A medical imaging intelligent diagnosis data analysis system based on big data

Abstract

The present invention discloses a medical image intelligent diagnosis data analysis system based on big data, which relates to the field of data processing technology and includes a medical image data acquisition module, a medical image database, a medical image data preprocessing module, a medical image data processing module and a medical image data diagnosis module. The medical image data acquisition module is used to acquire medical image data of patients, the medical image database stores historical medical image data of all patients, the medical image data preprocessing module is used to preprocess the acquired medical image data, and input the processed medical image data into the medical image data processing module. The present invention solves the problems that manual processing is time-consuming and labor-intensive, tiny lesions are difficult to highlight, lesion boundaries are blurred, the clarity of medical images is affected by the patient's condition, and comprehensive consideration needs to be given to the patient's disease condition.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to a medical image intelligent diagnosis data analysis system based on big data. Background Art

[0002] Medical imaging refers to the technology and process of obtaining images of internal tissues of the human body or a part of the human body in a non-invasive manner for medical treatment or medical research. In medical diagnosis, medical imaging is a commonly used diagnostic method. When doctors diagnose a condition by looking at the patient's medical images, they sometimes need to perform manual processing to make the lesion area in the medical image appear more clearly. The main methods of manual processing include image enhancement, image restoration, image segmentation and image fusion. These manual processing methods have an important impact on improving the accuracy of diagnosis.

[0003] At present, in medical diagnosis, doctors often need to manually process images to make the lesion area clearer, but this method has many problems. On the one hand, doctors have limited energy and face many patients. Manual image processing is time-consuming and laborious, which reduces diagnostic efficiency. Long-term work can easily lead to fatigue, which reduces the accuracy of diagnosis. On the other hand, different lesions appear differently in images. It is difficult to highlight tiny lesions manually. When the boundaries of lesions are blurred, manual processing is still not conducive to diagnosis. In addition, the clarity of medical images is affected by the patient's condition. Doctors need to observe from multiple angles, which is time-consuming. In addition, doctors need to consider factors such as the patient's age and the length of the disease when diagnosing, because the same disease manifests differently in different patients, which puts higher demands on accurate diagnosis. In view of this, we propose a medical image intelligent diagnosis data analysis system based on big data. Summary of the invention

[0004] The purpose of the present invention is to provide a medical imaging intelligent diagnosis data analysis system based on big data. The technical problem to be solved is to improve the efficiency and accuracy of medical image processing, to solve the technical problems that manual processing is time-consuming and labor-intensive, tiny lesions are difficult to highlight, lesion boundaries are blurred, the clarity of medical images is affected by the patient's condition, and the need to be comprehensively considered in combination with the patient's disease condition.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a medical image intelligent diagnosis data analysis system based on big data, comprising a medical image data acquisition module, a medical image database, a medical image data preprocessing module, a medical image data processing module and a medical image data diagnosis module; The medical image data acquisition module is used to acquire medical image data of patients; The medical imaging database stores historical medical imaging data of all patients; The medical image data preprocessing module is used to preprocess the collected medical image data and input the processed medical image data into the medical image data processing module; The medical image data processing module includes an image contrast contour construction module, and the image contrast contour construction module is used to construct a contrast contour for the processed medical image data; The medical image data diagnosis module diagnoses medical image diseases based on the contrast contour construction data of the medical image output by the medical image data processing module.

[0006] Wherein, the medical imaging data diagnosis module includes a first case information set establishment module, a first matching value calculation module, a disease course prediction value calculation module and a first data set comparison module; The first case information set establishment module is used to establish the patient's condition level; The first matching value calculation module is used to calculate the matching value of the patient case established by the first case information set establishment module value; The disease course prediction value calculation module is used to calculate the disease course prediction value of the patient case established by the first case information set establishment module. calculate; The first data set comparison module predicts the course of disease of the patient case according to The data is constructed by comparing the contours of the medical images output by the image data processing module to diagnose medical image diseases and determine the type of disease in the patient.

[0007] The present invention uses big data technology to comprehensively consider the patient's condition based on massive historical data and scientific algorithms, and automatically collects, pre-processes and analyzes medical imaging data, freeing doctors from the cumbersome manual image processing process, allowing them to devote more energy to the review of the system's diagnosis results and comprehensive analysis of the condition. Doctors do not need to manually process images one by one, which greatly saves time and can quickly give diagnosis results. It not only reduces the physical and mental labor intensity of doctors, but also reduces manpower input, optimizes the allocation of medical resources, avoids misjudgments caused by human negligence or experience limitations, greatly improves the accuracy of diagnosis, and allows doctors to handle more patient cases per unit time, significantly improving overall diagnostic efficiency.

