Liver cirrhosis pathology recognition system based on artificial intelligence

Abstract

The invention, which relates to the technical field of liver cirrhosis identification, discloses an artificial intelligence-based liver cirrhosis pathology identification system comprising a data acquisition module, a data preprocessing module, a classification evaluation module, a time change module, a risk prediction module and a data updating module. The classification evaluation module evaluates the cirrhosis risk according to the image data and the biochemical index data, the time change module is used for comparing influence factors in the image data and the biochemical index data at different times, and the classification evaluation module analyzes the image and biochemical index data, quantifies the cirrhosis risk score and improves the risk evaluation accuracy. The time change module monitors the change of scores along with time, the risk score change rate is monitored, the influence weight is adjusted in real time, the sensitivity and accuracy of the system are enhanced, finally, the scores are integrated through the risk prediction module, a threshold value is set, subjective judgment is avoided, the recognition accuracy is improved, high-risk patients are recognized in advance, and the risk prediction efficiency is improved. The liver cancer occurrence rate is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of liver cirrhosis recognition, and in particular to a liver cirrhosis pathology recognition system based on artificial intelligence. Background Art

[0002] Cirrhosis is a common clinical chronic progressive liver disease, which is specifically defined as the histological development of fibrous bands around regenerative nodules caused by chronic liver damage. As cirrhosis further develops, it can lead to portal hypertension and end-stage liver disease, with a high mortality rate. Early detection and treatment of cirrhosis is very necessary. Clinically, we can also find that a small number of chronic hepatitis B virus patients directly transform into liver cancer, but the vast majority of patients gradually progress on the basis of hepatitis B cirrhosis and eventually develop liver cancer. Therefore, in clinical practice, it is necessary to establish a predictor to assess the risk of liver cancer in hepatitis B cirrhosis, and to predict the risk factor of liver cancer in patients at an early stage, which can reduce the incidence of liver cancer in patients with hepatitis B cirrhosis, and has practical significance and clinical promotion value.

[0003] At present, most patients' condition gradually progresses on the basis of hepatitis B cirrhosis and eventually develops liver cancer. This is greatly affected by the subjective judgment of doctors. Different doctors may interpret the imaging data differently, resulting in inaccurate evaluation results. At the same time, there is a lack of real-time analysis of changes in cirrhosis data, making it difficult to take preventive measures in advance, thereby increasing the incidence of liver cancer in patients with hepatitis B cirrhosis. Summary of the invention

[0004] The purpose of the present invention is to provide a liver cirrhosis pathology recognition system based on artificial intelligence, which solves the problems raised in the above-mentioned background technology.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a liver cirrhosis pathology recognition system based on artificial intelligence, comprising a data acquisition module, a data preprocessing module, a classification evaluation module, a time change module, a risk prediction module and a data update module; The data acquisition module is used to collect patient data, which includes imaging data and biochemical index data; The classification assessment module assesses the risk of liver cirrhosis based on the imaging data and biochemical index data to obtain a risk score R based on the biochemical index data. sh and the risk score R based on imaging data yx ; The time change module is used to compare the influencing factors in the image data and biochemical index data at different times, obtain the change rate of each influencing factor, and set a threshold for it. When one of the influencing factors exceeds the threshold, its weight is automatically adjusted to ensure the sensitivity and accuracy of the system; The risk prediction module is based on the risk score R based on the biochemical index data. sh and the risk score R based on imaging data yx , and the comprehensive risk score R at time t + Δt is obtained based on the changes all (t+Δt), and set a threshold for it, and judge the patient's liver cirrhosis condition in the future according to the threshold, so that preventive measures can be taken in advance.

