includes:

[0088] Acquiring pixel parameters of the copper deposition area, wherein the pixel parameters include the number of pixels and the pixel volume;

[0089] The copper deposition volume is obtained by calculating the product of the number of pixels and the pixel volume;

[0090] It should be noted that the product formula of the number of pixels and the pixel volume is: ,

[0091] Where: , is the pixel volume, which is used to represent the physical volume of each pixel and is determined by the spatial resolution of the image. , and are the resolutions of the image of the copper deposition area in the X, Y, and Z directions, respectively, and N is the number of pixels in the copper deposition area. is the copper deposition volume;

[0092] It should also be noted that the calculation logic of the basal ganglia volume is consistent with the copper deposition volume mentioned above, except that the basal ganglia volume is calculated based on the first segmented image; wherein the calculation formula of the volume ratio is: ;Where: is the volume ratio, is the basal ganglia volume;

[0093] It should be understood that the volume change of the basal ganglia area is an important indicator for assessing the degree of damage in Wilson disease patients and can be used to reflect pathological changes. Subsequently, by exploring the relationship between volume changes and patients' cognition or neurological disorders, it can be used to assess the degree of damage to patients caused by lesions of different degrees.

[0094] S104: Acquire physiological characteristic data of the patient, input the physiological characteristic data and the volume ratio into a pre-trained convolutional neural network model to obtain a quantitative score of cognitive or neural function, and evaluate the degree of damage of the patient affected by Wilson's disease according to the quantitative score of cognitive or neural function; the degree of damage includes mild, moderate and severe;

[0095] The physiological characteristic data include but are not limited to age, gender, years of education and body mass index (BMI), etc.

[0096] In implementation, the training method of the convolutional neural network model is as follows:

[0097] Collecting historical loss degree training data, and dividing the historical loss degree training data into a loss degree training set and a loss degree test set; the historical loss degree training data includes loss degree reflection feature data and its corresponding quantitative score;

[0098] Wherein, the loss degree reflecting characteristic data includes physiological characteristic data and volume ratio;

[0099] It should be noted that: the historical loss degree training data is collected by technicians based on clinical data, and the quantitative score is obtained by technicians after quantitatively scoring the patients based on the cognitive or neurological function scoring scale;

[0100] Wherein, the cognitive or neurological function rating scale includes but is not limited to the Montreal Cognitive Assessment Scale and the Hamilton Depression Rating Scale;

[0101] It should be understood that: genetic diseases in which copper accumulates in various parts of the body due to abnormal copper metabolism, especially in the brain and liver. Excessive accumulation of copper in the brain, especially the basal ganglia, can cause a series of neurological symptoms, including movement disorders, cognitive decline, psychiatric symptoms, etc.

[0102] Construct a convolutional neural network for performing regression tasks, use the loss degree reflection feature data in the loss degree training set as the input of the convolutional neural network, and use the label of the loss degree reflection feature data as the output of the convolutional neural network, train the convolutional neural network, and obtain an initial convolutional neural network;

[0103] Use the loss degree test set to verify the model of the initial convolutional neural network, and output the initial convolutional neural network with a test accuracy greater than or equal to the preset test accuracy as the trained convolutional neural network model;

[0104] In practice, the step of assessing the extent of impairment of a patient affected by Wilson's disease includes:

[0105] Setting a quantitative scoring threshold, wherein the quantitative scoring threshold includes Td1 and Td2, Td1>Td2>0;

[0106] comparing the quantitative scores of cognitive or neurological function to quantitative score thresholds;

[0107] If the quantitative score of cognitive or neurological function is greater than the quantitative score threshold Td1, the patient is judged to have severe damage due to Wilson's disease;

[0108] If the quantitative score of cognitive or neurological function is less than or equal to the quantitative score threshold Td1, and greater than or equal to the quantitative score threshold Td2, the degree of damage to the patient affected by Wilson's disease is judged to be moderate;

[0109] If the quantitative score of cognitive or neurological function is less than the quantitative score threshold Td2, the patient is judged to have a mild degree of impairment due to Wilson's disease;

