Diagnosis management system based on brain CT image analysis

Abstract

The invention belongs to the technical field of brain CT image diagnosis, and discloses a diagnosis management system based on brain CT image analysis. Comprising an acquisition module used for acquiring basic information, clinical information and brain CT images of a patient; the diagnosis module is used for comparing the brain CT image of the patient with a pre-constructed normal brain model to obtain abnormal brain information, and inputting the abnormal brain information into a pre-constructed brain disease diagnosis model to obtain a diagnosis result; according to the invention, tiny structural abnormality, vasculopathy and brain sulcus change in a brain CT image can be captured, and the accuracy and efficiency of brain disease diagnosis are improved; the diagnosis result of the patient can be matched to generate a disease development path, so that a doctor and the patient are helped to know the clinical stage of an illness state, and the disease development speed of the patient can be adjusted by setting the gain factor and the inhibition factor.

Description

Technical Field

[0001] The present invention relates to the technical field of brain CT image diagnosis, and more specifically, to a diagnosis management system based on brain CT image analysis. Background Art

[0002] With the rapid development of medical imaging technology, brain CT images have become an indispensable and important means of diagnosing brain diseases. Brain CT images play a vital role in modern medical diagnosis. Their high resolution and detailed anatomical structure images enable doctors to accurately evaluate brain anatomy, lesions and functional status. With the development of computer vision and image processing technology, automated diagnostic systems based on computer vision and image processing technology have gradually emerged, providing new possibilities for the accurate diagnosis of brain diseases.

[0003] However, traditional brain CT image analysis methods are easily affected by doctors' personal experience and subjective judgment, which greatly restricts the efficiency and accuracy of brain disease diagnosis. This is especially challenging when identifying tiny structural abnormalities, vascular lesions, and subtle changes in brain grooves, which may increase the risk of misdiagnosis or missed diagnosis.

[0004] In view of this, the present invention proposes a diagnosis management system based on brain CT image analysis to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: a diagnosis and management system based on brain CT image analysis, comprising an acquisition module for acquiring basic information, clinical information and brain CT images of patients;

[0006] A diagnosis module is used to compare the patient's brain CT image with a pre-built normal brain model to obtain abnormal brain information, and input the abnormal brain information into a pre-built brain disease diagnosis model to obtain a diagnosis result;

[0007] A generation module is used to build a diagnosis database and generate a disease development path based on the matching of the diagnosis results with the diagnosis database;

[0008] The comparison module is used to collect the patient's clinical information in real time, obtain real-time clinical information, and compare the real-time clinical information with the disease development path to obtain the comparison result;

[0009] The update module is used to preset the deviation sequence, compare the comparison result with the deviation sequence, and determine whether to update the disease development path.

[0010] Furthermore, in the acquisition module,

[0011] Obtaining the patient's basic information through electronic medical records, including age, height, weight, and family genetic disease history;

[0012] Obtaining clinical information of the patient through medical examination equipment, electronic medical records, historical patient medical records and patient complaints, wherein the data corresponding to the clinical information includes body temperature, blood pressure, blood lipids, cholesterol, heart rate, past medical history and patient's head symptoms;

[0013] Obtain the patient's brain CT image, and associate the patient's basic information and clinical information with the brain CT image.

[0014] Furthermore, the steps of constructing the normal brain model are:

[0015] Preset age stages, and divide normal brain CT images of different ages according to age stages;

[0016] Extract and associate features of n normal brain CT images within the same age stage, and obtain the normal brain model corresponding to the age stage;

[0017] The method of extracting and associating features of n normal brain CT images in the same age group is as follows:

[0018] The reference three-dimensional coordinates are established based on the midpoint of the AC-PC line, and the position distribution range of each part in n normal brain CT images on the reference three-dimensional coordinates is obtained to obtain the part distribution characteristics;

[0019] Obtain the volume distribution range of each part in n normal brain CT images to obtain volume features;

[0020] Obtain the grayscale range of each part in n normal brain CT images to obtain grayscale features;

[0021] Obtain the length, diameter or depth and extension direction of veins, arteries and sulci in n normal brain CT images and take the average as the line to obtain the three-dimensional structure line;

[0022] The location distribution characteristics, volume characteristics, grayscale characteristics, and three-dimensional structure lines of normal brain CT images corresponding to the same age stage are correlated to obtain a normal brain model corresponding to the age stage.