[0006] Optionally, the classification assessment module assessment process includes assessing the risk of liver cirrhosis based on biochemical indicator data and imaging data, wherein the assessment process based on biochemical indicator data is as follows: ; Where R sh is the risk score based on biochemical indicator data, sh represents biochemical indicator data; L1 is the intercept one, which represents the risk base value without the influence of other factors; ALT pat is the patient's trough aminotransferase value; ALT nor is the intraglutathione transaminase threshold; is the weight of intravalley aminotransferase, ranging from 0 to 1; AST pat is the patient's aspartate aminotransferase value; AST nor is the aspartate aminotransferase threshold; α2 is the impact weight of aspartate aminotransferase, ranging from 0 to 1; TBLL pat is the patient's albumin value; TBLL nor is the albumin threshold; α3 is the albumin impact weight, ranging from 0 to 1; Risk score R based on biochemical index data sh The higher the value, the greater the risk of liver cirrhosis. Biochemical index data directly reflects liver function and damage; The evaluation process based on image data is as follows: ; Where R yx It is a risk score based on imaging data; L2 is intercept two; CH pat is the stiffness of the patient's liver; CH nor is the liver stiffness threshold; β1 is the liver stiffness weight; CDpat is the patient's liver volume; CD nor is the liver volume threshold; β2 is the liver volume weight; Risk score R based on imaging data yx Provides liver structure and morphology information, R yx The higher the value, the greater the risk of cirrhosis. By evaluating the risk of cirrhosis based on biochemical indicator data and imaging data, doctors can understand the impact of each factor on the risk of cirrhosis.

[0007] Optionally, the time change module includes a change rate assessment based on biochemical indicator data and based on image data, wherein the change rate assessment process based on biochemical indicator data is as follows: ; Where ΔR sh,i is the risk score change rate based on the i-th data in the biochemical index data; R sh,i (t) is the risk score of the i-th data in the biochemical index data at the current time; R sh,i (t-Δt) is the risk score of the i-th data in the biochemical index data at time t-Δt; Δt is the time interval in months; The value range of i is 1 to 3. When i is 1, it indicates aspartate aminotransferase, when i is 2, it indicates aspartate aminotransferase, and when i is 3, it indicates albumin. Set ΔR sh,i The change threshold is Y1 sh,i , when ΔR sh,i Greater than Y1 sh,i When , and i is 1, it means that the fluctuation of the risk score change rate of the intra-valley aminotransferase value in the biochemical index data is too large. At this time, the intra-valley aminotransferase weight α1 is adjusted. The adjustment process is as follows: ; Among them, xα1 is the adjusted aminotransferase weight in valley; k1 is the adjustment coefficient 1, α1+k1≤1; The evaluation process based on the rate of change of image data is as follows: ; Where ΔR yx,n It is the risk score change rate based on the nth data in the imaging data; R yx,n (t) is the risk score of the nth data in the image data at the current time; R yx,n(t-Δt) is the risk score of the nth data in the imaging data at time t-Δt; Δt is the time interval; The value range of n is 1 to 2. When n is 1, it indicates liver stiffness, and when n is 2, it indicates liver volume. yx,n The change threshold 2 is Y2 yx,n , when ΔR yx,n Greater than Y2 yx,n When , and n is 1, it means that the fluctuation of the liver stiffness risk score in the imaging data is too large. At this time, the liver stiffness weight β1 is adjusted. The adjustment process is as follows: ; Among them, xβ1 is the adjusted internal liver stiffness weight; k2 is the adjustment coefficient 2, β1+k2≤1; By calculating the risk score change rate of each biochemical indicator data and imaging data, we can accurately understand the changes in the factors affecting cirrhosis, set thresholds for each biochemical indicator data and imaging data separately, and adjust the impact weight of each data in real time according to the risk score change rate, so that the system can automatically adapt to the impact of data changes on cirrhosis.