[0008] Preferably, preprocessing the collected medical image data includes: segmenting the collected medical image data to obtain the course data of all patients. , Age of onset data and medical imaging data , perform data anomaly removal and data processing on these data; Among them, the data processing adopts the deviation standardization formula: , where Represented as raw data, and are the maximum and minimum values ​​in the original data respectively.

[0009] A method for analyzing medical image intelligent diagnosis data based on big data comprises the following steps: S1: Acquire and collect the patient's medical imaging data through the medical imaging data acquisition module ,in, Indicates the order of collecting medical imaging data, and the collected data is stored in the medical imaging database; S2: The medical image data preprocessing module processes the medical data collected by the medical image data acquisition module. Perform pre-processing; S3: Extract medical imaging data of all patients , and the medical imaging data of all patients Construct comparative contours, medical imaging data of all patients The number of medical images is , where Indicates Medical images, Indicates the number of medical images; S4: The onset age data of all patients in the medical database , Number of medical images , medical history data , image contrast contour Establishing a first case information set as an attribute in a first case information data set of a first case information set building module; S5: The number of medical records in the first case information dataset is taken as a set , the number of images of the patient is counted as The age of patients is , the age of all patients in the statistical sample is , the course of disease is , indicating the number of images is The course of disease at that time, through mathematical analysis, the relationship between age, course of disease, and incidence rate of the patients in the sample was obtained; S6: Extract the image contrast contour of the patient , and import into the first medical record information data set; S7: Extract the patient's medical imaging data The pixel values ​​of all pixels in ; S8: Extracting medical imaging data The medical image in the image is counted and the number of all feature points on the medical image is counted. The number of feature points is , the distance between two feature points is marked as ; S9: The patient's age of onset is collected by extracting the first medical record information data set , the patient's disease course prediction value and the number of feature points Substitute it into the preset formula to get the predicted value of the pixel value , and the statistics of Pixel values ​​and predicted values Make comparisons; If the relationship between the predicted value and the pixel value is positive, the predicted value is rounded and marked as 1; if the relationship between the predicted value and the pixel value is negative, the predicted value is rounded and marked as 0, and all the marks are constructed to obtain the predicted contrast contour. ; S10: Judgment and Are they the same? If not, then Re-extract feature points from all pixels in the image and reconstruct the predicted contrast contour. ,until and The same as the patient's diagnosis information in the first case information data set; S11: Contrast the predicted contour of the image Extract the patient's data, import it into the medical imaging data processing module, and extract the predicted contrast contours and disease course prediction values ​​of all images in the first case information set , and then count the images in the first case information set as The patient's disease course is , the course of disease is The number of images is ,in The relationship between the size of is positive, and it is substituted into the disease course estimation formula The patient's estimated course of disease is obtained , and stored in the first case data set; S12: Extract the patient medical image data collected by the medical image data collection module and input it into the In the model, we add the activation function softmax function And the cross entropy function, get the final disease diagnosis model , all disease diagnosis models Combine to get a set of diagnostic models , and stored in the medical image data processing module.

[0010] Preferably, in step S2, the collected medical data Preprocessing is performed, including: S201: Collect the medical imaging data Perform data segmentation to obtain the course data of all patients , Age of onset data of all patients and medical imaging data of all patients ; S202: Medical data for all patients , Age of onset data of all patients and medical imaging data of all patients Perform data exception processing; S203: Medical course data for all patients , Age of onset data of all patients and medical imaging data of all patients Perform data digitization processing.

[0011] Preferably, in step S3, the medical imaging data of all patients Construct a comparative profile, specifically: S301: Establishing a coordinate system, taking all pixel points of the medical image as coordinate points in the coordinate system, and constructing a dot matrix in the coordinate system according to the number of pixel points of the medical image data; S302: Medical imaging data of all patients The operation of step S301 is performed; S303: Compare the pixel values ​​of all pixel coordinates with the average pixel value of the medical image Perform the difference and take the absolute value of the difference to get the difference set ; S304: Difference set Greater than the preset threshold As feature points, the set of feature points is used as the medical imaging data of all patients. Image contrast contour .

