A prediction system for liver transplantation

Abstract

The present invention relates to the field of medical information technology, and in particular to a prediction system for post-operative liver transplantation. The system includes a serum creatinine level monitoring module, a liver protein evaluation module, a liver tumor detection module, and a post-operative rehabilitation prediction module, which can obtain the original health data of patients after liver transplantation through a blood oxygen sensor; pre-process the original health data of patients after liver transplantation to obtain the health data of patients after liver transplantation; and use a serum creatinine determination kit to extract the serum creatinine level of the health data of patients after liver transplantation to obtain serum creatinine level data. The present invention uses data processing technology, pattern recognition technology, and machine learning technology to perform a comprehensive analysis of rehabilitation indicators after liver transplantation based on serum creatinine level change trend information, liver protein evaluation data, and liver tumor growth information; and achieves a comprehensive post-operative indicator evaluation and analysis of patients, thereby improving the accuracy of rehabilitation prediction after liver transplantation.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to a prediction system for liver transplantation. Background Art

[0002] In the early days of post-liver transplant prediction, researchers first obtained the patient's medical data, and then predicted the degree of recovery after liver transplantation based on the patient's medical data based on the researchers' experience. With the application of machine learning technology in the field of liver transplantation, it is possible to perform statistical analysis on the patient's medical data, and perform clinical analysis on the patient data obtained through statistics, and then build a post-operative prediction model to predict and analyze the post-liver transplant situation. As a result, post-liver transplant prediction has gradually shifted from traditional manual prediction to automatic prediction. However, existing post-liver transplant prediction technologies often rely on simplified models or fixed physiological parameters, which make it difficult to conduct comprehensive post-operative indicator evaluation of patients, resulting in low accuracy in post-liver transplant recovery prediction, thus affecting the research on post-liver transplant prediction. Summary of the invention

[0003] Based on this, it is necessary to provide a prediction system for post-liver transplantation to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a prediction system for liver transplantation includes the following modules:

[0005] The serum creatinine level monitoring module includes a blood oxygen sensor, a serum creatinine determination kit and a biochemical monitor; the original health data of the patient after liver transplantation is obtained through the blood oxygen sensor; the original health data of the patient after liver transplantation is preprocessed to obtain the health data of the patient after liver transplantation; the serum creatinine level of the patient's health data after liver transplantation is extracted using the serum creatinine determination kit to obtain the serum creatinine level data; the change of the serum creatinine level data is monitored using the biochemical monitor to generate the serum creatinine level change data; the change trend of the serum creatinine level change data is analyzed to obtain the serum creatinine level change trend information;

[0006] A liver protein evaluation module, including a protein detection kit, is used to identify the alpha-fetoprotein level characteristics of the patient's health data after liver transplantation to generate alpha-fetoprotein level characteristics; use the protein detection kit to calculate the postoperative liver protein content based on the alpha-fetoprotein level characteristics to obtain the postoperative liver protein content; perform liver protein evaluation on the postoperative liver protein content to generate liver protein evaluation data;

[0007] The liver tumor detection module includes an image scanner; the image scanner is used to scan the post-liver transplantation morphological image of the patient's post-liver transplantation health data to obtain a liver morphological structure image; the post-liver transplantation volume of the liver morphological structure image is measured to generate post-liver transplantation volume data; the liver tumor feature is identified on the liver morphological structure image according to the post-liver transplantation volume data to obtain liver tumor feature data; the growth of the liver tumor feature data is analyzed to generate liver tumor growth information;

[0008] A postoperative rehabilitation prediction module includes a display screen; the postoperative rehabilitation prediction module is used to measure postoperative rehabilitation indicators after liver transplantation based on serum creatinine level change trend information, liver protein evaluation data and liver tumor growth information, and generate postoperative rehabilitation indicator data after liver transplantation; a postoperative rehabilitation prediction model after liver transplantation is constructed through postoperative rehabilitation indicator data after liver transplantation to obtain a postoperative rehabilitation prediction model after liver transplantation; postoperative rehabilitation prediction is performed on the patient's postoperative health data after liver transplantation based on the postoperative rehabilitation prediction model to generate postoperative rehabilitation prediction data; postoperative rehabilitation prediction data is visualized using a display screen to generate a postoperative rehabilitation report after liver transplantation.

[0009] The present invention monitors the serum creatinine level of patients after liver transplantation through a serum creatinine level monitoring module, and can obtain the original health data of patients after liver transplantation; pre-process the original data so that the pre-processed data can better reflect the health information of patients after liver transplantation; further, by extracting serum creatinine level data, it can monitor the change of serum creatinine level; trend analysis is performed on the change data to clarify the change trend information of serum creatinine level. Through the liver protein evaluation module, the alpha-fetoprotein level characteristics in the health data of patients after liver transplantation are identified, and the alpha-fetoprotein level characteristics can be identified; the postoperative liver protein content is measured for the alpha-fetoprotein level characteristics, so that the postoperative liver protein content is clarified, and then the liver protein is evaluated and analyzed, thereby obtaining liver protein evaluation data; it is helpful to determine the protein synthesis function of the liver in the subsequent process. Through the liver tumor detection module, the morphological image scanning of the patient's post-liver transplantation health data can be performed to obtain the liver morphological structure image; the liver morphological structure image is used to measure the post-liver transplantation volume, and the post-liver transplantation volume data is clearly calculated; then the liver tumor characteristic information is identified, thereby obtaining the liver tumor characteristic data; the growth situation of the liver tumor characteristic data is analyzed, so that the liver tumor growth information is mastered, providing data support for monitoring the health status of the liver. Through the postoperative rehabilitation prediction module, the post-liver transplantation rehabilitation index is determined by comprehensively utilizing the serum creatinine level change trend information, liver protein assessment data and liver tumor growth information, so as to clarify the post-liver transplantation rehabilitation index data; based on the post-liver transplantation rehabilitation index data, a post-liver transplantation rehabilitation prediction model is constructed, so that the post-liver transplantation health data of the patient can be predicted for post-operative rehabilitation, and the post-operative rehabilitation prediction data can be obtained; the post-operative rehabilitation prediction data is visualized, so that the post-liver transplantation rehabilitation report can be visualized, providing researchers with intuitive rehabilitation prediction information. Therefore, the present invention uses data processing technology, pattern recognition technology and machine learning technology to conduct a comprehensive analysis of rehabilitation indicators after liver transplantation based on serum creatinine level change trend information, liver protein assessment data and liver tumor growth information; to achieve a comprehensive postoperative indicator evaluation and analysis of patients, thereby improving the accuracy of rehabilitation prediction after liver transplantation.

[0010] Preferably, the serum creatinine level monitoring module includes the following functions:

[0011] Obtain the original health data of patients after liver transplantation through blood oxygen sensors;

[0012] Performing data standardization on the original health data of patients after liver transplantation to obtain the health data of patients after liver transplantation;

[0013] Extract blood sample data from the patient's post-liver transplant health data to obtain post-operative blood sample data; use a serum creatinine determination kit to perform serum creatinine feature marking on the post-operative blood sample data to generate a serum creatinine marking feature; determine the serum creatinine content of the post-operative blood sample data based on the serum creatinine marking feature to obtain serum creatinine level data;

[0014] Extracting peak and valley values ​​of serum creatinine level data to generate serum creatinine peak and valley values; calculating peak and valley differences of serum creatinine peak and valley values ​​to obtain serum creatinine peak and valley difference data; using a biochemical monitor and based on the serum creatinine peak and valley difference data, identifying changes in serum creatinine level data to generate serum creatinine level change data;

[0015] The serum creatinine level change data were analyzed for change trends to obtain serum creatinine level change trend information.

[0016] The present invention obtains the original health data of patients after liver transplantation to provide basic information for subsequent data analysis; performs data standardization on the original data to eliminate the deviation caused by different data sources and units, thereby improving the comparability and accuracy of the data; further, extracts blood sample data from the health data of patients after liver transplantation to obtain specific postoperative blood sample data; performs serum creatinine feature marking on the blood sample data to help identify and distinguish serum creatinine-related data; performs serum creatinine content measurement on the postoperative blood sample data according to the serum creatinine marking feature to clarify the serum creatinine level data, thereby reflecting the key indicators of the patient's renal function to the serum Extracting peak and valley values ​​of creatinine level data can clarify the peak and valley values ​​of serum creatinine, which helps to identify extreme changes in serum creatinine levels; calculating the peak and valley difference of serum creatinine peak and valley values, and then obtaining the peak and valley difference of serum creatinine, can reveal the maximum fluctuation range of serum creatinine levels; identifying changes in serum creatinine level data based on serum creatinine peak and valley difference data, generating serum creatinine level change data, and recording the dynamic changes of serum creatinine levels in detail; performing change trend analysis on serum creatinine level change data, so as to grasp the change trend information of serum creatinine levels; providing a data basis for monitoring changes in patient renal function, which helps to evaluate the patient's recovery after liver transplantation.

[0017] Preferably, the change trend analysis of serum creatinine level change data includes:

[0018] Extract the change sequence features of the serum creatinine level change data to generate the serum creatinine change sequence features; identify the time features of the serum creatinine change sequence features to obtain the serum creatinine time features;

[0019] Aligning the serum creatinine content time series with the serum creatinine time characteristics and the serum creatinine level data to obtain the serum creatinine content time data;

[0020] Performing content difference fluctuation detection on serum creatinine peak-valley difference data to generate content difference fluctuation data; marking difference fluctuation time points on serum creatinine content time data based on the content difference fluctuation data to obtain content difference fluctuation time data;

[0021] The fluctuation periodicity of the content difference fluctuation time data is determined to obtain the content fluctuation periodicity data; the serum creatinine content trend characteristics are identified on the serum creatinine level change data through the content fluctuation periodicity data to generate the serum creatinine content trend characteristics;

[0022] A serum creatinine level change trend graph is constructed according to the trend characteristics of the serum creatinine content to obtain a serum creatinine level change trend graph; a serum creatinine content image inflection point is identified on the serum creatinine level change trend graph to generate a serum creatinine content change inflection point; a serum creatinine content image slope is identified on the serum creatinine level change trend graph in combination with the serum creatinine content change inflection point to obtain a serum creatinine content change slope;

[0023] The change trend of the serum creatinine level change trend graph is determined based on the serum creatinine content change inflection point and the serum creatinine content change slope to obtain the serum creatinine level change trend information.