[0023] Furthermore, in the three-dimensional structure line,

[0024] Perform line division at the inflection points or intersections of each line in the three-dimensional structure line to obtain m divided line segments;

[0025] Each segment is associated with its corresponding length range or depth range, extension direction range, and grayscale mean variance characteristics;

[0026] The grayscale mean variance feature is obtained as follows:

[0027] Preset the unit line segment expansion area, obtain the maximum average grayscale threshold and the maximum grayscale variance threshold of the unit line segment expansion area, associate the maximum average grayscale threshold and the maximum grayscale variance threshold with the unit line segment expansion area of ​​the corresponding line segment, and obtain the line grayscale mean variance feature;

[0028] The maximum average grayscale threshold is the maximum grayscale value among the n average grayscale values ​​corresponding to the same dividing line segment in n normal brain CT images;

[0029] The maximum grayscale variance threshold is the maximum grayscale variance threshold among the n grayscale variances corresponding to the same dividing line segment in n normal brain CT images.

[0030] Furthermore, the step of obtaining abnormal brain information includes:

[0031] Select a normal brain model corresponding to the patient based on the patient's basic information;

[0032] Compare the patient's brain CT images with the normal brain model in terms of location distribution, volume, grayscale, and three-dimensional structure lines;

[0033] The brain areas that are not within the location distribution characteristics, volume characteristics, grayscale characteristics, three-dimensional structure lines, and line grayscale mean variance characteristics of the normal brain model are marked to obtain abnormal brain information.

[0034] Furthermore, the step of inputting abnormal brain information into a pre-constructed brain disease diagnosis model to obtain a diagnosis result includes:

[0035] Build and train models for diagnosing brain diseases;

[0036] Input abnormal brain information into the trained brain disease diagnosis model;

[0037] The brain disease diagnosis model outputs the diagnosis results corresponding to the abnormal brain information.

[0038] Furthermore, the step of constructing a diagnostic database and generating a disease development path based on matching the diagnostic result with the diagnostic database includes:

[0039] Obtain clinical information and brain CT images corresponding to different types of brain diseases, associate the same type of brain disease with the corresponding clinical information and brain CT images, and then obtain a diagnostic database;

[0040] The diagnosis result is input into the diagnosis database, and a conventional diagnosis path corresponding to the diagnosis result in the diagnosis database is obtained based on a similarity algorithm;

[0041] Obtain the clinical stage corresponding to the diagnosis result, obtain the path nodes that the diagnosis result has passed based on the clinical stage, and obtain the patient's disease progression speed through the passed path nodes;

[0042] Collect and determine the patient's gain factor and inhibition factor, as well as the gain speed and inhibition speed corresponding to the gain factor and inhibition factor, respectively;

[0043] The time between adjacent path nodes is adjusted based on the disease development speed, gain speed and inhibition speed, thereby obtaining the disease development path;

[0044] The adjustment methods are:

[0045] The actual disease development speed is obtained by adding the disease development speed to the gain speed and subtracting the inhibition speed. If the actual disease development speed is greater than zero, other inhibition factors are added until the actual disease development speed is equal to zero or less than zero.

[0046] Furthermore, the conventional diagnostic path is obtained in the following manner:

[0047] Acquire clinical information and brain CT images of the same brain disease at different times, and construct a conventional development path according to the chronological order of the clinical information, wherein a path node is added at a position corresponding to the time when a clinical stage is reached on the conventional development path, and the path node includes clinical information corresponding to the clinical stage of the brain disease;

[0048] Obtaining the gain factor and inhibition factor of the brain disease in the conventional development path, and obtaining the gain speed or inhibition speed corresponding to each gain factor or each inhibition factor between adjacent path nodes;

[0049] The gain factor is a factor that increases the development speed of the brain disease at adjacent path nodes; the inhibition factor is a factor that reduces the development speed of the brain disease at adjacent path nodes.

