Artificial intelligence-based liver cancer incidence rule analysis and early prediction method

Abstract

This invention relates to the field of medical data processing technology, specifically to an artificial intelligence-based method for analyzing the incidence patterns and early prediction of liver cancer. The method acquires basic information and examination indicators of the subjects; classifies the subjects into feature clusters based on the basic information; analyzes the disease conditions of historical patients within the feature cluster to which the subject belongs, and the correlation between the subject and historical patients, to determine the subject's disease prediction parameters; when the disease prediction parameters meet monitoring conditions, a preset prediction method is used to predict the risk of the subject, yielding risk prediction parameters. This invention improves the accuracy of liver cancer prediction for subjects.

Description

Technical Field

[0001] This invention relates to the field of medical data processing technology, specifically to an artificial intelligence-based method for analyzing the incidence patterns of liver cancer and for early prediction. Background Technology

[0002] Liver cancer is one of the most common malignant tumors, with a high incidence and mortality rate. Because liver cancer often has an insidious onset and no obvious symptoms in the early stages, most patients are diagnosed at an advanced stage, missing the best treatment opportunity and resulting in a low 5-year survival rate. Therefore, in-depth analysis of the pathogenesis of liver cancer and early prediction are of great significance.

[0003] The prediction of liver cancer usually relies mainly on the doctor's experience in judging the patient's clinical characteristics, laboratory indicators and imaging results. However, this traditional experience judgment is affected by the doctor's clinical experience and the limitations of a single indicator. The occurrence and development of liver cancer is affected by multiple factors, including genetic factors, lifestyle, and history of liver disease. It is difficult to fully and accurately capture the complex interrelationships between different factors and their comprehensive impact on the onset of liver cancer by relying on a single indicator. It is also difficult to reflect the characteristics of liver cancer lesions reflected by changes in different indicators, and it affects the accuracy and reliability of early prediction of liver cancer in patients. Summary of the Invention

[0004] To address the technical problem of low accuracy in early prediction of liver cancer in patients, the present invention aims to provide a method for analyzing the pathogenesis and early prediction of liver cancer based on artificial intelligence. The specific technical solution adopted is as follows: In a first aspect, embodiments of the present invention provide a method for analyzing the incidence patterns and early prediction of liver cancer based on artificial intelligence, the method comprising: Obtain basic information and examination indicators of the individuals to be tested; Based on basic information, the individuals to be tested are divided into feature clusters; the condition of historical patients in the feature cluster to which the individual to be tested belongs and the correlation between the individual to be tested and historical patients are analyzed to determine the predictive parameters of the individual's condition. When the disease prediction parameters meet the monitoring conditions, the risk prediction parameters are obtained by predicting the risk of the person to be tested using a preset prediction method. The preset prediction method is as follows: analyze the progression characteristics of the test indicators before and after the threshold of the test subject's examination indicators to determine the lesion progression parameters; adjust the acquisition time of the examination indicators through the lesion progression parameters, acquire the test subject's examination indicator sequence in segments, analyze the upward trend of adjacent examination indicator sequences to obtain the degree of hepatocyte damage; combine the lesion progression parameters, the degree of hepatocyte damage and the disease prediction parameters to determine the risk prediction parameters of the test subject.

[0005] Furthermore, the analysis of the patient's condition in the feature cluster to which the test subject belongs, the correlation between the test subject and historical patients, and the determination of the test subject's condition prediction parameters include: Acquire medical images of the person to be tested; the examination indicators include: AFP value; The distribution of tumor features and the magnitude of AFP values ​​in the medical images were analyzed to determine the severity index of the patient's condition. Based on the severity indicators of historical patients in the characteristic cluster to which the test subject belongs, and the correlation between the test subject and historical patients, the predictive parameters of the test subject's condition are determined.

[0006] Furthermore, the analysis of the distribution of tumor features and the magnitude of AFP values ​​in the medical images to determine the severity index of the patient's condition includes: Obtain the average diameter and number of tumors in medical images; The severity index of the patient's condition is determined based on the average diameter, the number of tumors, and the real-time AFP value of the patient; the average diameter, the number of tumors, and the real-time AFP value of the patient are all positively correlated with the severity index.

[0007] Furthermore, the monitoring condition is that the disease prediction parameter is greater than a preset risk threshold.