[0012] Preferably, the first medical record information data set includes a first case information data, and the first case information data includes the onset age data of all patients. , Number of medical images , medical history data And image contrast contour .

[0013] Preferably, the relationship between the patient's age, course of disease and incidence rate in step S5 is obtained by the following mathematical analysis, specifically, , where and is the weight value, As a constant, the patient's age is and course of disease Substitute into the formula to get the patient's disease course prediction value .

[0014] Preferably, in step S8, by using a preset formula Get the first The predicted pixel value of the pixel , the predicted value With pixel value Compare and get the predicted contrast profile ,until and Same, will and Import into the medical image data processing module.

[0015] Preferably, in step S9, by using a preset formula , get the predicted value of the pixel value , where Indicates that the pixel value is pixel values.

[0016] Preferably, the cross entropy function in step S12 is: ; In the formula, Indicates The samples belong to The probability of classification, i.e. the predicted value, represents the predicted probability distribution, represents the actual probability distribution, The value of , which means each pixel and If they are the same, the value is 0 if they are different, and 1 if they are the same.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention uses big data technology to comprehensively consider the patient's condition based on massive historical data and scientific algorithms, and automatically collects, preprocesses and analyzes medical imaging data, freeing doctors from the cumbersome manual image processing process, allowing them to devote more energy to the review of the system's diagnostic results and comprehensive assessment of the condition. Doctors do not need to manually process images one by one, which greatly saves time and can quickly give diagnostic results. It not only reduces the physical and mental labor intensity of doctors, but also reduces manpower input, optimizes the allocation of medical resources, avoids misjudgments caused by human negligence or experience limitations, greatly improves the accuracy of diagnosis, and allows doctors to handle more patient cases per unit time, significantly improving overall diagnostic efficiency.

[0018] 2. The present invention utilizes image contrast contour construction and big data sample matching. By constructing fine image contrast contours and complex data analysis models, it can dig out subtle abnormalities hidden in the image, accurately capture the characteristics of various lesions, not miss any tiny lesions, clearly outline the boundaries of lesions, and highlight the boundary contours of the lesion area, thereby allowing the image data processing module and the image data diagnosis module to focus more attention on the lesion area, reducing the possibility of the lesion area being neglected during the processing process and improving the accuracy of the diagnosis results.

[0019] 3. The present invention analyzes the correlation between age, disease course and morbidity of all patients in the patient data set, and then obtains the patient's disease course prediction value by calculation, so that the disease course prediction value can be obtained according to the patient's personal situation, thereby further improving the accuracy of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0021] Embodiment 1: Figure 1 As shown, the present invention relates to a medical image intelligent diagnosis data analysis system based on big data, including a medical image data acquisition module, a medical image database, a medical image data preprocessing module, a medical image data processing module and a medical image data diagnosis module; The medical image data acquisition module is used to acquire medical image data of patients; The medical imaging database stores historical medical imaging data of all patients; The medical image data preprocessing module is used to preprocess the collected medical image data and input the processed medical image data into the medical image data processing module; In an embodiment of the present invention, preprocessing the collected medical image data includes: segmenting the collected medical image data to obtain the course data of all patients. , Age of onset data and medical imaging data , perform data anomaly removal and data processing on these data; Among them, the data processing adopts the deviation standardization formula: , where Represented as raw data, and are the maximum and minimum values ​​in the original data respectively; The medical image data processing module includes an image contrast contour construction module, and the image contrast contour construction module is used to construct a contrast contour for the processed medical image data; The medical image data diagnosis module diagnoses medical image diseases based on the contrast contour construction data of the medical image output by the medical image data processing module.

[0022] Wherein, the medical imaging data diagnosis module includes a first case information set establishment module, a first matching value calculation module, a disease course prediction value calculation module and a first data set comparison module; The first case information set establishment module is used to establish the patient's condition level; The first matching value calculation module is used to calculate the matching value of the patient case established by the first case information set establishment module value; The disease course prediction value calculation module is used to calculate the disease course prediction value of the patient case established by the first case information set establishment module. calculate; The first data set comparison module predicts the course of disease of the patient case according to The data is constructed by comparing the contours of the medical images output by the image data processing module to diagnose medical image diseases and determine the type of disease in the patient.