[0024] The present invention analyzes serum creatinine level change data, specifically extracts serum creatinine change sequence features, and can describe in detail the changes of serum creatinine levels over time; further, by identifying the time features in the serum creatinine change sequence features, the time characteristics of serum creatinine level changes can be revealed; the serum creatinine time features are aligned with the serum creatinine level data, and the serum creatinine level is accurately associated with the time point of its occurrence, so as to provide time series information for subsequent serum creatinine content trend analysis; the serum creatinine peak-to-valley difference data is subjected to content difference fluctuation detection, so as to reveal the maximum fluctuation range of the serum creatinine level; the serum creatinine content time data is subjected to difference fluctuation time point marking based on the content difference fluctuation data, so as to clarify the specific time point of the serum creatinine level fluctuation; the content difference fluctuation time data is subjected to fluctuation periodicity determination, which is helpful to identify serum The invention discloses a method for determining the periodic characteristics of creatinine level changes; identifying the trend characteristics of serum creatinine content on serum creatinine level change data through periodic data of content fluctuation, so as to present the overall trend of serum creatinine level changes; constructing a serum creatinine level change trend graph based on the trend characteristics of serum creatinine content, which can intuitively show the trend of serum creatinine level changes over time; identifying the inflection point of serum creatinine content image on the serum creatinine level change trend graph, which can clearly identify the inflection point of serum creatinine content change, thereby revealing the key turning point of serum creatinine level change trend; identifying the slope of serum creatinine content image on the serum creatinine level change trend graph in combination with the inflection point of serum creatinine content change, so as to show the rate of change of serum creatinine level; based on the inflection point of serum creatinine content change and the slope of serum creatinine content change, the trend of serum creatinine level change can be determined, and the information on serum creatinine level change trend can be obtained.

[0025] Preferably, the liver protein assessment module includes the following functions:

[0026] Performing alpha-fetoprotein characteristic marking on the patient's health data after liver transplantation to obtain alpha-fetoprotein marking characteristics; extracting alpha-fetoprotein concentration values ​​from the patient's health data after liver transplantation according to the alpha-fetoprotein marking characteristics to generate alpha-fetoprotein concentration values; measuring alpha-fetoprotein levels on the alpha-fetoprotein concentration values ​​to obtain alpha-fetoprotein level characteristics;

[0027] The postoperative liver function type of the patient's health data after liver transplantation is identified to generate postoperative liver function type data; based on the characteristics of alpha-fetoprotein levels, the postoperative liver protein function is extracted from the postoperative liver function type data to obtain postoperative liver protein function data; the liver protein content of the postoperative liver protein function data is measured using a protein detection kit to obtain the postoperative liver protein content;

[0028] The postoperative liver protein content is measured by the total serum protein content to generate the postoperative serum total protein content; the postoperative liver protein content is measured by the albumin content to obtain the postoperative albumin content; the postoperative liver protein content is measured by the globulin content to generate the postoperative globulin content;

[0029] The liver synthetic capacity was evaluated based on the postoperative serum total protein content and the postoperative albumin content to obtain the liver synthetic capacity evaluation data; the liver metabolic function was evaluated based on the liver synthetic capacity evaluation data and the postoperative albumin content to generate the liver metabolic function evaluation data; the liver synthetic capacity evaluation data and the liver metabolic function evaluation data were integrated to obtain the liver protein evaluation data.

[0030] The present invention marks the alpha-fetoprotein characteristics of the patient's health data after liver transplantation, and can identify and track the characteristic data of the alpha-fetoprotein; according to the marking characteristics, the alpha-fetoprotein concentration value is extracted, and the alpha-fetoprotein concentration value is obtained, providing quantitative data for the subsequent alpha-fetoprotein level measurement; the alpha-fetoprotein concentration value is measured to reflect the specific level of alpha-fetoprotein in the patient; further, the postoperative liver function type of the patient's health data after liver transplantation is identified, and the postoperative liver function type data is clarified, which is helpful to understand the overall functional state of the liver; based on the alpha-fetoprotein level characteristics, the postoperative liver protein function is extracted from the postoperative liver function type data, so as to provide a direct indicator of the liver's ability to synthesize protein; the liver protein content of the postoperative liver protein function data is measured, which can reveal the liver's ability to synthesize protein; The postoperative liver protein content is measured by the total serum protein content, which can reflect the total protein level in the serum; at the same time, the albumin content is measured for the postoperative liver protein content to clarify the specific content information of albumin in the serum; the globulin content is measured for the postoperative liver protein content to obtain the information of the globulin level in the serum; the liver synthesis capacity is evaluated based on the postoperative serum total protein content and the postoperative albumin content, which is helpful to evaluate the overall ability of the liver to synthesize protein; the liver metabolic function is evaluated based on the liver synthesis capacity evaluation data and the postoperative albumin content to clarify the metabolic function status of the liver; the liver protein evaluation data is obtained by integrating the liver synthesis capacity evaluation data and the liver metabolic function evaluation data, which can comprehensively reflect the synthesis and metabolic functions of the liver and provide indicator information for evaluating the liver health status of patients after liver transplantation.

[0031] Preferably, the liver tumor detection module includes the following functions:

[0032] Extract medical images after liver transplantation from the patient's health data after liver transplantation to generate medical image data after liver transplantation; perform image contrast enhancement on the medical image data after liver transplantation to obtain enhanced medical image data after liver transplantation;

[0033] The liver transplantation medical image enhancement data is scanned by an image scanner to obtain a liver morphological structure image;

[0034] Slice the liver morphology structure image to generate liver morphology image slice data; calculate the area of ​​the liver morphology image slice data to obtain liver morphology slice area data; accumulate the slice thickness of the liver morphology slice area data to generate volume data after liver transplantation;

[0035] Based on the volume data after liver transplantation, liver tumor characteristics are identified in the liver morphological structure image to obtain liver tumor characteristic data;

[0036] The liver tumor characteristic data is analyzed for growth to generate liver tumor growth information.

[0037] The present invention extracts medical images from the patient's post-liver transplant health data to obtain post-liver transplant medical image data, which can provide basic data for subsequent image analysis; performs image contrast enhancement processing on the post-liver transplant medical image data, which can more clearly display the fine structure of the liver, which is helpful to improve the accuracy of subsequent analysis; further, performs morphological image scanning on the liver transplant medical image enhancement data, intuitively displays the morphology and structure of the liver, and thus provides intuitive image data for the evaluation of liver volume and morphology; performs slicing processing on the liver morphological structure image, which is helpful for a more detailed analysis of the local area of ​​the liver; performs area calculation on the liver morphological image slice data, which can obtain liver morphological slice area data; accumulates the slice thickness of the liver morphological slice area data, thereby obtaining post-liver transplant volume data; accurately calculates the volume of the liver after liver transplantation, which is helpful for subsequent evaluation of the size and functional status of the liver; based on the post-liver transplant volume data, performs liver tumor feature recognition on the liver morphological structure image, which can identify the tumor features existing in the liver; performs growth analysis on the liver tumor feature data, which can clarify the growth dynamics of the liver tumor.

[0038] Preferably, the post-liver transplantation morphological image scanning of the liver transplantation medical image enhancement data includes:

[0039] Performing liver structural feature calibration on the liver transplantation medical image enhancement data to obtain liver structural features; performing liver contour edge detection on the liver transplantation medical image enhancement data according to the liver structural features to obtain liver contour edge information;

[0040] The structural geometric features of the liver structure features are calculated based on the liver contour edge information to obtain the liver postoperative structural geometric features; the liver vascular structure features are extracted from the liver postoperative structural geometric features to generate the liver vascular structure features;

[0041] The liver vascular distribution is identified by the liver vascular structure characteristics, and the liver vascular distribution data is obtained;

[0042] The liver morphological structure image after liver transplantation is obtained by scanning the liver structural features after liver transplantation through an image scanner according to the liver postoperative structural geometric features and liver vascular structural features.

[0043] The present invention calibrates liver structural features of liver transplantation medical image enhancement data, which is helpful to identify and analyze the specific structure of the liver; performs liver contour edge detection on liver transplantation medical image enhancement data according to the liver structural features, so as to clarify the boundary and shape of the liver; performs structural geometric feature calculation on liver structural features based on liver contour edge information, and can quantitatively describe the liver morphology; extracts liver vascular structural features based on liver postoperative structural geometric features, and clarifies the distribution of blood vessels in the liver; identifies the liver vascular distribution based on liver structural features through liver vascular structural features, and can reveal the layout and direction of blood vessels in the liver; performs postoperative morphological image scanning on liver structural features based on liver postoperative structural geometric features and liver vascular structural features, and intuitively displays the morphology and structure of the liver.

[0044] Preferably, the liver tumor feature recognition based on the liver morphological structure image after liver transplantation volume data includes:

[0045] Extract volume values ​​of liver transplantation postoperative volume data to generate liver transplantation postoperative volume values; calculate volume range of liver transplantation postoperative volume values ​​to obtain liver transplantation postoperative volume range data;

[0046] Perform liver morphology abnormality detection on the liver morphology structure image according to the liver postoperative volume range data to obtain liver morphology abnormality data; perform liver surface abnormal protrusion structure feature recognition on the liver morphology structure image based on the liver morphology abnormality data to generate surface abnormal protrusion structure features;

[0047] The structural uplift degree of the abnormal surface uplift structure feature is measured to obtain the structural uplift degree data; the distribution density of the abnormal surface uplift structure feature is calculated to generate the uplift distribution density data;

[0048] Liver tumor markers are identified based on the structural protrusion degree data and the protrusion distribution density data to obtain liver tumor marker data; liver tumor characteristics are identified on the liver morphological structure image based on the liver tumor marker data to obtain liver tumor characteristic data.