[0008] Optionally, the risk prediction module evaluation process is as follows: ; Where R all (t + Δt) is the comprehensive risk score at time t + Δt; R sh It is a risk score based on biochemical marker data; ΔR sh is the risk score change rate of the biochemical index data; W1 is the weight of biochemical index data; R yx It is a risk score based on imaging data; ΔR yx is the rate of change of risk score of imaging data; W2 is the image data weight; Setting R all The risk threshold of (t+Δt) is Z, R all The larger (t+Δt) is, the more severe the patient's liver cirrhosis will be in the future. all When (t+Δt) is greater than Z, it means that the patient's risk of cirrhosis is too high at time t+Δt, and preventive measures need to be taken in advance. By predicting the comprehensive risk score in the future, doctors can identify high-risk patients in advance and develop corresponding monitoring and intervention plans.

[0009] Compared with the current technology, the beneficial effects of the present invention are as follows: Effect 1. The present invention uses a classification assessment module to assess the risk of liver cirrhosis based on imaging data and biochemical indicator data, and obtains a risk score R based on biochemical indicator data. sh and the risk score R based on imaging data yx , analyze the patient's liver cirrhosis from different aspects, and quantify the results so that doctors and patients can understand, provide data support for decision-making, improve the accuracy of identification results, and then use the time change module to calculate the risk score R according to the biochemical index data sh and the risk score R based on imaging data yx , calculate the changes in biochemical index data and imaging data over time, and finally use the risk prediction module to calculate the risk score R based on the biochemical index data. sh and the risk score R based on imaging data yx By integrating and combining the changes over time, a comprehensive risk score for the future is obtained, and a threshold is set for it so that doctors can identify the risk of cirrhosis, avoid the influence of doctors' subjective judgment, improve the accuracy of cirrhosis pathology identification, and achieve the effect of cirrhosis risk prediction. It can identify high-risk patients in advance, take preventive measures in time, avoid deterioration, and effectively reduce the incidence of liver cancer in patients with hepatitis B cirrhosis.

[0010] Effect 2: The present invention can accurately understand the changes in the factors affecting cirrhosis by calculating the risk score change rate of each data in the biochemical indicator data and the imaging data, and set a threshold for each data in the biochemical indicator data and the imaging data separately. When the threshold is exceeded, the influence weight can be adjusted in real time, so that the system can adapt to the impact of data changes on cirrhosis, automatically adjust the results of the classification evaluation module, and improve the sensitivity and accuracy of the system.

[0011] Optionally, the data preprocessing module includes data cleaning, data standardization and feature selection.

[0012] Optionally, the patient data includes personal information data, and the personal information data includes the patient's age, gender and medical history.

[0013] Optionally, the data updating module is used to regularly update the patient data and liver cirrhosis standard data to ensure the accuracy and timeliness of the data.

[0014] Optionally, the system also includes an interactive interface module, and both doctors and patients can log in to the interactive interface module to transmit data to the system, and patients can upload personal information data through the interactive interface module. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a block diagram of the system modules of the present invention. DETAILED DESCRIPTION

[0016] 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.

[0017] Embodiment 1: See also Figure 1 ,This implementation provides an artificial intelligence-based liver cirrhosis pathology recognition system, including a data acquisition module, a data preprocessing module, a classification evaluation module, a time change module, a risk prediction module and a data update module; The data collection module is used to collect patient data, which includes imaging data and biochemical index data. The patient data also includes personal information data, which includes the patient's age, gender and medical history; It also includes an interactive interface module, and both doctors and patients can log in to the interactive interface module to transmit data to the system, such as patients uploading personal information data through the interactive interface module, and doctors uploading patient liver cirrhosis data analysis through the interactive interface module; The classification assessment module includes a classification assessment module and a risk prediction module; The classification assessment module assesses the risk of liver cirrhosis based on imaging data and biochemical index data, and obtains a risk score R based on biochemical index data. sh and the risk score R based on imaging data yx ; The time change module is used to compare the influencing factors in the image data and biochemical index data at different times, obtain the change rate of each influencing factor, and set a threshold for it. When one of the influencing factors exceeds the threshold, its weight is automatically adjusted to ensure the sensitivity and accuracy of the system; The risk prediction module is used to predict the risk of sh and the risk score R based on imaging data yx , and the comprehensive risk score R at time t + Δt is obtained based on the changes all (t+Δt), and set a threshold for it, and judge the patient's liver cirrhosis condition in the future according to the threshold, so that preventive measures can be taken in advance.