[0110] It is understandable that: when the degree of impairment of a patient affected by Wilson's disease is judged to be mild, it usually means that the patient's cognitive and mental state are affected, but the impact is minor, and the patient can still maintain basic daily life; specifically, it is manifested as: mild cognitive problems, such as occasional memory impairment or inattention, but it will not significantly affect work or social interaction; mild speech or expression difficulties, such as inability to find appropriate words or poor expression in certain situations, but these problems will not seriously affect communication; mood swings are small, and occasional anxiety or mild depression may occur, but the patient is usually able to self-regulate and keep daily activities normal; when the degree of impairment of a patient affected by Wilson's disease is judged to be moderate, it usually means that the patient's cognitive and mental functions are more significantly affected and have begun to interfere with all aspects of daily life; specifically, it is manifested as: moderate memory impairment, the patient may have difficulty recalling past events or maintaining long-term memory, but can still maintain a certain degree of independence; executive function decline, including weakened decision-making ability, problem-solving ability and planning ability, the patient may have difficulty making decisions or handling complex tasks, and may need help from others to complete certain daily tasks; speech or communication difficulties worsen, such as the patient may Speech disorders occur more frequently, communication takes more time, and comprehension may be slightly affected; emotional problems are obvious, such as more obvious anxiety, depression or emotional instability, and medication or psychological support may be needed to relieve symptoms; when the degree of damage caused by Wilson disease is judged to be severe, it usually means that the patient's cognition and mental state have been severely damaged, resulting in significant obstacles in the patient's daily life; specifically, it is manifested as: severe cognitive impairment, such as obvious short-term and long-term memory loss, the patient may often forget important information or be unable to remember recent events, seriously affecting his ability to live independently; severe impairment of executive function, the patient's decision-making ability, planning ability, judgment ability and organizational ability are significantly reduced, and may be completely unable to complete daily tasks independently and require round-the-clock care; loss of speech and communication function, the patient may show severe slurred speech or inability to understand what others say, resulting in inability to communicate effectively; significant mental symptoms, such as severe anxiety, depression, hallucinations or delusions, etc., the patient may need long-term psychiatric treatment and drug intervention; all-round functional loss, the patient may be completely unable to take care of himself in basic life, such as dressing, eating, washing, etc., and requires comprehensive care from others;

[0111] The precise segmentation method can not only help doctors accurately identify the basal ganglia area, but also provide a reliable basis for the quantification of copper deposition areas. Furthermore, based on the identification and quantification of copper deposition areas, the complex data relationship between copper deposition and cognitive or neural function is further analyzed and mined, providing a more accurate and comprehensive tool for assessing the degree of damage to patients affected by Wilson's disease. Compared with traditional methods, the present invention can more comprehensively reveal the actual impact of copper deposition, help doctors evaluate patients' conditions from multiple dimensions, and provide more personalized treatment plans. It enables doctors to make more accurate judgments on the condition in clinical practice, helps doctors achieve earlier and more accurate diagnosis in clinical practice, and adjust treatment plans in a timely manner, significantly improving the treatment effect of Wilson's disease patients.

[0112] Example 2: Please refer to Figure 2 As shown, based on the same inventive concept, this embodiment discloses a Wilson disease damage degree assessment system based on artificial intelligence. For details not provided in this embodiment, please refer to the description of the relevant parts in Example 1. The system includes:

[0113] The data acquisition module 210 is used for the first MRI image and the second MRI image of the patient, acquiring geometric features about the basal ganglia from the first MRI image, and preliminarily segmenting a first segmented image about the basal ganglia from the second MRI image according to the geometric features;

[0114] An image segmentation module 220 is used to obtain texture features of the basal ganglia in the first MRI image, and perform a second round of segmentation on the first segmented image according to the texture features to obtain a second segmented image, wherein the texture features are gray level co-occurrence matrix features;

[0115] A region identification module 230 is used to divide the copper deposition region in the basal ganglia according to the first segmented image and the second segmented image, and after determining the copper deposition volume according to the copper deposition region, calculate the volume ratio of the copper deposition volume to the basal ganglia volume;

[0116] The damage assessment module 240 is used to obtain the patient's physiological characteristic data, input the physiological characteristic data and volume ratio into a pre-trained convolutional neural network model to obtain a quantitative score of cognitive or neural function, and assess the degree of damage to the patient affected by Wilson's disease based on the size of the quantitative score of cognitive or neural function; the degree of damage includes mild, moderate and severe.