[0049] The present invention extracts volume values ​​from volume data after liver transplantation to provide volume quantification data for subsequent volume analysis; performs volume range measurement on liver volume values ​​after surgery to define a reasonable interval of liver volume, providing a basis for evaluating the liver volume range; performs liver morphology abnormality detection on liver morphology structural images based on liver postoperative volume range data, which is helpful to identify whether the liver has volume abnormalities, thereby providing data support for further liver health assessment; performs liver morphology structural image recognition of abnormal liver surface protrusion structural features based on liver morphology abnormality data, thereby clarifying the abnormal surface protrusion structural situation; performs structural protrusion degree measurement on the abnormal surface protrusion structural features, which can quantify the degree of liver surface protrusion; at the same time, performs distribution density calculation on the abnormal surface protrusion structural features, which can reveal the distribution of liver surface protrusions; performs liver tumor marker recognition based on structural protrusion degree data and protrusion distribution density data, which can clearly identify liver tumor markers; performs liver tumor feature recognition on liver morphology structural images based on liver tumor marker data, thereby obtaining liver tumor feature data.

[0050] Preferably, the growth analysis of liver tumor characteristic data includes:

[0051] Performing time series extraction on liver tumor characteristic data to generate liver tumor characteristic time series information; tracking liver tumor characteristic changes on the liver tumor characteristic time series information to obtain liver tumor characteristic change data;

[0052] Performing shape and volume measurement on the liver tumor characteristic change data to obtain liver tumor shape and volume data; quantifying the growth rate of the liver tumor shape and volume data to generate the liver tumor shape growth rate;

[0053] Extracting liver tumor surface features from liver tumor feature change data to obtain liver tumor surface feature data; identifying surface smoothness of liver tumor surface feature data to generate liver tumor surface smoothness;

[0054] According to the smoothness of the liver tumor surface, the liver tumor tissue vascular characteristics are identified on the liver tumor characteristic data to obtain the liver tumor tissue vascular characteristics; the vascular distribution area of ​​the liver tumor tissue vascular characteristics is detected to generate the liver tumor vascular distribution area; based on the liver tumor vascular distribution area, the liver tumor fibrosis degree is determined on the liver tumor characteristic change data to obtain the liver tumor fibrosis degree;

[0055] The growth rate of liver tumor shape, liver tumor surface smoothness and liver tumor fibrosis degree are combined with liver tumor growth conditions to obtain liver tumor growth information.

[0056] The present invention extracts time series from liver tumor characteristic data, which helps to track the changes of liver tumor characteristics over time. Tracking the changes of liver tumor characteristics on these time series information makes it possible to record the dynamic changes of liver tumor characteristics in detail; performing shape and volume measurement on liver tumor characteristic change data, clarifying the volume information of liver tumor, and providing a quantitative basis for evaluating tumor size; quantifying the growth rate of liver tumor shape and volume data, so as to reveal the speed of liver tumor volume change; extracting liver tumor surface characteristics from liver tumor characteristic change data, which can clarify the liver tumor surface characteristics; performing surface smoothness identification on surface characteristic data, so as to reveal the smoothness of liver tumor surface, which helps to analyze the morphological characteristics of tumor; performing liver tumor tissue vascular characteristics identification on liver tumor characteristic data according to the liver tumor surface smoothness, performing vascular distribution area detection on liver tumor tissue vascular characteristics, which can clarify the liver tumor vascular distribution area, and then reveal the distribution of blood vessels in tumor; determining the liver tumor fibrosis degree on liver tumor characteristic change data based on the liver tumor vascular distribution area, so as to evaluate the fibrosis level inside the tumor; merging the liver tumor shape growth rate, liver tumor surface smoothness and liver tumor fibrosis degree, so as to obtain liver tumor growth information.

[0057] Preferably, the postoperative rehabilitation prediction module includes the following functions:

[0058] Determine the post-liver transplant rehabilitation index based on the serum creatinine level change trend information, liver protein assessment data and liver tumor growth information to generate post-liver transplant rehabilitation index data;

[0059] A post-liver transplant rehabilitation prediction model is constructed based on the post-liver transplant rehabilitation index data to obtain a post-liver transplant rehabilitation prediction pre-model; and the post-liver transplant rehabilitation prediction pre-model is trained using the post-liver transplant rehabilitation index data to obtain a post-liver transplant rehabilitation prediction training model;

[0060] Perform model cross-validation evaluation on the post-liver transplant rehabilitation prediction training model to obtain model evaluation data; adjust model parameters of the post-liver transplant rehabilitation prediction training model based on the model evaluation data to obtain the post-liver transplant rehabilitation prediction model;

[0061] Inputting the patient's post-liver transplant health data into a post-liver transplant rehabilitation prediction model to predict the post-liver transplant risk, thereby obtaining post-liver transplant risk prediction data; performing risk degree prediction on the post-liver transplant risk prediction data, thereby obtaining post-liver transplant risk degree prediction data;

[0062] According to the postoperative risk prediction data, the postoperative rehabilitation prediction model is used to predict the patient's health data after liver transplantation, and the postoperative rehabilitation prediction data is generated;

[0063] The postoperative rehabilitation prediction data is visualized and processed in a graphical form using a display screen to obtain a postoperative rehabilitation prediction chart; a visual report is compiled for the postoperative rehabilitation prediction chart to obtain a liver transplant postoperative rehabilitation report.

[0064] The present invention can measure the rehabilitation index after liver transplantation by comprehensively analyzing the trend information of serum creatinine level changes, liver protein evaluation data and liver tumor growth information, and then obtain the rehabilitation index data after liver transplantation; so as to provide key quantitative indicators for evaluating the postoperative rehabilitation status of patients; construct a postoperative rehabilitation prediction model after liver transplantation using the rehabilitation index data after liver transplantation, and provide a basic framework for subsequent rehabilitation prediction; train the postoperative rehabilitation prediction model after liver transplantation, and learn and predict the postoperative rehabilitation trend of patients; perform model cross-validation evaluation on the postoperative rehabilitation prediction training model after liver transplantation, so as to improve the accuracy of the evaluation model; adjust the model parameters of the postoperative rehabilitation prediction training model after liver transplantation through the model evaluation data, and obtain the final postoperative rehabilitation after liver transplantation. The prediction model enables a more accurate prediction of the patient's recovery; inputting the patient's post-liver transplant health data into the post-liver transplant rehabilitation prediction model can predict the post-liver transplant risk, thereby obtaining post-liver transplant risk prediction data; predicting the risk level of the post-liver transplant risk prediction data helps to identify the risk situation faced by the patient after surgery; based on the post-operative risk level prediction data and combined with the post-liver transplant rehabilitation prediction model, the patient's post-liver transplant health data can be used for post-operative rehabilitation prediction, thereby generating post-operative rehabilitation prediction data; visualizing the post-operative rehabilitation prediction data into charts to intuitively display the patient's recovery trend; and compiling a visual report of the post-operative rehabilitation prediction chart to enable researchers to more intuitively understand the post-liver transplant rehabilitation report.

[0065] Preferably, the determination of recovery indicators after liver transplantation based on serum creatinine level change trend information, liver protein assessment data and liver tumor growth information includes:

[0066] The serum creatinine level change trend information is used to track the change in creatinine content to obtain creatinine content change tracking data; the patient's post-liver transplant health data is used to measure the liver synergistic function to obtain liver synergistic function index data based on the creatinine content change tracking data;

[0067] According to the liver protein evaluation data, the liver synthesis function of the patient's post-liver transplantation health data is evaluated to generate a liver synthesis function index; the liver metabolic function of the patient's post-liver transplantation health data is measured by the liver protein evaluation data to obtain a liver metabolic function index; the liver synthesis function index and the liver metabolic function index are integrated to generate liver synthesis-metabolic function index data;

[0068] Based on the liver tumor growth information, the liver tumor abnormality is measured on the patient's health data after liver transplantation to obtain liver tumor abnormality index data;

[0069] The liver synergistic function index data, liver synthesis-metabolic function index data and liver tumor abnormality index data are combined into the post-liver transplantation rehabilitation index data to generate the post-liver transplantation rehabilitation index data.

[0070] The present invention tracks the trend information of serum creatinine level changes, and can accurately reflect the dynamic changes of the patient's serum creatinine level; uses the tracking data to measure the liver synergistic function of the patient's health data after liver transplantation, and clarifies the liver synergistic function action index information; evaluates the liver synthesis function of the patient's health data after liver transplantation based on the liver protein evaluation data, and can reflect the liver's ability to synthesize protein; at the same time, further analysis of the liver protein evaluation data obtains liver metabolic function indicators, and then reveals the functional state of the liver in the metabolic process; based on the liver tumor growth information, the patient's health data after liver transplantation is measured for liver tumor abnormalities, and liver tumor abnormality index data is obtained, which can identify tumor abnormalities existing in the liver; the liver synergistic function index data, the liver synthesis-metabolic function index data and the liver tumor abnormality index data are merged to obtain the post-liver transplantation rehabilitation index data, so as to comprehensively reflect the patient's post-liver transplantation rehabilitation state, and provide a comprehensive index system for evaluating the patient's rehabilitation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A schematic diagram of a process flow for a prediction system after liver transplantation;

[0072] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0073] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0074] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0075] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0076] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0077] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0078] To achieve this, please refer to Figures 1 to 3 , a prediction system for liver transplantation, the system comprising the following modules:

[0079] The serum creatinine level monitoring module includes a blood oxygen sensor, a serum creatinine determination kit and a biochemical monitor; the original health data of the patient after liver transplantation is obtained through the blood oxygen sensor; the original health data of the patient after liver transplantation is preprocessed to obtain the health data of the patient after liver transplantation; the serum creatinine level of the patient's health data after liver transplantation is extracted using the serum creatinine determination kit to obtain the serum creatinine level data; the change of the serum creatinine level data is monitored using the biochemical monitor to generate the serum creatinine level change data; the change trend of the serum creatinine level change data is analyzed to obtain the serum creatinine level change trend information;

[0080] A liver protein evaluation module, including a protein detection kit, is used to identify the alpha-fetoprotein level characteristics of the patient's health data after liver transplantation to generate alpha-fetoprotein level characteristics; use the protein detection kit to calculate the postoperative liver protein content based on the alpha-fetoprotein level characteristics to obtain the postoperative liver protein content; perform liver protein evaluation on the postoperative liver protein content to generate liver protein evaluation data;

[0081] The liver tumor detection module includes an image scanner; the image scanner is used to scan the post-liver transplantation morphological image of the patient's post-liver transplantation health data to obtain a liver morphological structure image; the post-liver transplantation volume of the liver morphological structure image is measured to generate post-liver transplantation volume data; the liver tumor feature is identified on the liver morphological structure image according to the post-liver transplantation volume data to obtain liver tumor feature data; the growth of the liver tumor feature data is analyzed to generate liver tumor growth information;

[0082] A postoperative rehabilitation prediction module includes a display screen; the postoperative rehabilitation prediction module is used to measure postoperative rehabilitation indicators after liver transplantation based on serum creatinine level change trend information, liver protein evaluation data and liver tumor growth information, and generate postoperative rehabilitation indicator data after liver transplantation; a postoperative rehabilitation prediction model after liver transplantation is constructed through postoperative rehabilitation indicator data after liver transplantation to obtain a postoperative rehabilitation prediction model after liver transplantation; postoperative rehabilitation prediction is performed on the patient's postoperative health data after liver transplantation based on the postoperative rehabilitation prediction model to generate postoperative rehabilitation prediction data; postoperative rehabilitation prediction data is visualized using a display screen to generate a postoperative rehabilitation report after liver transplantation.