[0050] Furthermore, the step of collecting the patient's clinical information in real time to obtain the real-time clinical information, and comparing the real-time clinical information with the disease development path to obtain the comparison result includes:

[0051] Preset the collection cycle and obtain the patient's real-time clinical information through the collection cycle;

[0052] Acquire the path node corresponding to the patient's real-time clinical information to obtain the real-time path node;

[0053] The real-time clinical information of the real-time pathway node is compared with the clinical information corresponding to the pathway node on the disease development pathway to obtain a comparison result.

[0054] Furthermore, the step of presetting the deviation sequence, comparing the comparison result with the deviation sequence, and determining whether to update the disease development pathway comprises:

[0055] A deviation interval in the deviation sequence is preset, and the data in the comparison result is compared with the corresponding deviation interval; if the data in the comparison result is within the deviation interval in the deviation sequence, the path nodes in the disease development path are replaced by real-time path nodes to update the path nodes in the disease development path; if the comparison result is not within the deviation interval in the deviation sequence, the path nodes in the disease development path continue to be maintained.

[0056] The technical effects and advantages of the diagnostic management system based on brain CT image analysis of the present invention are as follows:

[0057] 1. By constructing a detailed normal brain model and performing a multi-dimensional comparison with the patient's brain CT images, abnormal brain information can be accurately identified, and then accurate diagnostic results can be output through the brain disease diagnosis model, which helps to capture tiny structural abnormalities, vascular lesions, and changes in brain sulci in brain CT images. This not only improves the accuracy and efficiency of brain disease diagnosis, but also helps to achieve precision medicine;

[0058] 2. By building a diagnostic database, it helps to match the patient's diagnostic results with the disease development path, which can help doctors and patients know the clinical stage of the disease. By setting gain factors and inhibitory factors, it helps to adjust the patient's disease development speed; through the comparison of real-time clinical information, it can trigger the update mechanism of the path nodes in the disease development path according to whether the real-time clinical information deviates from the expected disease development path, so as to timely detect changes in the patient's condition and ensure that doctors can make more accurate diagnosis and treatment decisions based on the latest clinical data. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a structural schematic diagram of a diagnosis and management system based on brain CT image analysis of the present invention;

[0060] Figure 2 A schematic diagram of the process of generating a disease development pathway in the present invention. DETAILED DESCRIPTION

[0061] 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.

[0062] Example 1

[0063] See also Figure 1 As shown, this embodiment is a diagnosis and management system based on brain CT image analysis, including:

[0064] An acquisition module is used to acquire the patient's basic information, clinical information and brain CT images (brain CT images are three-dimensional images);

[0065] The diagnosis module is used to compare the patient's brain CT image with a pre-built normal brain model to obtain abnormal brain information, and input the abnormal brain information into a pre-built brain disease diagnosis model to obtain a diagnosis result.

[0066] In the acquisition module,

[0067] Obtain the patient's basic information through electronic medical records, including age, height, weight, and family genetic disease history;

[0068] Obtain the patient's clinical information through medical examination equipment, electronic medical records, historical patient medical records and patient complaints. The data corresponding to the clinical information include body temperature, blood pressure, blood lipids, cholesterol, heart rate, past medical history and the patient's head symptoms (for example, headache, dizziness, confusion, etc., the head symptoms are quantified, and the quantification method can be presence or absence, recorded as 1 and 0 respectively, or the head symptoms can be graded according to preset degree levels, and the degree level is used to achieve standardization of clinical information);

[0069] Obtain the patient's brain CT image, and associate the patient's basic information and clinical information with the brain CT image.