[0018] More specifically, in this embodiment: after the data acquisition module acquires the patient data, the patient data is divided into imaging data and biochemical index data according to different data types, and then the classification evaluation module evaluates the risk of liver cirrhosis according to the imaging data and the biochemical index data to obtain a risk score R based on the biochemical index data. sh and the risk score R based on imaging data yx Since both biochemical indicator data and imaging data can reflect the patient's liver cirrhosis condition, but biochemical indicator data usually directly reflects liver function and damage, imaging data provides information on liver structure and morphology. Imaging data can identify anatomical changes, tumors or degree of fibrosis of the liver, so that the patient's liver cirrhosis condition can be analyzed from different aspects and the results can be quantified for doctors and patients to understand. It can provide more detailed and targeted risk assessments, provide data support for decision-making, and improve the accuracy of identification results.

[0019] Then, the risk score R of the biochemical index data is calculated by the time change module. sh and the risk score R based on imaging data yx , calculate the time changes of each influencing factor in the biochemical index data and imaging data. The biochemical index data include glutamic transaminase, aspartate transaminase and albumin, and the imaging data include liver stiffness, liver volume, etc., and set a separate change threshold for each data. When one of the data exceeds the threshold, it means that the fluctuation range of the data is too large, which will increase the impact on liver cirrhosis. At this time, the influence weight is automatically adjusted, so that the system can capture the real-time changes of liver cirrhosis data and improve the sensitivity of the system. Finally, the risk score R based on the biochemical index data is used for risk prediction. sh and the risk score R based on imaging data yx The data are integrated and assigned different weights, and then combined with the rate of change data to obtain a comprehensive risk score for the future. This comprehensive risk score can more comprehensively represent the patient's liver cirrhosis condition and set a threshold for it so that doctors can identify liver cirrhosis. Cirrhosis can be identified through data analysis to avoid the influence of doctors' subjective judgment, improve the accuracy of liver cirrhosis pathology identification, and achieve the effect of liver cirrhosis risk prediction by combining the rate of change. High-risk patients can be identified in advance, so that preventive measures can be taken in time to avoid deterioration, effectively reducing the incidence of liver cancer in patients with hepatitis B cirrhosis.

[0020] Furthermore, the evaluation process of the classification evaluation module includes evaluating the risk of liver cirrhosis based on biochemical index data and imaging data, wherein the evaluation process based on biochemical index data is as follows: ; Where R shis the risk score based on biochemical indicator data, sh represents biochemical indicator data; L1 is the intercept one, which represents the risk base value without the influence of other factors; ALT pat is the patient's trough aminotransferase value; ALT nor is the intraglutathione transaminase threshold; is the weight of intravalley aminotransferase, ranging from 0 to 1; AST pat is the patient's aspartate aminotransferase value; AST nor is the aspartate aminotransferase threshold; α2 is the impact weight of aspartate aminotransferase, ranging from 0 to 1; TBLL pat is the patient's albumin value; TBLL nor is the albumin threshold; α3 is the albumin impact weight, ranging from 0 to 1; Risk score R based on biochemical index data sh The higher the value, the greater the risk of liver cirrhosis. Biochemical index data directly reflects liver function and damage; The evaluation process based on image data is as follows: ; Where R yx It is a risk score based on imaging data; L2 is intercept two; CH pat is the patient's liver stiffness; CH nor is the liver stiffness threshold; β1 is the liver stiffness weight; CD pat is the patient's liver volume; CD nor is the liver volume threshold; β2 is the liver volume weight; In actual evaluation, specific factors in biochemical index data and imaging data can be adjusted according to specific circumstances. For example, biochemical index data also include total bilirubin, alkaline phosphatase and alanine aminotransferase, and imaging data also include liver surface roughness, liver density and esophageal varices. Specifically, the risk score R based on imaging data yx Provides liver structure and morphology information, R yxThe higher the value, the greater the risk of cirrhosis. By evaluating the risk of cirrhosis based on biochemical indicator data and imaging data, doctors can understand the impact of each factor on the risk of cirrhosis, provide data support for subsequent decision-making, reduce the impact of subjective judgment, and improve the accuracy of pathology identification.