[0083] The present invention monitors the serum creatinine level of patients after liver transplantation through a serum creatinine level monitoring module, and can obtain the original health data of patients after liver transplantation; pre-process the original data so that the pre-processed data can better reflect the health information of patients after liver transplantation; further, by extracting serum creatinine level data, it can monitor the change of serum creatinine level; trend analysis is performed on the change data to clarify the change trend information of serum creatinine level. Through the liver protein evaluation module, the alpha-fetoprotein level characteristics in the health data of patients after liver transplantation are identified, and the alpha-fetoprotein level characteristics can be identified; the postoperative liver protein content is measured for the alpha-fetoprotein level characteristics, so that the postoperative liver protein content is clarified, and then the liver protein is evaluated and analyzed, thereby obtaining liver protein evaluation data; it is helpful to determine the protein synthesis function of the liver in the subsequent process. Through the liver tumor detection module, the morphological image scanning of the patient's post-liver transplantation health data can be performed to obtain the liver morphological structure image; the liver morphological structure image is used to measure the post-liver transplantation volume, and the post-liver transplantation volume data is clearly calculated; then the liver tumor characteristic information is identified, thereby obtaining the liver tumor characteristic data; the growth situation of the liver tumor characteristic data is analyzed, so that the liver tumor growth information is mastered, providing data support for monitoring the health status of the liver. Through the postoperative rehabilitation prediction module, the post-liver transplantation rehabilitation index is determined by comprehensively utilizing the serum creatinine level change trend information, liver protein assessment data and liver tumor growth information, so as to clarify the post-liver transplantation rehabilitation index data; based on the post-liver transplantation rehabilitation index data, a post-liver transplantation rehabilitation prediction model is constructed, so that the post-liver transplantation health data of the patient can be predicted for post-operative rehabilitation, and the post-operative rehabilitation prediction data can be obtained; the post-operative rehabilitation prediction data is visualized, so that the post-liver transplantation rehabilitation report can be visualized, providing researchers with intuitive rehabilitation prediction information. Therefore, the present invention uses data processing technology, pattern recognition technology and machine learning technology to conduct a comprehensive analysis of rehabilitation indicators after liver transplantation based on serum creatinine level change trend information, liver protein assessment data and liver tumor growth information; to achieve a comprehensive postoperative indicator evaluation and analysis of patients, thereby improving the accuracy of rehabilitation prediction after liver transplantation.

[0084] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a system for predicting liver transplantation after surgery according to the present invention. In this example, the system for predicting liver transplantation after surgery includes the following modules:

[0085] S1: Serum creatinine level monitoring module, including a blood oxygen sensor, a serum creatinine determination kit and a biochemical monitor; obtaining the original health data of the patient after liver transplantation through the blood oxygen sensor; pre-processing the original health data of the patient after liver transplantation to obtain the health data of the patient after liver transplantation; using the serum creatinine determination kit to extract the serum creatinine level of the patient's health data after liver transplantation to obtain the serum creatinine level data; using the biochemical monitor to monitor the change of the serum creatinine level data to generate the serum creatinine level change data; performing a change trend analysis on the serum creatinine level change data to obtain the serum creatinine level change trend information;

[0086] In an embodiment of the present invention, an electronic health record (EHR) system is used to collect the patient's medical information, specifically including various postoperative physiological indicators and laboratory test results, and the patient's medical information is stored to obtain the patient's original health data after liver transplantation; the patient's original health data after liver transplantation is preprocessed, specifically, the patient's original health data after liver transplantation is denoised, missing value processed and normalized to obtain accurate and usable patient's health data after liver transplantation; the feature extraction technology is used to extract the serum creatinine level physiological indicator from the patient's health data after liver transplantation to obtain the serum creatinine level physiological indicator; combined with the serum creatinine level physiological indicator The serum creatinine level result is tested according to the standard to generate the serum creatinine level result; the serum creatinine level data is monitored by time series analysis technology, and the serum creatinine level changes over time are specifically analyzed to identify trends, seasonality and other changes in the data; for example, by collecting the serum creatinine level data of patients at different time points after surgery, the changes in serum creatinine levels are observed; for example, during the postoperative recovery period, the serum creatinine level will show a downward trend; the linear regression technology is used to calculate the rate of change of serum creatinine level over time, and the rate of change of serum creatinine level is visualized through a line graph, so as to obtain the serum creatinine level change trend information.

[0087] S2: Liver protein evaluation module, including a protein detection kit, used to identify the alpha-fetoprotein level characteristics of the patient's health data after liver transplantation and generate alpha-fetoprotein level characteristics; use the protein detection kit to calculate the postoperative liver protein content based on the alpha-fetoprotein level characteristics to obtain the postoperative liver protein content; perform liver protein evaluation on the postoperative liver protein content to generate liver protein evaluation data;

[0088] In an embodiment of the present invention, a feature labeling technology is used to label the alpha-fetoprotein level characteristics of the patient's health data after liver transplantation, and the alpha-fetoprotein level feature labeling information is used to extract the patient's alpha-fetoprotein level characteristics to generate alpha-fetoprotein level characteristics; the alpha-fetoprotein physiological index of the patient is identified on the alpha-fetoprotein level characteristics to obtain the patient's alpha-fetoprotein physiological index; the alpha-fetoprotein content of the patient's alpha-fetoprotein physiological index is measured using the physiological index measurement technology to obtain the alpha-fetoprotein content measurement value; specifically, the iBAQ value in the alpha-fetoprotein content measurement value is extracted to obtain the patient's alpha-fetoprotein expression level; the postoperative liver protein content is calculated based on the patient's alpha-fetoprotein expression to obtain the postoperative liver protein content; the postoperative liver protein content is evaluated for liver protein using the physiological index enrichment analysis technology; for example, the enrichment degree of liver protein is calculated using bioinformatics software to generate liver protein evaluation data.

[0089] S3: liver tumor detection module, including an image scanner; scanning the post-liver transplantation morphological image of the patient's post-liver transplantation health data through the image scanner to obtain a liver morphological structure image; measuring the post-liver transplantation volume of the liver morphological structure image to generate post-liver transplantation volume data; identifying liver tumor features in the liver morphological structure image based on the post-liver transplantation volume data to obtain liver tumor feature data; analyzing the growth of the liver tumor feature data to generate liver tumor growth information;

[0090] In an embodiment of the present invention, ultrasound elastic imaging technology is used to perform morphological image scanning on the liver of a patient after liver transplantation; specifically scanning the structural characteristic information of the liver is helpful to evaluate the health status of the liver; for example, ultrasound elastic imaging technology is used to obtain a patient's liver structural image, which includes the shape of the liver, the size of the liver, and the smoothness of the surface; three-dimensional visualization technology is used to perform volume measurement on the liver morphological structure image, specifically to measure the positional relationship between the postoperative tumor and the surrounding important structures, and to accurately calculate the tumor volume; for example, the three-dimensional visualization technology is used to calculate that the patient's liver volume after liver transplantation is 1200 cubic centimeters m; deep learning technology, especially convolutional neural network (CNN), is used to identify liver tumor characteristics from liver morphological and structural images; for example, a model built based on the ResNet18 algorithm is used to distinguish mild, moderate and severe fatty degeneration grades in ultrasound examination image data, with accuracies of 0.85, 0.91 and 0.93 respectively, thereby obtaining liver tumor characteristic data; multimodal non-enhanced magnetic resonance imaging (MRI) technology is used to analyze liver tumor growth; specifically, non-enhanced MRI technology is used to monitor the morphology, size and surrounding fibrosis of patients' liver tumors, thereby obtaining liver tumor growth information.

[0091] S4: A postoperative rehabilitation prediction module, including a display screen; the postoperative rehabilitation prediction module is used to measure postoperative rehabilitation indicators after liver transplantation according to the trend information of serum creatinine level changes, liver protein assessment data and liver tumor growth information, and generate postoperative rehabilitation indicator data after liver transplantation; a postoperative rehabilitation prediction model after liver transplantation is constructed through the postoperative rehabilitation indicator data after liver transplantation, and a postoperative rehabilitation prediction model after liver transplantation is obtained; postoperative rehabilitation prediction is performed on the patient's postoperative health data after liver transplantation based on the postoperative rehabilitation prediction model, and postoperative rehabilitation prediction data is generated; the postoperative rehabilitation prediction data is visualized using the display screen, and a postoperative rehabilitation report after liver transplantation is generated.