[0070] The steps to construct a normal brain model are:

[0071] Preset age stages (e.g., childhood, youth, adulthood, and old age) and divide normal brain CT images of different ages according to age stages;

[0072] Extract and associate features of n normal brain CT images within the same age stage, and obtain the normal brain model corresponding to the age stage;

[0073] It should be noted that by dividing and building a normal brain model by age, the system can support personalized diagnosis. The brain structure and function of different age groups are different, so using a normal brain model that matches the patient's age for comparison can improve the accuracy and pertinence of the diagnosis;

[0074] Among them, the method of extracting and associating features of n normal brain CT images in the same age stage is as follows:

[0075] The reference three-dimensional coordinates are established based on the midpoint of the AC-PC line (intercommissural line), and the midpoint is used as the origin of the reference three-dimensional coordinates. The position distribution range of each part in n normal brain CT images on the reference three-dimensional coordinates is obtained (the position distribution range of each part is the superposition range of the range of this part in n normal brain CT images), and the part distribution characteristics are obtained;

[0076] Obtain the volume distribution range of each part in n normal brain CT images (composed of the maximum volume and minimum volume of each part) to obtain the volume feature;

[0077] Obtain the grayscale range of each part in n normal brain CT images (composed of the minimum grayscale value and the maximum grayscale value of each part) to obtain the grayscale feature;

[0078] Obtain the length, diameter or depth and extension direction of veins, arteries and cerebral sulci in n normal brain CT images and take the average as the line to obtain the three-dimensional structure line, in which the lines are all pipeline-shaped and the cerebral sulci are incompletely closed pipeline-shaped;

[0079] The location distribution characteristics, volume characteristics, grayscale characteristics, and three-dimensional structure lines of normal brain CT images corresponding to the same age stage are correlated to obtain a normal brain model corresponding to the age stage.

[0080] In the three-dimensional structure line,

[0081] Perform line division at the inflection points or intersections of each line in the three-dimensional structure line to obtain m divided line segments;

[0082] Each segment is associated with its corresponding length range (consisting of the minimum length range and the maximum length range), diameter range (consisting of the minimum diameter and the maximum diameter) or depth range (consisting of the lowest sulcus depth and the deepest sulcus depth), extension direction range (the superposition angle range of n extension directions) and grayscale mean variance features;

[0083] The grayscale mean variance feature is obtained as follows:

[0084] Preset a unit segment expansion area (i.e., the area obtained by opening and expanding the outer wall midline of a segment of unit length, generally in a rectangular shape), obtain the maximum average grayscale threshold and the maximum grayscale variance threshold of the unit segment expansion area, associate the maximum average grayscale threshold and the maximum grayscale variance threshold with the unit segment expansion area of ​​the corresponding segment, and obtain the line grayscale mean variance feature;

[0085] The maximum average grayscale threshold is the maximum grayscale value among the n average grayscale values ​​corresponding to the same segment in n normal brain CT images; the average grayscale value is obtained by adding the grayscale values ​​of all pixels in the unit segment expansion area and then dividing it by the total number of pixels to obtain the average grayscale value of the unit segment expansion area;

[0086] The maximum grayscale variance threshold is the maximum grayscale variance threshold among the n grayscale variances corresponding to the same dividing line segment in n normal brain CT images; the grayscale variance is obtained by accumulating the square of the difference between the grayscale value of each pixel point in the expanded area of ​​the unit line segment and the average grayscale value, and then dividing it by the total number of pixels.

[0087] Steps to obtain abnormal brain information include:

[0088] Select a normal brain model corresponding to the patient based on the patient's basic information;

[0089] Compare the patient's brain CT images with the normal brain model in terms of location distribution, volume, grayscale, and three-dimensional structure lines;

[0090] The brain areas that are not within the location distribution characteristics, volume characteristics, grayscale characteristics, three-dimensional structure lines, and line grayscale mean variance characteristics of the normal brain model are marked to obtain abnormal brain information. If no abnormal brain information is obtained, the patient's brain is considered to be in a normal state.

[0091] The step of inputting abnormal brain information into a pre-built brain disease diagnosis model to obtain a diagnosis result includes:

[0092] Build and train models for diagnosing brain diseases;

[0093] Input abnormal brain information into the trained brain disease diagnosis model;

[0094] The brain disease diagnosis model outputs the diagnosis results corresponding to the abnormal brain information;

[0095] The process of building and training the brain disease diagnosis model is as follows:

[0096] The brain CT images with labeled brain disease types and their corresponding abnormal brain information are used as abnormal brain feature data. The abnormal brain feature data of several different disease types constitute the abnormal brain feature data set. The abnormal brain feature data set is divided into a training set, a validation set, and a test set. The ratios of the training set, the validation set, and the test set are 70%, 15%, and 15% respectively.