[0117] Example 3: Please refer to Figure 3As shown, this embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the artificial intelligence-based Wilson disease damage degree assessment method provided by any one of the above methods.

[0118] Since the electronic device introduced in the content of this embodiment is an electronic device used to implement a method for assessing the degree of damage of Wilson's disease based on artificial intelligence in the embodiment of this application, based on the method for assessing the degree of damage of Wilson's disease based on artificial intelligence introduced in the embodiment of this application, a person skilled in the art can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not described in detail here. As long as a person skilled in the art implements an electronic device used in a method for assessing the degree of damage of Wilson's disease based on artificial intelligence in the embodiment of this application, it belongs to the scope of protection of this application.

[0119] Embodiment 3: This embodiment discloses a computer-readable storage medium, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements any one of the artificial intelligence-based Wilson disease damage degree assessment methods provided by the above methods.

[0120] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.

[0121] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired network or a wireless network. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD) or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0122] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0123] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0124] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units 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 units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0126] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0127] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0128] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for assessing the degree of damage of Wilson's disease based on artificial intelligence, characterized in that: The method comprises: Acquire a first MRI image and a second MRI image of the patient, acquire geometric features about the basal ganglia from the first MRI image, and preliminarily segment a first segmented image about the basal ganglia from the second MRI image according to the geometric features; Acquire texture features of the basal ganglia in the first MRI image, and perform a second round of segmentation on the first segmented image according to the texture features to obtain a second segmented image, wherein the texture features are gray-level co-occurrence matrix features; The copper deposition area in the basal ganglia is divided according to the first segmented image and the second segmented image, and after the copper deposition volume is determined according to the copper deposition area, the volume ratio of the copper deposition volume to the basal ganglia volume is calculated; The patient's physiological characteristic data is obtained, and the physiological characteristic data and volume ratio are input into a pre-trained convolutional neural network model to obtain a quantitative score of cognitive or neurological function, and the degree of damage to the patient affected by Wilson's disease is evaluated according to the size of the quantitative score of cognitive or neurological function; the degree of damage includes mild, moderate and severe.

2. The artificial intelligence-based Wilson disease damage assessment method according to claim 1, characterized in that: The step of preliminarily segmenting a first segmented image of the basal ganglia from the second MRI image comprises: Using a K-means clustering algorithm to distinguish pixels of the second MRI image, and clustering the pixels to form regions as candidate regions, thereby obtaining a plurality of candidate regions; Obtaining a first feature parameter set about geometric features of each candidate region, wherein the first feature parameter set includes a plurality of geometric feature vectors, including but not limited to a position feature vector of the basal ganglia relative to the brain region, a shape feature vector of the basal ganglia, an area feature vector of the basal ganglia, and a perimeter feature vector of the basal ganglia; Retrieving a second feature parameter set related to geometric features in the first MRI image; Calculate the similarity index between the first feature parameter and the second feature parameter of each candidate region by using a preconfigured similarity measurement model to obtain multiple similarity indexes; ； Where: represents the similarity index, represents the kth geometric eigenvector in the first eigenparameter set, represents the kth geometric eigenvector in the second eigenparameter set, represents the number of geometric eigenvectors; The multiple similarity indexes are sorted from large to small according to the numerical values, and the candidate area corresponding to the first sorted similarity index is used as the first segmented image.

3. The artificial intelligence-based Wilson disease damage assessment method according to claim 2, characterized in that: The second round of segmentation of the first segmented image comprises: Using an edge detection algorithm to extract boundaries between different levels in the first segmented image, a plurality of boundaries are obtained; Marking the hierarchical area surrounded by every two boundaries as a candidate area, and obtaining multiple candidate areas; Extracting gray-level co-occurrence matrix features of the basal ganglia in the first MRI image and gray-level co-occurrence matrix features of each candidate region; Marking the gray level co-occurrence matrix features of the basal ganglia in the first MRI image as first gray level co-occurrence matrix features, and marking the gray level co-occurrence matrix features of the candidate region as second gray level co-occurrence matrix features; Calculate the cosine similarity coefficient between the first gray level co-occurrence matrix feature and the second gray level co-occurrence matrix feature to obtain a plurality of cosine similarity coefficients; The multiple cosine similarity coefficients are sorted from large to small in value, and the candidate area corresponding to the first sorted cosine similarity coefficient is marked as the second segmented image.