[0092] In an embodiment of the present invention, a multivariate linear regression analysis method is used to measure the rehabilitation index after liver transplantation based on the serum creatinine level change trend information, liver protein evaluation data and liver tumor growth information, and the influence of each indicator on postoperative rehabilitation is specifically calculated, thereby generating the rehabilitation index data after liver transplantation; for example, the patient's rehabilitation index score is 85 points, of which the serum creatinine level change trend contributes 30 points, the liver protein evaluation result contributes 40 points, and the liver tumor growth information contributes 15 points. A machine learning algorithm, such as a random forest algorithm, is used to construct a post-liver transplant rehabilitation prediction model based on rehabilitation index data; the data set is divided into a training set and a test set, and the training set is used to train the model to obtain a post-liver transplant rehabilitation prediction model that can perform rehabilitation prediction based on the input rehabilitation index data; based on the post-liver transplant rehabilitation prediction model, the patient's post-liver transplant health data is predicted, specifically, the patient's serum creatinine level changes, liver protein assessment data, and liver tumor growth information are input, and the model outputs post-operative rehabilitation prediction data; for example, the model predicts that the post-operative recovery probability of patient P001 is 85%; data visualization tools, such as Python's matplotlib library, are used to display the post-operative rehabilitation prediction data in the form of charts, and a post-liver transplant rehabilitation report is generated, which specifically displays the patient's recovery probability and visualization results of related indicators.

[0093] Preferably, the serum creatinine level monitoring module includes the following functions:

[0094] Obtain the original health data of patients after liver transplantation through blood oxygen sensors;

[0095] Performing data standardization on the original health data of patients after liver transplantation to obtain the health data of patients after liver transplantation;

[0096] Extract blood sample data from the patient's post-liver transplant health data to obtain post-operative blood sample data; use a serum creatinine determination kit to perform serum creatinine feature marking on the post-operative blood sample data to generate a serum creatinine marking feature; determine the serum creatinine content of the post-operative blood sample data based on the serum creatinine marking feature to obtain serum creatinine level data;

[0097] Extracting peak and valley values ​​of serum creatinine level data to generate serum creatinine peak and valley values; calculating peak and valley differences of serum creatinine peak and valley values ​​to obtain serum creatinine peak and valley difference data; using a biochemical monitor and based on the serum creatinine peak and valley difference data, identifying changes in serum creatinine level data to generate serum creatinine level change data;

[0098] The serum creatinine level change data were analyzed for change trends to obtain serum creatinine level change trend information.

[0099] In an embodiment of the present invention, the patient's post-liver transplant health data is collected through a hospital information system (HIS), specifically including serum creatinine levels, liver protein assessment results, and liver tumor growth information; the patient's post-liver transplant health raw data is standardized, specifically using the Z-score standardization method to convert the data into a standard normal distribution form with a mean of 0 and a standard deviation of 1; for example, if the patient's serum creatinine level is 120 μmol / L, after Z-score standardization, its standardized value is -1.5 (assuming that the mean value of the indicator is 100 μmol / L and the standard deviation is 20 μmol / L); blood sample data is extracted from the patient's post-liver transplant health data, specifically the serum creatinine level; a database query statement is used to select the patient's serum creatinine level field; and the serum creatinine level data is feature-marked using statistical analysis software; for example, endogenous creatinine clearance is calculated. The serum creatinine level data are obtained by calculating the peak-to-valley value of the serum creatinine level data and using Python's Pandas library for data analysis to determine the peak and valley points by finding the local maximum and minimum values ​​in the data. The peak-to-valley values ​​are calculated by difference calculation. For example, if the patient's serum creatinine level is 150 μmol / L at the peak and 100 μmol / L at the valley, the peak-to-valley difference is 50 μmol / L. Based on the serum creatinine peak-to-valley difference data, the time series analysis method is used to identify the changes in the serum creatinine level data. For example, by analyzing the changes in the patient's serum creatinine level over time, its change trend is identified. The trend analysis of the serum creatinine level change data is performed, and the long-term change trend of the serum creatinine level is analyzed by using linear regression technology.

[0100] Preferably, the change trend analysis of serum creatinine level change data includes:

[0101] Extract the change sequence features of the serum creatinine level change data to generate the serum creatinine change sequence features; identify the time features of the serum creatinine change sequence features to obtain the serum creatinine time features;

[0102] Aligning the serum creatinine content time series with the serum creatinine time characteristics and the serum creatinine level data to obtain the serum creatinine content time data;

[0103] Performing content difference fluctuation detection on serum creatinine peak-valley difference data to generate content difference fluctuation data; marking difference fluctuation time points on serum creatinine content time data based on the content difference fluctuation data to obtain content difference fluctuation time data;

[0104] The fluctuation periodicity of the content difference fluctuation time data is determined to obtain the content fluctuation periodicity data; the serum creatinine content trend characteristics are identified on the serum creatinine level change data through the content fluctuation periodicity data to generate the serum creatinine content trend characteristics;

[0105] A serum creatinine level change trend graph is constructed according to the trend characteristics of the serum creatinine content to obtain a serum creatinine level change trend graph; a serum creatinine content image inflection point is identified on the serum creatinine level change trend graph to generate a serum creatinine content change inflection point; a serum creatinine content image slope is identified on the serum creatinine level change trend graph in combination with the serum creatinine content change inflection point to obtain a serum creatinine content change slope;

[0106] The change trend of the serum creatinine level change trend graph is determined based on the serum creatinine content change inflection point and the serum creatinine content change slope to obtain the serum creatinine level change trend information.

[0107] In an embodiment of the present invention, a time series analysis technique, specifically an autoregressive moving average (ARMA) technique, is used to extract sequence features of serum creatinine level data; for example, ARMA is used to analyze the patient's serum creatinine level time series data to extract features such as trend, seasonality and random components; a clustering algorithm is used to identify the time features of the extracted serum creatinine change sequence features; for example, the patient's serum creatinine change sequence features are divided into different time clusters to identify the creatinine level change patterns in different time periods; the serum creatinine time features are aligned with the serum creatinine level data through database management technology; specifically, an SQL query statement is used to match the patient's serum creatinine time features with the corresponding creatinine level data; a local extreme value detection technique is used to perform content difference fluctuation detection on the serum creatinine peak-to-valley difference data, and the patient's serum creatinine peak-to-valley difference data is analyzed to identify time points with larger fluctuations; the content difference fluctuation data is marked at time points; and through periodic analysis technology, The fluctuation time data of the difference in serum creatinine content of the patient is analyzed by Fourier transform to determine its fluctuation periodicity; the trend analysis technology is used to identify the trend characteristics of serum creatinine content for the content fluctuation periodicity data; for example, the serum creatinine content fluctuation periodicity data of the patient is analyzed by linear regression to obtain the trend characteristics of serum creatinine content; data visualization technology is used, such as Python's matplotlib library, to construct a serum creatinine level change trend graph according to the serum creatinine content trend characteristics; image processing technology, such as the Canny edge detection algorithm, is used to identify the inflection point of the serum creatinine content image for the serum creatinine level change trend graph; slope calculation technology, such as numerical differentiation, is used to identify the slope of the serum creatinine content image for the inflection point of the serum creatinine content change, specifically, the slope of the patient's serum creatinine content change is calculated by numerical differentiation; the change trend of the serum creatinine level change trend graph is determined based on the serum creatinine content change inflection point and the serum creatinine content change slope, and finally the serum creatinine level change trend information is obtained.

[0108] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Schematic diagram of the functional flow of the liver protein evaluation module. In this embodiment, the functions of the liver protein evaluation module include:

[0109] S21: performing alpha-fetoprotein feature marking on the patient's post-liver transplant health data to obtain an alpha-fetoprotein marking feature; performing alpha-fetoprotein concentration value extraction on the patient's post-liver transplant health data according to the alpha-fetoprotein marking feature to generate an alpha-fetoprotein concentration value; performing alpha-fetoprotein level measurement on the alpha-fetoprotein concentration value to obtain an alpha-fetoprotein level feature;

[0110] S22: Identify the postoperative liver function type of the patient's health data after liver transplantation to generate postoperative liver function type data; extract the postoperative liver protein function of the postoperative liver function type data based on the alpha-fetoprotein level characteristics to obtain postoperative liver protein function data; use a protein detection kit to measure the liver protein content of the postoperative liver protein function data to obtain the postoperative liver protein content;

[0111] S23: the postoperative liver protein content is measured by the serum total protein content to generate the postoperative serum total protein content; the postoperative liver protein content is measured by the albumin content to obtain the postoperative albumin content; the postoperative liver protein content is measured by the globulin content to generate the postoperative globulin content;

[0112] S24: Evaluate the liver synthesis capacity based on the postoperative serum total protein content and the postoperative albumin content to obtain liver synthesis capacity evaluation data; evaluate the liver metabolic function based on the liver synthesis capacity evaluation data and the postoperative albumin content to generate liver metabolic function evaluation data; integrate the liver synthesis capacity evaluation data and the liver metabolic function evaluation data to obtain liver protein evaluation data.

[0113] In an embodiment of the present invention, feature labeling technology is used to label the health data of patients after liver transplantation with alpha-fetoprotein features; the health data of patients after liver transplantation are analyzed according to the alpha-fetoprotein labeling features; and the alpha-fetoprotein concentration value is extracted using a laboratory information system (LIS), for example, the alpha-fetoprotein concentration value of the patient obtained from the LIS is 20 ng / mL; the alpha-fetoprotein level is measured using luminescent immunoassay technology, for example, the alpha-fetoprotein level of the patient is determined to be 20 ng / mL, and the alpha-fetoprotein level feature is generated; and the classification algorithm in machine learning is used to classify the patient's liver transplantation Postoperative liver function type identification is performed based on postoperative health data after transplantation; for example, the liver function type is identified as "normal" or "abnormal"; based on the characteristics of alpha-fetoprotein levels, postoperative liver protein function is extracted from postoperative liver function type data using logistic regression analysis; serum total protein content is measured for postoperative liver protein content, and the serum total protein content in postoperative liver protein content is specifically measured; albumin content is measured for postoperative liver protein content, and the albumin content in postoperative liver protein content is specifically measured; globulin content is measured for postoperative liver protein content, and the globulin content in postoperative liver protein content is specifically measured;

[0114] The liver synthetic capacity was evaluated based on the postoperative serum total protein content and the postoperative albumin content to obtain the liver synthetic capacity evaluation data; the liver metabolic function was evaluated based on the liver synthetic capacity evaluation data and the postoperative albumin content to generate the liver metabolic function evaluation data; the liver synthetic capacity evaluation data and the liver metabolic function evaluation data were integrated to obtain the liver protein evaluation data.