[0097] A fully connected neural network is used as the basic framework of the brain disease diagnosis model. The basic framework includes an input layer, a hidden layer, and an output layer. The input layer neurons are defined as abnormal brain feature data. There are two hidden layers, each with 64 and 32 neurons respectively. The hidden layer uses ReLU as the activation function. The output layer is defined as M neurons, which corresponds to the number of brain disease types. The output layer uses softmax as the activation function.

[0098] Define the loss function for the brain disease diagnosis model:

[0099]

[0100] Where N is the number of abnormal brain feature data, M is the number of brain disease types, and yij is an indication of whether the i-th abnormal brain feature data belongs to the j-th brain disease type (1 if it does, and 0 if it does not). Predict the probability that the i-th abnormal brain feature data belongs to the j-th type of brain disease for the brain disease diagnosis model;

[0101] For example, during the training process, the learning rate is set to 0.001, the batch size is 32, and the number of iterations is 100. The performance of the model is evaluated on the validation set after each round of iteration. If the performance does not improve after multiple consecutive rounds of iterations, the training is stopped early. After the training is completed, the performance of the diagnostic model is evaluated on the test set by calculating its accuracy.

[0102] In this embodiment, by establishing the reference three-dimensional coordinates based on the midpoint of the AC-PC line, the distribution characteristics, volume characteristics and grayscale characteristics of each part of the normal brain in the three-dimensional space can be accurately quantified, thereby constructing a highly refined normal brain model, which significantly enhances the ability to identify abnormal brain information during the comparison process; by extracting three-dimensional structural lines and dividing lines, length, depth, extension direction and grayscale mean variance characteristics, it is helpful to capture tiny structural abnormalities, vascular lesions and changes in brain grooves in brain CT images, so that the system can accurately locate and identify these brain abnormalities, which is beneficial to the diagnosis of brain diseases such as vascular blockage and atherosclerosis; by using the constructed normal brain model as a comparison template, it is possible to achieve rapid and automated comparison of the patient's brain CT images, improve diagnostic efficiency, reduce errors caused by manual intervention and subjective judgment of doctors, and thus enhance the accuracy of identifying abnormal brain information in brain CT images.

[0103] Example 2

[0104] See also Figure 1 and Figure 2 As shown, this embodiment is a diagnosis and management system based on brain CT image analysis, including:

[0105] A generation module is used to build a diagnosis database and generate a disease development path based on the matching of the diagnosis results with the diagnosis database;

[0106] The comparison module is used to collect the patient's clinical information in real time, obtain real-time clinical information, and compare the real-time clinical information with the disease development path to obtain the comparison result;

[0107] The update module is used to preset the deviation sequence, compare the comparison result with the deviation sequence, and determine whether to update the disease development path.

[0108] The steps of constructing a diagnostic database and generating a disease development path based on matching the diagnostic results with the diagnostic database include:

[0109] Obtain clinical information and brain CT images corresponding to different types of brain diseases, associate the same type of brain disease with the corresponding clinical information and brain CT images, and then obtain a diagnostic database;

[0110] The diagnosis result is input into the diagnosis database, and the conventional diagnosis path (most similar) corresponding to the diagnosis result in the diagnosis database is obtained based on the similarity algorithm (Euclidean distance algorithm or cosine similarity algorithm);

[0111] Obtain the clinical stage corresponding to the diagnosis result, obtain the path nodes that the diagnosis result has passed based on the clinical stage, and obtain the patient's disease progression speed through the passed path nodes;

[0112] Collect and determine the patient's gain factor and inhibition factor, as well as the gain speed and inhibition speed corresponding to the gain factor and inhibition factor, respectively;

[0113] The time between adjacent path nodes is adjusted based on the disease development speed, gain speed and inhibition speed, thereby obtaining the disease development path;

[0114] The adjustment methods are:

[0115] The actual disease development speed is obtained by adding the disease development speed to the gain speed and subtracting the inhibition speed. If the actual disease development speed is greater than zero, other inhibition factors are added until the actual disease development speed is equal to zero or less than zero.