[0008] Furthermore, the step of analyzing the progression characteristics of the examination indicators of the subjects before and after the threshold is exceeded, and determining the parameters of lesion progression, includes: The progression characteristics of AFP sequences before and after threshold breakthrough in the subjects were analyzed separately to determine the pre- and post-progression parameters of AFP values; the lesion progression parameters of the subjects were determined by combining the pre- and post-progression parameters.

[0009] Furthermore, the progression characteristics of AFP sequences after threshold breach in the subjects were analyzed to determine the subsequent progression parameters of AFP values, including: Obtain all AFP values ​​in the AFP sequence that are greater than a preset threshold, and use them as the target AFP values; Determine the frequency and maximum consecutive length of the target AFP value; Obtain the length of the time series sequence after the first occurrence of the target AFP value, and use it as the length of the sequence after the breakthrough. The proportion of the maximum continuous length to the length of the sequence after the breakthrough is used as the breakthrough progress value; Analyze the overall magnitude of the target AFP value exceeding the preset threshold to determine the reference value after the breakthrough. By combining the number of occurrences of the target AFP value, the breakthrough progress value, and the reference value after the breakthrough, the subsequent progress parameters of the AFP value are obtained.

[0010] Furthermore, the progression characteristics of the AFP sequence before threshold breach in the subjects were analyzed to determine the early progression parameters of the AFP value, including: Obtain the length of the time series sequence before the first occurrence of the target AFP value, and use it as the length of the pre-breakthrough sequence; Calculate the average difference between all adjacent AFP values ​​in the time series before the target AFP value first appears, and use it as a reference value before the breakthrough. Based on the pre-breakthrough sequence length and the pre-breakthrough reference value, the preceding progression parameters of the AFP value are obtained; wherein, the pre-breakthrough sequence length is negatively correlated with the preceding progression parameters, and the pre-breakthrough reference value is positively correlated with the preceding progression parameters.

[0011] Furthermore, by adjusting the acquisition time of the examination indicators using the lesion progression parameters, acquiring the examination indicator sequence of the subject in segments, and analyzing the upward trend of adjacent examination indicator sequences to obtain the degree of hepatocellular damage, the following steps are taken: The inspection indicators include: ALT value and AST value; The duration of liver function index acquisition is adjusted using the aforementioned disease progression parameters to obtain the adjustment duration. Based on the aforementioned adjustment duration, the ALT and AST sequences of the subjects were obtained in segments, and the upward trend of the ALT and AST sequences was analyzed to determine the degree of hepatocyte damage.

[0012] Furthermore, the analysis of the ascending trends of ALT and AST sequences yields the degree of hepatocyte damage, including: Analyze the relative upward trend of ALT values ​​in the ALT sequence to obtain the first upward trend; By analyzing the relative upward trend of AST values ​​in the AST sequence, a second upward trend is obtained; By comparing the real-time ALT and real-time AST values ​​of the subjects, the relative changes in the indicators can be obtained; The degree of hepatocyte damage is obtained by combining the first upward trend, the second upward trend, and the relative change of the indicator; wherein the first upward trend, the second upward trend, and the relative change of the indicator are all positively correlated with the degree of hepatocyte damage.

[0013] Furthermore, the determination of risk prediction parameters for the test subject by combining the disease progression parameters, the degree of hepatocyte damage, and the disease prediction parameters includes: Based on the disease progression parameters and the degree of hepatocyte damage, predictive adjustment weights are determined; based on the predictive adjustment weights, the disease prediction parameters are weighted to obtain risk prediction parameters.

[0014] Secondly, an artificial intelligence-based system for analyzing the incidence patterns and early prediction of liver cancer is provided. This system includes the following modules: The information acquisition module is used to acquire the basic information and examination indicators of the person to be tested; The preliminary prediction module is used to classify the test subjects into feature clusters based on basic information; analyze the condition of historical patients in the feature cluster to which the test subject belongs, and the correlation between the test subject and historical patients, to determine the condition prediction parameters of the test subject; The risk prediction module is used to predict the risk of individuals under test by using a preset prediction method when the disease prediction parameters meet the monitoring conditions. The preset prediction method is as follows: analyze the progression characteristics of the test indicators before and after the threshold of the test subject's examination indicators to determine the lesion progression parameters; adjust the acquisition time of the examination indicators through the lesion progression parameters, acquire the test subject's examination indicator sequence in segments, analyze the upward trend of adjacent examination indicator sequences to obtain the degree of hepatocyte damage; combine the lesion progression parameters, the degree of hepatocyte damage and the disease prediction parameters to determine the risk prediction parameters of the test subject.

[0015] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the various possible implementations of the first aspect.