[0023] The present invention uses big data technology to comprehensively consider the patient's condition based on massive historical data and scientific algorithms, and automatically collects, pre-processes and analyzes medical imaging data, freeing doctors from the cumbersome manual image processing process, allowing them to devote more energy to the review of the system's diagnosis results and comprehensive analysis of the condition. Doctors do not need to manually process images one by one, which greatly saves time and can quickly give diagnosis results. It not only reduces the physical and mental labor intensity of doctors, but also reduces manpower input, optimizes the allocation of medical resources, avoids misjudgments caused by human negligence or experience limitations, greatly improves the accuracy of diagnosis, and allows doctors to handle more patient cases per unit time, significantly improving overall diagnostic efficiency.

[0024] Embodiment 2: A method for analyzing medical image intelligent diagnosis data based on big data, comprising the following steps: S1: Acquire and collect the patient's medical imaging data through the medical imaging data acquisition module ,in, Indicates the order of collecting medical imaging data, and the collected data is stored in the medical imaging database; S2: The medical image data preprocessing module processes the medical data collected by the medical image data acquisition module. Perform pre-processing; As another embodiment of the present invention, the collected medical data in step S2 Preprocessing is performed, including: S201: Collect the medical imaging data Perform data segmentation to obtain the course data of all patients , Age of onset data of all patients and medical imaging data of all patients ; S202: Medical data for all patients , Age of onset data of all patients and medical imaging data of all patients Perform data exception processing; S203: Medical course data for all patients , Age of onset data of all patients and medical imaging data of all patients Carry out data digitization processing; S3: Extract medical imaging data of all patients , and the medical imaging data of all patients Perform comparative contour construction; As another embodiment of the present invention, the medical imaging data of all patients The number of medical images is , where Indicates Medical images, Indicates the number of medical images; Wherein, in step S3, the medical imaging data of all patients are Construct a comparative profile, specifically: S301: Establishing a coordinate system, taking all pixel points of the medical image as coordinate points in the coordinate system, and constructing a dot matrix in the coordinate system according to the number of pixel points of the medical image data; S302: Medical imaging data of all patients The operation of step S301 is performed; S303: Compare the pixel values ​​of all pixel coordinates with the average pixel value of the medical image Perform the difference and take the absolute value of the difference to get the difference set ; S304: Difference set Greater than the preset threshold As feature points, the set of feature points is used as the medical imaging data of all patients. Image contrast contour ; S4: The onset age data of all patients in the medical database , Number of medical images , medical history data , image contrast contour Establishing a first case information set as an attribute in a first case information data set of a first case information set building module; As another embodiment of the present invention, the first medical record information data set includes a first case information data, and the first case information data includes the onset age data of all patients. , Number of medical images , medical history data And image contrast contour ; S5: The number of medical records in the first case information dataset is taken as a set , the number of images of the patient is counted as The age of patients is , the age of all patients in the statistical sample is , the course of disease is , indicating the number of images is The course of disease at that time, through mathematical analysis, the relationship between age, course of disease, and incidence rate of the patients in the sample was obtained; As another embodiment of the present invention, the relationship between the patient's age, course of disease, and incidence rate in step S5 is obtained by the following mathematical analysis, specifically, , where and is the weight value, As a constant, the patient's age is and course of disease Substitute into the formula to get the patient's disease course prediction value ; The present invention analyzes the correlation between age, disease course and morbidity of all patients in the patient data set, and then obtains the patient's disease course prediction value by calculation, so that the disease course prediction value can be obtained according to the patient's personal situation, thereby further improving the accuracy of diagnosis.