[0115] Preferably, the liver tumor detection module includes the following functions:

[0116] Extract medical images after liver transplantation from the patient's health data after liver transplantation to generate medical image data after liver transplantation; perform image contrast enhancement on the medical image data after liver transplantation to obtain enhanced medical image data after liver transplantation;

[0117] The liver transplantation medical image enhancement data is scanned by an image scanner to obtain a liver morphological structure image;

[0118] Slice the liver morphology structure image to generate liver morphology image slice data; calculate the area of ​​the liver morphology image slice data to obtain liver morphology slice area data; accumulate the slice thickness of the liver morphology slice area data to generate volume data after liver transplantation;

[0119] Based on the volume data after liver transplantation, liver tumor characteristics are identified in the liver morphological structure image to obtain liver tumor characteristic data;

[0120] The liver tumor characteristic data is analyzed for growth to generate liver tumor growth information.

[0121] In an embodiment of the present invention, medical image data of patients after liver transplantation are extracted from a hospital information system using medical image archiving and communication system (PACS) technology; for example, CT or MRI image data of patients after surgery are obtained from the PACS system, thereby obtaining a medical image data set after liver transplantation; an MRI enhancement method based on visual attention is used to enhance the medical image data of liver transplantation by adaptive contrast adjustment and image fusion; for example, for the MRI image of the patient, an adaptive attenuation weight matrix is ​​used to introduce the image of the preserved illumination into the model of the attenuation image as compensation, and finally generate enhanced medical image data of liver transplantation; an ultrasound examination technology is used to perform morphological image scanning on the enhanced medical image data to obtain a liver morphological structure image; a computer-aided ( The method comprises the following steps: using the image segmentation technology in the CAD system to slice the liver morphological structure image; specifically, segmenting the patient's liver morphological structure image into multiple slices to obtain liver morphological image slice data; calculating the patient's volume data after liver transplantation based on the liver morphological image slice area data and by accumulating the thickness of each slice; specifically, multiplying the patient's liver morphological slice area data with the corresponding slice thickness and accumulating them to generate the volume data after liver transplantation; utilizing the feature recognition technology to identify liver tumor features of the liver morphological structure image based on the volume data after liver transplantation to obtain liver tumor feature data; utilizing the time series analysis technology to analyze the growth of the liver tumor feature data; specifically, generating liver tumor growth information by analyzing the changes of the patient's liver tumor feature data over time.

[0122] Preferably, the post-liver transplantation morphological image scanning of the liver transplantation medical image enhancement data includes:

[0123] Performing liver structural feature calibration on the liver transplantation medical image enhancement data to obtain liver structural features; performing liver contour edge detection on the liver transplantation medical image enhancement data according to the liver structural features to obtain liver contour edge information;

[0124] The structural geometric features of the liver structure features are calculated based on the liver contour edge information to obtain the liver postoperative structural geometric features; the liver vascular structure features are extracted from the liver postoperative structural geometric features to generate the liver vascular structure features;

[0125] The liver vascular distribution is identified by the liver vascular structure characteristics, and the liver vascular distribution data is obtained;

[0126] The liver morphological structure image after liver transplantation is obtained by scanning the liver structural features after liver transplantation through an image scanner according to the liver postoperative structural geometric features and liver vascular structural features.

[0127] In an embodiment of the present invention, 3D image processing technology is used to calibrate liver structural features of liver transplantation medical image enhancement data to obtain liver structural feature data; edge detection technology is used to detect liver contour edges of liver transplantation medical image enhancement data to obtain liver contour edge information; based on liver contour edge information, image segmentation technology in a computer-aided (CAD) system is used to calculate structural geometric features of liver structural features to obtain liver postoperative structural geometric features; grayscale correction, noise filtering, and enhancement sharpening are specifically performed on liver transplantation medical image enhancement data; image processing technology is used to extract liver vascular structural features based on liver postoperative structural geometric features, specifically three-dimensional segmentation is performed on CT images of liver vessels, thereby obtaining liver vascular structural features; based on liver vascular structural features, image processing technology is used to identify liver vascular distribution of liver structural features, specifically by matching key points in multiple images to identify vascular distribution, thereby obtaining liver vascular distribution data; based on liver postoperative structural geometric features and liver vascular structural features, ultrasound examination technology is used to scan liver structural features for postoperative morphological images of liver structural features to obtain liver morphological structural images, specifically detecting the characteristics of the liver or vascular surface.

[0128] Preferably, the liver tumor feature recognition based on the liver morphological structure image after liver transplantation volume data includes:

[0129] Extract volume values ​​of liver transplantation postoperative volume data to generate liver transplantation postoperative volume values; calculate volume range of liver transplantation postoperative volume values ​​to obtain liver transplantation postoperative volume range data;

[0130] Perform liver morphology abnormality detection on the liver morphology structure image according to the liver postoperative volume range data to obtain liver morphology abnormality data; perform liver surface abnormal protrusion structure feature recognition on the liver morphology structure image based on the liver morphology abnormality data to generate surface abnormal protrusion structure features;

[0131] The structural uplift degree of the abnormal surface uplift structure feature is measured to obtain the structural uplift degree data; the distribution density of the abnormal surface uplift structure feature is calculated to generate the uplift distribution density data;

[0132] Liver tumor markers are identified based on the structural protrusion degree data and the protrusion distribution density data to obtain liver tumor marker data; liver tumor characteristics are identified on the liver morphological structure image based on the liver tumor marker data to obtain liver tumor characteristic data.

[0133] In an embodiment of the present invention, computed tomography (CT) or magnetic resonance imaging (MRI) technology is used to extract volume values ​​of the patient's volume data after liver transplantation; the volume range of the liver's postoperative volume value is measured to obtain the liver's postoperative volume range data;

[0134] By using abnormality detection technology, the postoperative liver volume value is compared with the preset normal liver volume range; the preset normal liver volume range is specifically set to 900 to 1500 cubic centimeters, and the volume difference between the two is compared to determine whether there is an abnormality. If the patient's postoperative liver volume value exceeds the normal range, it will be marked as abnormal liver morphology data; based on the abnormal liver morphology data, computed tomography (CT) is used to identify the structural features of abnormal protrusions on the liver surface; for example, the smoothness of the surface of the patient's liver morphology structure image is scanned; for the non-smooth area, the abnormal protrusion is identified, thereby obtaining the structural features of the abnormal surface protrusion; for the structural features of the abnormal surface protrusion, the edge detection algorithm in the image processing technology, such as the Canny algorithm, is used. Method for determining the degree of structural protrusion; for example, processing the abnormal protrusion area of ​​the patient by the Canny algorithm, specifically measuring its protrusion degree, and obtaining the data of the degree of structural protrusion; using image analysis software, calculating the distribution density of the surface abnormal protrusion structure characteristics; specifically calculating the density of the abnormal protrusion structure of the patient according to its distribution in the liver; based on the data of the degree of structural protrusion and the data of the protrusion distribution density, using statistical methods and feature recognition technology, identifying liver tumor markers; specifically identifying liver tumor marker data, including human carbonyl reductase, Golgi phosphoprotein and tumor-specific growth factor, etc., to obtain liver tumor marker data; performing liver tumor feature recognition on the liver morphological structure image according to the liver tumor marker data to obtain liver tumor feature data.

[0135] Preferably, the growth analysis of liver tumor characteristic data includes:

[0136] Performing time series extraction on liver tumor characteristic data to generate liver tumor characteristic time series information; tracking liver tumor characteristic changes on the liver tumor characteristic time series information to obtain liver tumor characteristic change data;

[0137] Performing shape and volume measurement on the liver tumor characteristic change data to obtain liver tumor shape and volume data; quantifying the growth rate of the liver tumor shape and volume data to generate the liver tumor shape growth rate;

[0138] Extracting liver tumor surface features from liver tumor feature change data to obtain liver tumor surface feature data; identifying surface smoothness of liver tumor surface feature data to generate liver tumor surface smoothness;

[0139] According to the smoothness of the liver tumor surface, the liver tumor tissue vascular characteristics are identified on the liver tumor characteristic data to obtain the liver tumor tissue vascular characteristics; the vascular distribution area of ​​the liver tumor tissue vascular characteristics is detected to generate the liver tumor vascular distribution area; based on the liver tumor vascular distribution area, the liver tumor fibrosis degree is determined on the liver tumor characteristic change data to obtain the liver tumor fibrosis degree;

[0140] The growth rate of liver tumor shape, liver tumor surface smoothness and liver tumor fibrosis degree are combined with liver tumor growth conditions to obtain liver tumor growth information.

[0141] In the embodiment of the present invention, feature extraction technology is used to extract features from the patient's CT image; specifically, time series information of features such as tumor size, shape and texture heterogeneity is extracted from the patient's image data; time series analysis technology is used to track the changes of the extracted liver tumor features; for example, liver tumor feature change data is obtained by analyzing the changes of the patient's tumor features over time; shape and volume of the liver tumor feature change data are measured to obtain liver tumor shape and volume data; growth rate of the liver tumor shape and volume data is quantified, specifically by analyzing the changes of the patient's tumor volume over time, quantifying and recording its growth rate, thereby obtaining the liver tumor shape growth rate; image processing technology, such as wavelet transform, is used to Extract the surface features of liver tumors; for example, analyze the surface of the patient's tumor through wavelet transform to extract surface feature data; use image analysis software to identify the surface smoothness of the liver tumor surface feature data; specifically identify the smoothness of the patient's liver tumor surface; use imaging genomics features and feature recognition technology to identify the vascular features of liver tumor tissue; for example, by analyzing the vascular information of the patient's liver tumor tissue, the vascular features of liver tumor tissue can be identified; use dynamic contrast enhanced MRI (DCE-MRI) technology to detect the vascular distribution area of ​​the vascular features of liver tumor tissue, and specifically detect the vascular distribution area of ​​the patient's liver tumor; use the Ishak scoring system to evaluate the degree of liver tumor fibrosis. The Ishak scoring system divides liver fibrosis assessment into 0 to 6 stages. Specifically, according to the Ishak scoring system, the images of the vascular distribution area of ​​the patient's liver tumor are analyzed to determine the stage of liver fibrosis, thereby obtaining the degree of liver tumor fibrosis; the growth rate of liver tumor shape, the smoothness of the liver tumor surface and the degree of liver tumor fibrosis are combined and analyzed to obtain liver tumor growth information.