4. The artificial intelligence-based Wilson disease damage assessment method according to claim 3, characterized in that: The step of dividing the copper deposition area in the basal ganglia comprises: After rigid registration is performed on the first segmented image and the second segmented image, an AND operation is performed on the first segmented image and the second segmented image to obtain an overlapping area and a non-overlapping area; The non-overlapping areas are marked as copper deposition areas.

5. The artificial intelligence-based Wilson disease damage assessment method according to claim 4, characterized in that: Determining the copper deposition volume according to the copper deposition area includes: Acquiring pixel parameters of the copper deposition area, wherein the pixel parameters include the number of pixels and the pixel volume; The copper deposition volume is obtained by calculating the product of the number of pixels and the pixel volume.

6. The artificial intelligence-based Wilson disease damage assessment method according to claim 5, characterized in that: The physiological characteristic data include but are not limited to age, gender, years of education and body mass index (BMI); Among them, the training method of the convolutional neural network model is as follows: Collecting historical loss degree training data, and dividing the historical loss degree training data into a loss degree training set and a loss degree test set; the historical loss degree training data includes loss degree reflection feature data and its corresponding quantitative score; Wherein, the loss degree reflecting characteristic data includes physiological characteristic data and volume ratio; Construct a convolutional neural network for performing regression tasks, use the loss degree reflection feature data in the loss degree training set as the input of the convolutional neural network, and use the label of the loss degree reflection feature data as the output of the convolutional neural network, train the convolutional neural network, and obtain an initial convolutional neural network; The initial convolutional neural network model is verified using the loss degree test set, and an initial convolutional neural network with a preset test accuracy is output as the trained convolutional neural network model.

7. The artificial intelligence-based Wilson disease damage assessment method according to claim 6, characterized in that: The patient is assessed for the extent of impairment caused by Wilson disease, including: Setting a quantitative scoring threshold, wherein the quantitative scoring threshold includes Td1 and Td2, Td1>Td2>0; comparing the quantitative scores of cognitive or neurological function to quantitative score thresholds; If the quantitative score of cognitive or neurological function is greater than the quantitative score threshold Td1, the patient is judged to have severe damage due to Wilson's disease; If the quantitative score of cognitive or neurological function is less than or equal to the quantitative score threshold Td1, and greater than or equal to the quantitative score threshold Td2, the degree of damage to the patient affected by Wilson's disease is judged to be moderate; If the quantitative score of cognitive or neurological function is less than the quantitative score threshold Td2, the degree of impairment of the patient affected by Wilson's disease is judged to be mild.

8. An artificial intelligence-based Wilson disease damage assessment system, characterized in that: The method is implemented based on the artificial intelligence-based Wilson disease damage assessment method according to any one of claims 1 to 7, and the system comprises: A data acquisition module, for the first MRI image and the second MRI image of the patient, acquires geometric features about the basal ganglia from the first MRI image, and preliminarily segments a first segmented image about the basal ganglia from the second MRI image according to the geometric features; An image segmentation module is used to obtain texture features of the basal ganglia in the first MRI image, and perform a second round of segmentation on the first segmented image according to the texture features to obtain a second segmented image, wherein the texture features are gray-level co-occurrence matrix features; A region identification module is used to divide the copper deposition region in the basal ganglia according to the first segmented image and the second segmented image, and after determining the copper deposition volume according to the copper deposition region, calculate the volume ratio of the copper deposition volume to the basal ganglia volume; The damage assessment module is used to obtain the patient's physiological characteristic data, input the physiological characteristic data and volume ratio into a pre-trained convolutional neural network model to obtain a quantitative score of cognitive or neural function, and evaluate the degree of damage to the patient affected by Wilson's disease based on the size of the quantitative score of cognitive or neural function; the degree of damage includes mild, moderate and severe.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the artificial intelligence-based Wilson disease damage degree assessment method described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the artificial intelligence-based Wilson disease damage degree assessment method according to any one of claims 1 to 7 is implemented.

Images