[0025] S6: Extract the image contrast contour of the patient , and import into the first medical record information data set; S7: Extract the patient's medical imaging data The pixel values ​​of all pixels in ; S8: Extracting medical imaging data The medical image in the image is counted and the number of all feature points on the medical image is counted. The number of feature points is , the distance between two feature points is marked as ; As another embodiment of the present invention, in step S8, by using a preset formula Get the first The predicted pixel value of the pixel , the predicted value With pixel value Compare and get the predicted contrast profile ,until and Same, will and Import into the medical imaging data processing module; S9: The patient's age of onset is collected by extracting the first medical record information data set , the patient's disease course prediction value and the number of feature points Substitute it into the preset formula to get the predicted value of the pixel value , and the statistics of Pixel values ​​and predicted values Make comparisons; If the relationship between the predicted value and the pixel value is positive, the predicted value is rounded and marked as 1; if the relationship between the predicted value and the pixel value is negative, the predicted value is rounded and marked as 0, and all the marks are constructed to obtain the predicted contrast contour. ; As another embodiment of the present invention, in step S9, by using a preset formula , get the predicted value of the pixel value , where Indicates that the pixel value is pixel value; S10: Judgment and Are they the same? If not, then Re-extract feature points from all pixels in the image and reconstruct the predicted contrast contour. ,until and The same as the patient's diagnosis information in the first case information data set; S11: Contrast the predicted contour of the image Extract the patient's data, import it into the medical imaging data processing module, and extract the predicted contrast contours and disease course prediction values ​​of all images in the first case information set , and then count the images in the first case information set as The patient's disease course is , the course of disease is The number of images is ,in The relationship between the magnitude of is positive, and it is substituted into the disease course estimation formula to obtain the patient's estimated disease course value , and stored in the first case data set; As another embodiment of the present invention, the disease course estimation formula in step S11 is: ; S12: Extract the patient medical image data collected by the medical image data collection module and input it into the In the model, we add the activation function softmax function And the cross entropy function, get the final disease diagnosis model , all disease diagnosis models Combine to get a set of diagnostic models , and stored in the medical image data processing module; Wherein, the cross entropy function in step S12 is: , where Indicates The samples belong to The probability of classification, i.e. the predicted value, represents the predicted probability distribution, represents the actual probability distribution, The value of , which means each pixel and If they are the same, the value is 0 if they are different, and 1 if they are the same; The present invention utilizes image contrast contour construction and big data sample matching. By constructing fine image contrast contours and complex data analysis models, it can dig out subtle abnormalities hidden in images, accurately capture various lesion characteristics, not miss any tiny lesions, clearly outline the lesion boundaries, and highlight the boundary contours of the lesion area, thereby allowing the image data processing module and the image data diagnosis module to focus more attention on the lesion area, reducing the possibility of the lesion area being neglected during the processing process and improving the accuracy of the diagnosis results.

[0026] The embodiments of the present invention disclose preferred embodiments, but are not limited thereto. A person skilled in the art can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. However, as long as they do not deviate from the spirit of the present invention, they are all within the protection scope of the present invention.

Claims

1. A medical image intelligent diagnosis data analysis system based on big data, characterized in that: include: A medical imaging data acquisition module, used for acquiring medical imaging data of patients; Medical imaging database, which stores historical medical imaging data of all patients; A medical image data preprocessing module, used for preprocessing the collected medical image data; The medical image data processing module includes an image contrast contour construction module, which is used to construct a contrast contour for the processed medical image data; Medical imaging data diagnosis module, which constructs data based on the contrast contours of medical images to diagnose medical imaging diseases; Wherein, the medical imaging data diagnosis module includes: A first case information set establishment module is used to establish the patient's condition level; The first matching value calculation module is used to calculate the matching value of the patient case established by the first case information set establishment module value; The disease course prediction value calculation module is used to calculate the disease course prediction value of the patient case established by the first case information set establishment module calculate; The first data set comparison module, based on the disease course prediction value The data is constructed by comparing the contours of medical images to diagnose medical imaging diseases and determine the type of disease in the patient.

2. According to the big data-based medical image intelligent diagnosis data analysis system of claim 1, it is characterized in that: The preprocessing includes: segmenting the collected medical image data to obtain the course data of all patients , Age of onset data and medical imaging data , and then perform data anomaly removal and data processing; Among them, the data processing adopts the deviation standardization formula: , where Represented as raw data, and are the maximum and minimum values ​​in the original data respectively.