[0116] The conventional diagnostic path is obtained as follows:

[0117] Obtain clinical information and brain CT images of the same brain disease at different times, and construct a conventional development path according to the chronological order of the clinical information, wherein a path node is added at the position corresponding to the time of reaching a clinical stage on the conventional development path, and the clinical stage is set by relevant personnel. For example, a clinical stage can be an increase or decrease of e% in each piece of information in the patient's clinical information, and e can be a positive or negative value; each path node corresponds to the clinical information and brain CT images corresponding to the brain disease at the clinical stage;

[0118] Obtaining the gain factor and inhibition factor of the brain disease in the conventional development path, obtaining the gain speed or inhibition speed corresponding to each gain factor or each inhibition factor between adjacent path nodes, the gain speed is the speed at which the disease development speed between adjacent path nodes without the gain factor increases, and the inhibition speed is the speed at which the disease development speed between adjacent path nodes without the inhibition factor decreases, and adding a gain path or inhibition path between two adjacent path nodes based on the gain speed and inhibition speed, respectively, wherein each gain factor and inhibition factor adds a gain path or inhibition path between two adjacent path nodes;

[0119] Gain factors are factors that increase the speed of brain disease development at adjacent pathway nodes, such as other diseases that have comorbidities or complications combined with the current brain disease, the combination of chronic diseases such as hypertension and diabetes with atherosclerosis, and the excessive activation of microglia in Alzheimer's disease (AD); inhibitory factors are factors that reduce the speed of brain disease development at adjacent pathway nodes, such as antioxidants, anti-inflammatory drugs, and neuroprotectants. In patients with Alzheimer's disease (AD), the use of antioxidants may help slow down the degeneration of neurons and the decline in cognitive function.

[0120] The steps of collecting the patient's clinical information in real time, obtaining the real-time clinical information, and comparing the real-time clinical information with the disease development path to obtain the comparison result include:

[0121] Preset the collection cycle and obtain the patient's real-time clinical information through the collection cycle;

[0122] Obtain the path node corresponding to the patient's real-time clinical information (match the real-time clinical information with each path node in the disease development path through a similarity algorithm to determine the most similar path node as the real-time path node) to obtain the real-time path node;

[0123] The real-time clinical information of the real-time pathway node is compared with the clinical information corresponding to the pathway node on the disease development pathway to obtain a comparison result.

[0124] The steps of presetting a deviation sequence, comparing the comparison result with the deviation sequence, and determining whether to update the disease development pathway include:

[0125] Preset the deviation interval in the deviation sequence (relevant personnel or doctors set corresponding deviation intervals for the data in the clinical information according to the actual situation. For example, the normal range of body temperature is 36.5 degrees to 37.3 degrees, and the preset deviation interval is 37.4 degrees to 40 degrees, that is, the deviation interval is [37.4-40], that is, when the body temperature exceeds this range, it is considered as a deviation. The deviation sequence consists of multiple deviation intervals), and compare the comparison results (data in the clinical information) with the deviation intervals in the deviation sequence (the deviation range corresponding to the data in the clinical information) one by one;

[0126] If the data in the comparison result is within the deviation interval in the deviation sequence, the pathway nodes in the disease development pathway are replaced by the real-time pathway nodes, thereby updating the pathway nodes in the disease development pathway;

[0127] If the comparison result is not within the deviation interval in the deviation sequence, the path node in the disease development path continues to be maintained.