[0142] As an example of the present invention, refer to Figure 3 As shown, Figure 1 The function flow diagram of the postoperative rehabilitation prediction module is as follows. In this embodiment, the functions of the postoperative rehabilitation prediction module include:

[0143] S41: Determine the post-liver transplant rehabilitation index based on the serum creatinine level change trend information, liver protein assessment data and liver tumor growth information to generate post-liver transplant rehabilitation index data;

[0144] S42: constructing a post-liver transplant rehabilitation prediction model based on the post-liver transplant rehabilitation index data to obtain a post-liver transplant rehabilitation prediction pre-model; and training the post-liver transplant rehabilitation prediction pre-model using the post-liver transplant rehabilitation index data to obtain a post-liver transplant rehabilitation prediction training model;

[0145] S43: performing a model cross-validation evaluation on the post-liver transplant rehabilitation prediction training model to obtain model evaluation data; adjusting model parameters of the post-liver transplant rehabilitation prediction training model using the model evaluation data to obtain a post-liver transplant rehabilitation prediction model;

[0146] S44: inputting the patient's post-liver transplant health data into a post-liver transplant rehabilitation prediction model to perform post-liver transplant risk prediction, thereby obtaining post-liver transplant risk prediction data; performing risk degree prediction on the post-liver transplant risk prediction data, thereby obtaining post-liver transplant risk degree prediction data;

[0147] S45: Based on the postoperative risk level prediction data, postoperative rehabilitation prediction is performed on the patient's postoperative health data after liver transplantation in combination with the postoperative rehabilitation prediction model after liver transplantation to generate postoperative rehabilitation prediction data;

[0148] S46: Visualize and graph the postoperative rehabilitation prediction data using a display screen to obtain a postoperative rehabilitation prediction graph; and compile a visualized report of the postoperative rehabilitation prediction graph to obtain a liver transplant postoperative rehabilitation report.

[0149] In an embodiment of the present invention, a statistical analysis method is used to comprehensively analyze the post-liver transplant rehabilitation index information of serum creatinine level change trend information, liver protein evaluation data and liver tumor growth information, so as to obtain post-liver transplant rehabilitation index data; a random forest (RF) algorithm is used to construct a post-liver transplant rehabilitation prediction pre-model, specifically, a random forest model containing 10 decision trees is constructed, each decision tree uses bootstrap samples during training, and the number of features randomly selected at each split node is half of the total features; the patient's historical rehabilitation index data is used to train the pre-model, specifically, the data set is divided into a feature set and a label, the feature set includes serum creatinine level, liver protein evaluation results and liver tumor growth, and the label is the rehabilitation result. The model was trained using the gradient boosting machine (GBM) algorithm, with a learning rate of 0.1, the number of trees of 100, and a maximum depth of 5, with the loss function being to minimize the mean square error. The trained model was evaluated using a 5-fold cross-validation, with the data set divided into five parts, one of which was set aside each time as a test set, and the rest as a training set. The accuracy, precision, recall, and F1 score of the model on the test set were calculated to obtain model evaluation data. The model parameters were adjusted based on the results of the cross-validation. For example, if the model's accuracy was low, the number of trees was increased to 200, or the learning rate was adjusted to 0.05 to improve the generalization ability of the model. After the parameter adjustment, cross-validation was performed again to determine the final model parameters. For example, the final GBM model parameters were a learning rate of 0.05, a number of trees of 200, and a maximum depth of 5. A post-operative rehabilitation prediction model for liver transplantation is obtained; the patient's post-operative health data after liver transplantation is input into the post-operative rehabilitation prediction model for liver transplantation risk prediction, and the post-operative risk prediction data for liver transplantation is obtained; the risk level of the post-operative risk prediction data for liver transplantation is predicted through logistic regression analysis, and the severity of the risk after liver transplantation is specifically analyzed to obtain the post-operative risk level prediction data; based on the post-operative risk level prediction data and in combination with the post-operative rehabilitation prediction model, the patient's post-operative rehabilitation data for liver transplantation is predicted, and the post-operative rehabilitation prediction data is generated; the post-operative rehabilitation prediction data is visualized and graphed, specifically using Python's matplotlib library to obtain a post-operative rehabilitation prediction chart; and Tableau software is used to compile a visual report in combination with the chart, and finally the post-operative rehabilitation report for liver transplantation is obtained.

[0150] Preferably, the determination of recovery indicators after liver transplantation based on serum creatinine level change trend information, liver protein assessment data and liver tumor growth information includes:

[0151] The serum creatinine level change trend information is used to track the change in creatinine content to obtain creatinine content change tracking data; the patient's post-liver transplant health data is used to measure the liver synergistic function to obtain liver synergistic function index data based on the creatinine content change tracking data;

[0152] According to the liver protein evaluation data, the liver synthesis function of the patient's post-liver transplantation health data is evaluated to generate a liver synthesis function index; the liver metabolic function of the patient's post-liver transplantation health data is measured by the liver protein evaluation data to obtain a liver metabolic function index; the liver synthesis function index and the liver metabolic function index are integrated to generate liver synthesis-metabolic function index data;

[0153] Based on the liver tumor growth information, the liver tumor abnormality is measured on the patient's health data after liver transplantation to obtain liver tumor abnormality index data;

[0154] The liver synergistic function index data, liver synthesis-metabolic function index data and liver tumor abnormality index data are combined into the post-liver transplantation rehabilitation index data to generate the post-liver transplantation rehabilitation index data.

[0155] In an embodiment of the present invention, trend monitoring technology is used to track the change of creatinine content based on the trend information of serum creatinine level, and obtain creatinine content change tracking data; based on the creatinine content change tracking data, combined with MELD scoring technology, the synergistic function of the patient's liver is evaluated; the specific MELD score includes three parameters: creatinine, bilirubin and international normalized ratio (INR), and the liver synergistic function index data is calculated; the synthetic function of the patient's liver is evaluated by the measurement data of serum total protein (TP) and serum albumin (ALB); for example, the patient's TP is 70g / L and ALB is 35g / L, and the data of the two contents are weighted to obtain the liver synthetic function index; urea nitrogen is measured by a fully automatic biochemical analyzer and uric acid index content, so as to evaluate the metabolic function of the patient's liver; for example, the patient's urea nitrogen is 20mg / dL and uric acid is 6mg / dL, indicating that the metabolic function is normal, and the liver metabolic function index is obtained; the liver synthesis function index and the liver metabolic function index are integrated to form the liver synthesis-metabolic function index data; based on the liver tumor growth information, the patient's tumor abnormality after liver transplantation is measured; for example, if the patient's liver tumor growth appears to be in a bulging state, the abnormal growth of the liver transplantation tumor is specifically identified, and the liver tumor abnormality index data is obtained; using data fusion technology, the liver synergistic function index data, the liver synthesis-metabolic function index data and the liver tumor abnormality index data are merged to form the rehabilitation index data after liver transplantation.

[0156] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0157] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A prediction system for liver transplantation, characterized in that: Includes the following modules: The serum creatinine level monitoring module includes a blood oxygen sensor, a serum creatinine determination kit and a biochemical monitor; the original health data of the patient after liver transplantation is obtained through the blood oxygen sensor; the original health data of the patient after liver transplantation is preprocessed to obtain the health data of the patient after liver transplantation; the serum creatinine level of the patient's health data after liver transplantation is extracted using the serum creatinine determination kit to obtain the serum creatinine level data; the change of the serum creatinine level data is monitored using the biochemical monitor to generate the serum creatinine level change data; the change trend of the serum creatinine level change data is analyzed to obtain the serum creatinine level change trend information; A liver protein evaluation module, including a protein detection kit, is used to identify the alpha-fetoprotein level characteristics of the patient's health data after liver transplantation to generate alpha-fetoprotein level characteristics; use the protein detection kit to calculate the postoperative liver protein content based on the alpha-fetoprotein level characteristics to obtain the postoperative liver protein content; perform liver protein evaluation on the postoperative liver protein content to generate liver protein evaluation data; The liver tumor detection module includes an image scanner; the image scanner is used to scan the post-liver transplantation morphological image of the patient's post-liver transplantation health data to obtain a liver morphological structure image; the post-liver transplantation volume of the liver morphological structure image is measured to generate post-liver transplantation volume data; the liver tumor feature is identified on the liver morphological structure image according to the post-liver transplantation volume data to obtain liver tumor feature data; the growth of the liver tumor feature data is analyzed to generate liver tumor growth information; A postoperative rehabilitation prediction module includes a display screen; the postoperative rehabilitation prediction module is used to measure postoperative rehabilitation indicators after liver transplantation based on serum creatinine level change trend information, liver protein evaluation data and liver tumor growth information, and generate postoperative rehabilitation indicator data after liver transplantation; a postoperative rehabilitation prediction model after liver transplantation is constructed through postoperative rehabilitation indicator data after liver transplantation to obtain a postoperative rehabilitation prediction model after liver transplantation; postoperative rehabilitation prediction is performed on the patient's postoperative health data after liver transplantation based on the postoperative rehabilitation prediction model to generate postoperative rehabilitation prediction data; postoperative rehabilitation prediction data is visualized using a display screen to generate a postoperative rehabilitation report after liver transplantation.

2. The system for predicting postoperative liver transplantation according to claim 1, characterized in that: The serum creatinine level monitoring module includes the following functions: Obtain the original health data of patients after liver transplantation through blood oxygen sensors; Performing data standardization on the original health data of patients after liver transplantation to obtain the health data of patients after liver transplantation; Extract blood sample data from the patient's post-liver transplant health data to obtain post-operative blood sample data; use a serum creatinine determination kit to perform serum creatinine feature marking on the post-operative blood sample data to generate a serum creatinine marking feature; determine the serum creatinine content of the post-operative blood sample data based on the serum creatinine marking feature to obtain serum creatinine level data; Extracting the peak and valley values ​​of serum creatinine level data to generate serum creatinine peak and valley values; calculating the peak and valley difference of serum creatinine peak and valley values ​​to obtain serum creatinine peak and valley difference data; Using a biochemical monitor, the serum creatinine level data is identified for changes based on the serum creatinine peak-to-valley difference data to generate serum creatinine level change data; The serum creatinine level change data were analyzed for change trends to obtain serum creatinine level change trend information.