[0021] Furthermore, the time change module includes a change rate assessment based on biochemical indicator data and based on image data, wherein the change rate assessment process based on biochemical indicator data is as follows: ; Where ΔR sh,i is the risk score change rate based on the i-th data in the biochemical index data; R sh,i (t) is the risk score of the i-th data in the biochemical index data at the current time; R sh,i (t-Δt) is the risk score of the i-th data in the biochemical index data at time t-Δt; Δt is the time interval in months; The value range of i is 1 to 3. When i is 1, it indicates aspartate aminotransferase, when i is 2, it indicates aspartate aminotransferase, and when i is 3, it indicates albumin. Set ΔR sh,i The change threshold is Y1 sh,i , when ΔR sh,i Greater than Y1 sh,i When , and i is 1, it means that the fluctuation of the risk score change rate of the intra-valley aminotransferase value in the biochemical index data is too large. At this time, the intra-valley aminotransferase weight α1 is adjusted. The adjustment process is as follows: ; Among them, xα1 is the adjusted aminotransferase weight in valley; k1 is the adjustment coefficient 1, α1+k1≤1; The evaluation process based on the rate of change of image data is as follows: ; Where ΔR yx,n It is the risk score change rate based on the nth data in the imaging data; R yx,n (t) is the risk score of the nth data in the image data at the current time; R yx,n (t-Δt) is the risk score of the nth data in the imaging data at time t-Δt; Δt is the time interval; The value range of n is 1 to 2. When n is 1, it indicates liver stiffness, and when n is 2, it indicates liver volume. yx,n The change threshold 2 is Y2yx,n , when ΔR yx,n Greater than Y2 yx,n When , and n is 1, it means that the fluctuation of the liver stiffness risk score in the imaging data is too large. At this time, the liver stiffness weight β1 is adjusted. The adjustment process is as follows: ; Among them, xβ1 is the adjusted internal liver stiffness weight; k2 is the adjustment coefficient 2, β1+k2≤1; Specifically, by calculating the risk score change rate of each data in the biochemical indicator data and imaging data, we can accurately understand the changes in the factors affecting cirrhosis, and set thresholds for each data in each biochemical indicator data and imaging data separately. According to the risk score change rate of each data, its influence weight can be adjusted in real time, so that the system can adapt to the impact of data changes on cirrhosis, automatically adjust the results of the classification evaluation module, and improve the sensitivity and accuracy of the system.

[0022] Furthermore, the risk prediction module evaluation process is as follows: ; Where R all (t + Δt) is the comprehensive risk score at time t + Δt; R sh It is a risk score based on biochemical marker data; ΔR sh is the risk score change rate of the biochemical index data; W1 is the weight of biochemical index data; R yx It is a risk score based on imaging data; ΔR yx is the rate of change of risk score of imaging data; W2 is the image data weight; ΔR sh The process is ΔR sh =(ΔR sh，1 +ΔR sh，2 +ΔR sh，3 ) / 3; ΔR yx The process is ΔR sh =(ΔR yx，1 +ΔR yx，2 ) / 2; Using explainable AI, such as LIME or SHAP, the system can convert patient cirrhosis data analysis results into visual charts, allowing doctors to better understand the patient's health status.