[0128] In this embodiment, by constructing a diagnostic database containing clinical information and brain CT images, the most similar conventional diagnostic path can be matched for each diagnostic result based on a similarity algorithm, and then a disease development path corresponding to the patient's diagnostic result is generated, which helps doctors or patients understand the clinical stage corresponding to the patient's current brain disease, so as to facilitate the corresponding treatment for the patient; by collecting the patient's clinical information in real time and dynamically comparing it with the disease development path, the changing trend of the patient's condition can be discovered in time; by introducing gain factors and inhibition factors, it is convenient to help doctors or patients understand the speed of disease development more intuitively and provide corresponding adjustment methods; by automatically judging whether the real-time clinical information deviates from the expected disease development path through a preset deviation sequence, and then triggering the update mechanism of the path node in the disease development path, the patient's condition changes can be discovered in time, ensuring that the doctor can make more accurate diagnosis and treatment decisions based on the latest clinical data, and the patient's disease progression can be understood more accurately. Once it is found that the patient's clinical information deviates from the path node corresponding to the expected disease development path, measures are taken immediately to intervene, thereby effectively avoiding the patient's condition deterioration or complications.

[0129] 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.

[0130] 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.

[0131] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0132] 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 diagnostic management system based on brain CT image analysis, characterized in that: include: The acquisition module is used to obtain the patient's basic information, clinical information and brain CT images; A diagnosis module is used to compare the patient's brain CT image with a pre-built normal brain model to obtain abnormal brain information, and input the abnormal brain information into a pre-built brain disease diagnosis model to obtain a diagnosis result; A generation module is used to build a diagnosis database and generate a disease development path based on the matching of the diagnosis results with the diagnosis database; The comparison module is used to collect the patient's clinical information in real time, obtain real-time clinical information, and compare the real-time clinical information with the disease development path to obtain the comparison result; The update module is used to preset the deviation sequence, compare the comparison result with the deviation sequence, and determine whether to update the disease development path.

2. The diagnostic management system based on brain CT image analysis according to claim 1, characterized in that: In the acquisition module, Obtaining the patient's basic information through electronic medical records, including age, height, weight, and family genetic disease history; Obtaining clinical information of the patient through medical examination equipment, electronic medical records, historical patient medical records and patient complaints, wherein the data corresponding to the clinical information includes body temperature, blood pressure, blood lipids, cholesterol, heart rate, past medical history and patient's head symptoms; Obtain the patient's brain CT image, and associate the patient's basic information and clinical information with the brain CT image.

3. The diagnostic management system based on brain CT image analysis according to claim 1, characterized in that: The steps of constructing the normal brain model are: Preset age stages, and divide normal brain CT images of different ages according to age stages; Extract and associate features of n normal brain CT images within the same age stage, and obtain the normal brain model corresponding to the age stage; The method of extracting and associating features of n normal brain CT images in the same age group is as follows: The reference three-dimensional coordinates are established based on the midpoint of the AC-PC line, and the position distribution range of each part in n normal brain CT images on the reference three-dimensional coordinates is obtained to obtain the part distribution characteristics; Obtain the volume distribution range of each part in n normal brain CT images to obtain volume features; Obtain the grayscale range of each part in n normal brain CT images to obtain grayscale features; Obtain the length, diameter or depth and extension direction of veins, arteries and sulci in n normal brain CT images and take the average as the line to obtain the three-dimensional structure line; The location distribution characteristics, volume characteristics, grayscale characteristics, and three-dimensional structure lines of normal brain CT images corresponding to the same age stage are correlated to obtain a normal brain model corresponding to the age stage.

4. The diagnostic management system based on brain CT image analysis according to claim 3 is characterized in that: In the three-dimensional structure line, Perform line division at the inflection points or intersections of each line in the three-dimensional structure line to obtain m divided line segments; Each segment is associated with its corresponding length range or depth range, extension direction range, and grayscale mean variance characteristics; The grayscale mean variance feature is obtained as follows: Preset the unit line segment expansion area, obtain the maximum average grayscale threshold and the maximum grayscale variance threshold of the unit line segment expansion area, associate the maximum average grayscale threshold and the maximum grayscale variance threshold with the unit line segment expansion area of ​​the corresponding line segment, and obtain the line grayscale mean variance feature; The maximum average grayscale threshold is the maximum grayscale value among the n average grayscale values ​​corresponding to the same dividing line segment in n normal brain CT images; The maximum grayscale variance threshold is the maximum grayscale variance threshold among the n grayscale variances corresponding to the same dividing line segment in n normal brain CT images.