3. A method for analyzing medical image intelligent diagnosis data based on big data, which is used in a medical image intelligent diagnosis data analysis system based on big data as claimed in any one of claims 1 to 2, characterized in that: The steps include: S1: Acquire and collect medical imaging data of patients ,The collected data are stored in the medical imaging database; S2: Collected medical data Perform pre-processing; S3: Extract medical imaging data of all patients , and the medical imaging data of all patients Perform comparative contour construction; S4: The onset age data of all patients in the medical database , Number of medical images , medical history data , image contrast contour Establishing a first case information set as an attribute in a first case information data set of a first case information set building module; S5: The number of medical records in the first case information dataset is taken as a set The age of patients onset is , the course of disease is , indicating the number of images is The course of disease at that time, through mathematical analysis, the relationship between age, course of disease, and incidence rate of the patients in the sample was obtained; S6: Extract the image contrast contour of the patient , and import into the first medical record information data set; S7: Extract the patient's medical imaging data The pixel values ​​of all pixels in ; S8: Extracting medical imaging data The medical images in the image are counted and the number of all feature points is counted. , the distance between two feature points is marked as ; S9: The patient's age of onset is collected by extracting the first medical record information data set , the patient's disease course prediction value And the number of feature points are brought into the preset formula to get the predicted value of the pixel value , and the statistics of Pixel values ​​and predicted values Make comparisons; If the relationship between the predicted value and the pixel value is positive, the predicted value is rounded to 1; If the relationship between the predicted value and the pixel value is negative, the predicted value is rounded to 0 and all the marks are constructed to obtain the predicted contrast contour. ; S10: Judgment and Are they the same? If not, then Re-extract feature points from all pixels in the image and reconstruct the predicted contrast contour. ,until and The same as the patient's diagnosis information in the first case information data set; S11: Contrast the predicted contour of the image Extract the patient's data, import it into the medical imaging data processing module, and extract the predicted contrast contours and disease course prediction values ​​of all images in the first case information set , and then count the images in the first case information set as The patient's disease course is , the course of disease is The number of images is ,in The relationship between the size of is positive, and it is substituted into the disease course estimation formula The patient's estimated course of disease is obtained , and stored in the first case data set; S12: Extract the patient medical image data collected by the medical image data collection module and input it into the In the model, we add the activation function softmax function And the cross entropy function, get the final disease diagnosis model , all disease diagnosis models Combine to get a set of diagnostic models , and stored in the medical image data processing module.

4. The method for analyzing medical image intelligent diagnosis data based on big data according to claim 3, characterized in that: In step S2, the collected medical data Preprocessing is performed, including: S201: Collect the medical imaging data Perform data segmentation to obtain the course data of all patients , Age of onset data of all patients and medical imaging data of all patients ; S202: Medical data for all patients , Age of onset data of all patients and medical imaging data of all patients Perform data exception processing; S203: Medical data for all patients , Age of onset data of all patients and medical imaging data of all patients Perform data digitization processing.

5. The method for analyzing medical image intelligent diagnosis data based on big data according to claim 4, characterized in that: In step S3, the medical imaging data of all patients are Construct a comparative profile, specifically: S301: Establishing a coordinate system, taking all pixel points of the medical image as coordinate points in the coordinate system, and constructing a dot matrix in the coordinate system according to the number of pixel points of the medical image data; S302: Medical imaging data of all patients The operation of step S301 is performed; S303: Compare the pixel values ​​of all pixel coordinate points with the average pixel value of the medical image Perform the difference and take the absolute value of the difference to get the difference set ; S304: Difference set Greater than the preset threshold As feature points, the set of feature points is used as the medical imaging data of all patients. Image contrast contour .

6. The method for analyzing medical image intelligent diagnosis data based on big data according to claim 5, characterized in that: The first medical record information data set includes a first case information data, and the first case information data includes the onset age data of all patients. , Number of medical images , medical history data And image contrast contour .

7. The method for analyzing medical image intelligent diagnosis data based on big data according to claim 6, characterized in that: The relationship between the patient's age, course of disease and incidence rate in step S5 is obtained by the following mathematical analysis, specifically, , where and is the weight value, As a constant, the patient's age is and course of disease Substitute into the formula to get the patient's disease course prediction value .

8. The method for analyzing medical image intelligent diagnosis data based on big data according to claim 7, characterized in that: In step S8, the preset formula Get the first The predicted pixel value of the pixel , the predicted value With pixel value Compare and get the predicted contrast profile ,until and Same, will and Import into the medical image data processing module.

9. The method for analyzing medical image intelligent diagnosis data based on big data according to claim 8, characterized in that: In step S9, the preset formula , get the predicted value of the pixel value , where Indicates that the pixel value is pixel values.

10. The method for analyzing medical image intelligent diagnosis data based on big data according to claim 9, characterized in that: The cross entropy function in step S12 is: ； In the formula, Indicates The samples belong to The probability of classification, i.e. the predicted value, represents the predicted probability distribution, represents the actual probability distribution, The value of , which means each pixel and If they are the same, the value is 0 if they are different, and 1 if they are the same.

Images