[0023] Setting R allThe risk threshold of (t+Δt) is Z, R all The larger (t+Δt) is, the more severe the patient's liver cirrhosis will be in the future. all When (t+Δt) is less than Z, it means that the patient's liver cirrhosis is at low risk in the future and the existing treatment intensity and examination cycle can be maintained. all When (t+Δt) is greater than Z, it means that the patient's risk of cirrhosis is too high at time t+Δt and is in a high-risk state. Preventive measures need to be taken in advance. By predicting the comprehensive risk score in the future, doctors can identify high-risk patients in advance and strengthen corresponding monitoring and intervention plans, so that they can take corresponding measures according to the risk situation, allocate medical resources more reasonably, avoid the deterioration of the disease, and effectively reduce the incidence of liver cancer in patients with hepatitis B cirrhosis.

[0024] Furthermore, the data preprocessing module includes data cleaning, data standardization and feature selection.

[0025] Specifically, data standardization involves standardizing the data of various biochemical indicators and imaging data, and feature selection involves extracting features of the liver area in the imaging data for subsequent data analysis. The data preprocessing module can significantly improve the quality of patient data and the accuracy of subsequent liver cirrhosis pathology identification.

[0026] Furthermore, the data update module is used to regularly update the patient data and liver cirrhosis standard data. The liver cirrhosis standard data includes the standard data of various influencing factors in the imaging data and biochemical index data. Regularly updating the data can ensure its accuracy and timeliness.

[0027] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The liver cirrhosis pathology recognition system based on artificial intelligence is characterized by: It includes data collection module, data preprocessing module, classification evaluation module, time change module, risk prediction module and data update module; The data acquisition module is used to collect patient data, which includes imaging data and biochemical index data; The classification assessment module assesses the risk of liver cirrhosis based on the imaging data and biochemical index data to obtain a risk score R based on the biochemical index data. sh and the risk score R based on imaging data yx ; The time change module is used to compare the influencing factors in the image data and biochemical index data at different times, obtain the change rate of each influencing factor, and set a threshold for it. When one of the influencing factors exceeds the threshold, its weight is automatically adjusted to ensure the sensitivity and accuracy of the system; The risk prediction module is based on the risk score R based on the biochemical index data. sh and the risk score R based on imaging data yx , and the comprehensive risk score R at time t + Δt is obtained based on the changes all (t+Δt), and set a threshold for it, and judge the patient's liver cirrhosis condition in the future according to the threshold, so that preventive measures can be taken in advance.

2. The liver cirrhosis pathology recognition system based on artificial intelligence according to claim 1 is characterized by: The evaluation process of the classification evaluation module includes evaluating the risk of liver cirrhosis based on biochemical indicator data and imaging data, wherein the evaluation process based on biochemical indicator data is as follows: ； Where R sh is the risk score based on biochemical indicator data, sh represents biochemical indicator data; L1 is the intercept one, which represents the risk base value without the influence of other factors; ALT pat is the patient's trough aminotransferase value; ALT nor is the intraglutathione transaminase threshold; is the weight of intravalley aminotransferase, ranging from 0 to 1; AST pat is the patient's aspartate aminotransferase value; AST nor is the aspartate aminotransferase threshold; α2 is the impact weight of aspartate aminotransferase, ranging from 0 to 1; TBLL pat is the patient's albumin value; TBLL nor is the albumin threshold; α3 is the albumin impact weight, ranging from 0 to 1; Risk score R based on biochemical index data sh The higher the value, the greater the risk of liver cirrhosis. Biochemical index data directly reflects liver function and damage; The evaluation process based on image data is as follows: ； Where R yx It is a risk score based on imaging data; L2 is intercept two; CH pat is the stiffness of the patient's liver; CH nor is the liver stiffness threshold; β1 is the liver stiffness weight; CD pat is the patient's liver volume; CD nor is the liver volume threshold; β2 is the liver volume weight; Risk score R based on imaging data yx Provides liver structure and morphology information, R yx The higher the value, the greater the risk of cirrhosis. By evaluating the risk of cirrhosis based on biochemical indicator data and imaging data, doctors can understand the impact of each factor on the risk of cirrhosis.