5. The diagnostic management system based on brain CT image analysis according to claim 1, characterized in that: The step of obtaining abnormal brain information comprises: Select a normal brain model corresponding to the patient based on the patient's basic information; Compare the patient's brain CT images with the normal brain model in terms of location distribution, volume, grayscale, and three-dimensional structure lines; The brain areas that are not within the location distribution characteristics, volume characteristics, grayscale characteristics, three-dimensional structure lines, and line grayscale mean variance characteristics of the normal brain model are marked to obtain abnormal brain information.

6. The diagnostic management system based on brain CT image analysis according to claim 1, characterized in that: The step of inputting abnormal brain information into a pre-constructed brain disease diagnosis model to obtain a diagnosis result comprises: Build and train models for diagnosing brain diseases; Input abnormal brain information into the trained brain disease diagnosis model; The brain disease diagnosis model outputs the diagnosis results corresponding to the abnormal brain information.

7. The diagnostic management system based on brain CT image analysis according to claim 1, characterized in that: The step of constructing a diagnostic database and generating a disease development path based on matching the diagnostic result with the diagnostic database includes: Obtain clinical information and brain CT images corresponding to different types of brain diseases, associate the same type of brain disease with the corresponding clinical information and brain CT images, and then obtain a diagnostic database; The diagnosis result is input into the diagnosis database, and a conventional diagnosis path corresponding to the diagnosis result in the diagnosis database is obtained based on a similarity algorithm; Obtain the clinical stage corresponding to the diagnosis result, obtain the path nodes that the diagnosis result has passed based on the clinical stage, and obtain the patient's disease progression speed through the passed path nodes; Collect and determine the patient's gain factor and inhibition factor, as well as the gain speed and inhibition speed corresponding to the gain factor and inhibition factor, respectively; The time between adjacent path nodes is adjusted based on the disease development speed, gain speed and inhibition speed, thereby obtaining the disease development path; The adjustment methods are: The actual disease development speed is obtained by adding the disease development speed to the gain speed and subtracting the inhibition speed. If the actual disease development speed is greater than zero, other inhibition factors are added until the actual disease development speed is equal to zero or less than zero.

8. The diagnostic management system based on brain CT image analysis according to claim 7, characterized in that: The conventional diagnostic path is obtained as follows: Acquire clinical information and brain CT images of the same brain disease at different times, and construct a conventional development path according to the chronological order of the clinical information, wherein a path node is added at a position corresponding to the time when a clinical stage is reached on the conventional development path, and the path node includes clinical information corresponding to the clinical stage of the brain disease; Obtaining the gain factor and inhibition factor of the brain disease in the conventional development path, and obtaining the gain speed or inhibition speed corresponding to each gain factor or each inhibition factor between adjacent path nodes; The gain factor is a factor that increases the development speed of the brain disease at adjacent path nodes; the inhibition factor is a factor that reduces the development speed of the brain disease at adjacent path nodes.

9. The diagnostic management system based on brain CT image analysis according to claim 1, characterized in that: The step of collecting the patient's clinical information in real time to obtain the real-time clinical information, and comparing the real-time clinical information with the disease development path to obtain the comparison result includes: Preset the collection cycle and obtain the patient's real-time clinical information through the collection cycle; Acquire the path node corresponding to the patient's real-time clinical information to obtain the real-time path node; The real-time clinical information of the real-time pathway node is compared with the clinical information corresponding to the pathway node on the disease development pathway to obtain a comparison result.

10. The diagnostic management system based on brain CT image analysis according to claim 2, characterized in that: The step of presetting the deviation sequence, comparing the comparison result with the deviation sequence, and determining whether to update the disease development pathway comprises: Preset a deviation interval in the deviation sequence, and compare the data in the comparison result with the corresponding deviation interval; If the data in the comparison result is within the deviation interval in the deviation sequence, the pathway nodes in the disease development pathway are replaced by the real-time pathway nodes, thereby updating the pathway nodes in the disease development pathway; If the comparison result is not within the deviation interval in the deviation sequence, the path node in the disease development path continues to be maintained.

Images