3. The system for predicting postoperative liver transplantation according to claim 2, characterized in that: The change trend analysis of serum creatinine level change data includes: Extract the change sequence features of the serum creatinine level change data to generate the serum creatinine change sequence features; identify the time features of the serum creatinine change sequence features to obtain the serum creatinine time features; Aligning the serum creatinine content time series with the serum creatinine time characteristics and the serum creatinine level data to obtain the serum creatinine content time data; Performing content difference fluctuation detection on serum creatinine peak-valley difference data to generate content difference fluctuation data; marking difference fluctuation time points on serum creatinine content time data based on the content difference fluctuation data to obtain content difference fluctuation time data; The fluctuation periodicity of the content difference fluctuation time data is determined to obtain the content fluctuation periodicity data; the serum creatinine content trend characteristics are identified on the serum creatinine level change data through the content fluctuation periodicity data to generate the serum creatinine content trend characteristics; A serum creatinine level change trend graph is constructed according to the trend characteristics of the serum creatinine content to obtain a serum creatinine level change trend graph; a serum creatinine content image inflection point is identified on the serum creatinine level change trend graph to generate a serum creatinine content change inflection point; a serum creatinine content image slope is identified on the serum creatinine level change trend graph in combination with the serum creatinine content change inflection point to obtain a serum creatinine content change slope; The change trend of the serum creatinine level change trend graph is determined based on the serum creatinine content change inflection point and the serum creatinine content change slope to obtain the serum creatinine level change trend information.

4. The system for predicting postoperative liver transplantation according to claim 1, characterized in that: The liver protein assessment module includes the following functions: Performing alpha-fetoprotein characteristic marking on the patient's health data after liver transplantation to obtain alpha-fetoprotein marking characteristics; extracting alpha-fetoprotein concentration values ​​from the patient's health data after liver transplantation according to the alpha-fetoprotein marking characteristics to generate alpha-fetoprotein concentration values; measuring alpha-fetoprotein levels on the alpha-fetoprotein concentration values ​​to obtain alpha-fetoprotein level characteristics; The postoperative liver function type of the patient's health data after liver transplantation is identified to generate postoperative liver function type data; based on the characteristics of alpha-fetoprotein levels, the postoperative liver protein function is extracted from the postoperative liver function type data to obtain postoperative liver protein function data; the liver protein content of the postoperative liver protein function data is measured using a protein detection kit to obtain the postoperative liver protein content; The postoperative liver protein content is measured by the total serum protein content to generate the postoperative serum total protein content; the postoperative liver protein content is measured by the albumin content to obtain the postoperative albumin content; the postoperative liver protein content is measured by the globulin content to generate the postoperative globulin content; The liver synthetic capacity was evaluated based on the postoperative serum total protein content and the postoperative albumin content to obtain the liver synthetic capacity evaluation data; the liver metabolic function was evaluated based on the liver synthetic capacity evaluation data and the postoperative albumin content to generate the liver metabolic function evaluation data; the liver synthetic capacity evaluation data and the liver metabolic function evaluation data were integrated to obtain the liver protein evaluation data.

5. The system for predicting postoperative liver transplantation according to claim 1, characterized in that: The liver tumor detection module includes the following functions: Extract medical images after liver transplantation from the patient's health data after liver transplantation to generate medical image data after liver transplantation; perform image contrast enhancement on the medical image data after liver transplantation to obtain enhanced medical image data after liver transplantation; The liver transplantation medical image enhancement data is scanned by an image scanner to obtain a liver morphological structure image; Slicing the liver morphology structure image to generate liver morphology image slice data; performing area calculation on the liver morphology image slice data to obtain liver morphology slice area data; The liver morphology slice area data were accumulated by slice thickness to generate the volume data after liver transplantation; Based on the volume data after liver transplantation, liver tumor characteristics are identified in the liver morphological structure image to obtain liver tumor characteristic data; The liver tumor characteristic data is analyzed for growth to generate liver tumor growth information.

6. The system for predicting postoperative liver transplantation according to claim 5, characterized in that: The method of performing post-liver transplantation morphological image scanning on the liver transplantation medical image enhancement data includes: Performing liver structural feature calibration on the liver transplantation medical image enhancement data to obtain the liver structural features; performing liver contour edge detection on the liver transplantation medical image enhancement data according to the liver structural features to obtain liver contour edge information; The structural geometric features of the liver structure features are calculated based on the liver contour edge information to obtain the liver postoperative structural geometric features; the liver vascular structure features are extracted from the liver postoperative structural geometric features to generate the liver vascular structure features; The liver vascular distribution is identified by the liver vascular structure characteristics, and the liver vascular distribution data is obtained; The liver morphological structure image after liver transplantation is obtained by scanning the liver structural features after liver transplantation through an image scanner according to the liver postoperative structural geometric features and liver vascular structural features.

7. The system for predicting postoperative liver transplantation according to claim 5, characterized in that: The method of identifying liver tumor features on liver morphological structure images based on volume data after liver transplantation includes: Extract volume values ​​of liver transplantation postoperative volume data to generate liver transplantation postoperative volume values; calculate volume range of liver transplantation postoperative volume values ​​to obtain liver transplantation postoperative volume range data; Perform liver morphology abnormality detection on the liver morphology structure image according to the liver postoperative volume range data to obtain liver morphology abnormality data; perform liver surface abnormal protrusion structure feature recognition on the liver morphology structure image based on the liver morphology abnormality data to generate surface abnormal protrusion structure features; The structural uplift degree of the abnormal surface uplift structure feature is measured to obtain the structural uplift degree data; the distribution density of the abnormal surface uplift structure feature is calculated to generate the uplift distribution density data; Liver tumor markers are identified based on the structural protrusion degree data and the protrusion distribution density data to obtain liver tumor marker data; liver tumor characteristics are identified on the liver morphological structure image based on the liver tumor marker data to obtain liver tumor characteristic data.

8. The system for predicting postoperative liver transplantation according to claim 5, characterized in that: The growth analysis of liver tumor characteristic data includes: Performing time series extraction on liver tumor characteristic data to generate liver tumor characteristic time series information; tracking liver tumor characteristic changes on the liver tumor characteristic time series information to obtain liver tumor characteristic change data; Performing shape and volume measurement on the liver tumor characteristic change data to obtain liver tumor shape and volume data; quantifying the growth rate of the liver tumor shape and volume data to generate the liver tumor shape growth rate; Extracting liver tumor surface features from liver tumor feature change data to obtain liver tumor surface feature data; identifying surface smoothness of liver tumor surface feature data to generate liver tumor surface smoothness; According to the smoothness of the liver tumor surface, the liver tumor tissue vascular characteristics are identified on the liver tumor characteristic data to obtain the liver tumor tissue vascular characteristics; the vascular distribution area of ​​the liver tumor tissue vascular characteristics is detected to generate the liver tumor vascular distribution area; based on the liver tumor vascular distribution area, the liver tumor fibrosis degree is determined on the liver tumor characteristic change data to obtain the liver tumor fibrosis degree; The growth rate of liver tumor shape, liver tumor surface smoothness and liver tumor fibrosis degree are combined with liver tumor growth conditions to obtain liver tumor growth information.

9. The system for predicting postoperative liver transplantation according to claim 1, characterized in that: The postoperative rehabilitation prediction module includes the following functions: Determine the post-liver transplant rehabilitation index based on the serum creatinine level change trend information, liver protein assessment data and liver tumor growth information to generate post-liver transplant rehabilitation index data; A post-liver transplant rehabilitation prediction model is constructed based on the post-liver transplant rehabilitation index data to obtain a post-liver transplant rehabilitation prediction pre-model; and the post-liver transplant rehabilitation prediction pre-model is trained using the post-liver transplant rehabilitation index data to obtain a post-liver transplant rehabilitation prediction training model; Perform model cross-validation evaluation on the post-liver transplant rehabilitation prediction training model to obtain model evaluation data; adjust model parameters of the post-liver transplant rehabilitation prediction training model based on the model evaluation data to obtain the post-liver transplant rehabilitation prediction model; The patient's post-liver transplant health data is input into a post-liver transplant rehabilitation prediction model to predict the post-liver transplant risk, thereby obtaining post-liver transplant risk prediction data; Predicting the risk level of the risk prediction data after liver transplantation to obtain the risk level prediction data after transplantation; According to the postoperative risk prediction data, the postoperative rehabilitation prediction model is used to predict the patient's health data after liver transplantation, and the postoperative rehabilitation prediction data is generated; The postoperative rehabilitation prediction data is visualized and processed in a graphical form using a display screen to obtain a postoperative rehabilitation prediction chart; a visual report is compiled for the postoperative rehabilitation prediction chart to obtain a liver transplant postoperative rehabilitation report.

10. The system for predicting postoperative liver transplantation according to claim 9, characterized in that: The determination of recovery indicators after liver transplantation based on serum creatinine level change trend information, liver protein assessment data and liver tumor growth information includes: The serum creatinine level change trend information is used to track the change in creatinine content to obtain creatinine content change tracking data; the patient's post-liver transplant health data is used to measure the liver synergistic function to obtain liver synergistic function index data based on the creatinine content change tracking data; According to the liver protein evaluation data, the liver synthesis function of the patient's post-liver transplantation health data is evaluated to generate a liver synthesis function index; the liver metabolic function of the patient's post-liver transplantation health data is measured by the liver protein evaluation data to obtain a liver metabolic function index; the liver synthesis function index and the liver metabolic function index are integrated to generate liver synthesis-metabolic function index data; Based on the liver tumor growth information, the liver tumor abnormality is measured on the patient's health data after liver transplantation to obtain liver tumor abnormality index data; The liver synergistic function index data, liver synthesis-metabolic function index data and liver tumor abnormality index data are combined into the post-liver transplantation rehabilitation index data to generate the post-liver transplantation rehabilitation index data.

Images