3. The liver cirrhosis pathology recognition system based on artificial intelligence according to claim 2 is characterized by: The time change module includes change rate assessment based on biochemical indicator data and image data, wherein the change rate assessment process based on biochemical indicator data is as follows: ； Where ΔR sh,i is the risk score change rate based on the i-th data in the biochemical index data; R sh,i (t) is the risk score of the i-th data in the biochemical index data at the current time; R sh,i (t-Δt) is the risk score of the i-th data in the biochemical index data at time t-Δt; Δt is the time interval in months; The value range of i is 1 to 3. When i is 1, it indicates aspartate aminotransferase, when i is 2, it indicates aspartate aminotransferase, and when i is 3, it indicates albumin. Set ΔR sh,i The change threshold is Y1 sh,i , when ΔR sh,i Greater than Y1 sh,i When , and i is 1, it means that the fluctuation of the risk score change rate of the intra-valley aminotransferase value in the biochemical index data is too large. At this time, the intra-valley aminotransferase weight α1 is adjusted. The adjustment process is as follows: ； Among them, xα1 is the adjusted intravalley aminotransferase weight; k1 is the adjustment coefficient 1, α1+k1≤1; The evaluation process based on the rate of change of image data is as follows: ； Where ΔR yx,n It is the risk score change rate based on the nth data in the imaging data; R yx,n (t) is the risk score of the nth data in the image data at the current time; R yx,n (t-Δt) is the risk score of the nth data in the imaging data at time t-Δt; Δt is the time interval; The value range of n is 1 to 2. When n is 1, it indicates liver stiffness, and when n is 2, it indicates liver volume. yx,n The change threshold 2 is Y2 yx,n , when ΔR yx,n Greater than Y2 yx,n When , and n is 1, it means that the fluctuation of the liver stiffness risk score in the imaging data is too large. At this time, the liver stiffness weight β1 is adjusted. The adjustment process is as follows: ； Where xβ1 is the adjusted internal liver stiffness weight; k2 is the adjustment coefficient 2, β1+k2≤1; By calculating the risk score change rate of each biochemical indicator data and imaging data, we can accurately understand the changes in the factors affecting cirrhosis, set thresholds for each biochemical indicator data and imaging data separately, and adjust the impact weight of each data in real time according to the risk score change rate, so that the system can automatically adapt to the impact of data changes on cirrhosis.

4. The liver cirrhosis pathology recognition system based on artificial intelligence according to claim 3 is characterized by: The risk prediction module evaluation process is as follows: ； Where R all (t + Δt) is the comprehensive risk score at time t + Δt; R sh It is a risk score based on biochemical marker data; ΔR sh is the risk score change rate of the biochemical index data; W1 is the weight of biochemical index data; R yx It is a risk score based on imaging data; ΔR yx is the rate of change of risk score of imaging data; W2 is the image data weight; Setting R all The risk threshold of (t+Δt) is Z, R all The larger (t+Δt) is, the more severe the patient's liver cirrhosis will be in the future. all When (t+Δt) is greater than Z, it means that the patient's risk of cirrhosis is too high at time t+Δt, and preventive measures need to be taken in advance. By predicting the comprehensive risk score in the future, doctors can identify high-risk patients in advance and develop corresponding monitoring and intervention plans.

5. The liver cirrhosis pathology recognition system based on artificial intelligence according to claim 4 is characterized in that: The data preprocessing module includes data cleaning, data standardization and feature selection.

6. The liver cirrhosis pathology recognition system based on artificial intelligence according to claim 1 is characterized by: The patient data includes personal information data, and the personal information data includes the patient's age, gender and medical history.

7. The liver cirrhosis pathology recognition system based on artificial intelligence according to claim 1 is characterized by: The data updating module is used to regularly update the patient data and liver cirrhosis standard data to ensure the accuracy and timeliness of the data.

8. The liver cirrhosis pathology recognition system based on artificial intelligence according to claim 6 is characterized by: It also includes an interactive interface module. Both doctors and patients can log in to the interactive interface module to transmit data to the system, and patients can upload personal information data through the interactive interface module.

Images