[0016] Fourthly, embodiments of the present invention provide a computer program product comprising: computer program code, which, when executed on a computer, causes the computer to perform the method described in the first aspect or any possible implementation thereof.

[0017] Fifthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the various possible implementations of the first aspect.

[0018] The embodiments of the present invention have at least the following beneficial effects: This invention integrates multi-source data, including basic information and examination indicators of historical patients, as well as the changing characteristics of multidimensional indicators continuously monitored in the subjects, to quantify the liver cancer lesion status reflected by each indicator. First, preliminary determination is made to obtain disease prediction parameters for the subjects. Then, when the disease prediction parameters meet the monitoring conditions, a preset prediction method is used to predict the risk of the subjects, obtaining risk prediction parameters. The preset prediction method analyzes the threshold exceedance status of the subjects and the changing trends of various examination indicators to obtain the risk prediction parameters for the subjects. This invention improves the accuracy of liver cancer prediction for the subjects, providing strong support for early prevention. Attached Figure Description

[0019] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating an artificial intelligence-based method for analyzing the pathogenesis and early prediction of liver cancer, provided in one embodiment of the present invention. Figure 2 This is a flowchart of a preset prediction method provided in one embodiment of the present invention; Figure 3 This is a system block diagram of an artificial intelligence-based liver cancer incidence pattern analysis and early prediction system provided in one embodiment of the present invention. Detailed Implementation

[0021] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the artificial intelligence-based liver cancer pathogenesis analysis and early prediction method proposed in this invention.

[0022] In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0023] In the description of the embodiments of the present invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" means two or more.

[0024] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0026] The embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0027] The following description, in conjunction with the accompanying drawings, details the specific scheme of the artificial intelligence-based liver cancer incidence pattern analysis and early prediction method provided by this invention.

[0028] Please see Figure 1 The diagram illustrates a flowchart of a method for analyzing and predicting the pathogenesis of liver cancer based on artificial intelligence, according to an embodiment of the present invention. The method includes the following steps: Step S100: Obtain the basic information and examination indicators of the person to be tested.

[0029] Obtain basic information about the person to be tested, including: age, gender, name, whether they drink alcohol, whether they smoke, years of drinking alcohol, years of smoking alcohol, frequency of smoking alcohol, frequency of drinking alcohol, and medical images; the medical images include: liver ultrasound, CT, MRI and other examination images and imaging reports; the images include the diameter and number of tumors.

[0030] Age: The older you are, the higher the risk of liver cancer, and the level of liver cancer increases accordingly; 30 years and below is classified as Level I; 31-45 years is classified as Level II; 46-60 years is classified as Level III; and 61 years and above is classified as Level IV.

[0031] Smoking habits: The longer the smoking duration and the higher the frequency, the greater the potential damage to the liver, and the higher the level of severity. The threshold is set as follows: No smoking = 0; Occasional smoking = Level I (smoking duration less than 3 years, and less than 4 times per week); Light smoking = Level II (smoking duration 3-10 years, or 5-7 times per week); Moderate smoking = Level III (smoking duration 11-20 years, or 7-15 times per week); Heavy smoking = Level IV (smoking duration > 20 years, and more than 16 times per week). Drinking habits: Similar to smoking, the duration and frequency of drinking are key considerations, along with the average daily amount of alcohol consumed. The following categories are defined: No alcohol consumption (0); Occasional drinking (Level I: <5 years of drinking, <2 times per week, <30 grams per day); Light drinking (Level II: 5-10 years of drinking, or 2-4 times per week, 30-50 grams per day); Moderate drinking (Level III: 11-20 years of drinking, or 5-7 times per week, 51-80 grams per day); Heavy drinking (Level IV: >20 years of drinking, >7 times per week, >80 grams per day).

[0032] Image recognition algorithms are used to analyze the medical images of each subject to be tested, and the size and number of tumor areas at the time of diagnosis are extracted. Generally speaking, the larger the diameter of the tumor, the more serious the condition. At the same time, the number of tumors is also important. Multiple tumors are more complex than single tumors. Multiple tumor lesions may accelerate the damage to liver function and increase the possibility of metastasis. Obtain the test indicators of the individuals to be tested, including: alpha-fetoprotein (AFP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and carcinoembryonic antigen (CEA).

[0033] In addition to the size and number of tumor areas in imaging, relevant laboratory test indicators are also key to assessing the severity of liver cancer in patients. Among them, alpha-fetoprotein (AFP) is an important marker for the diagnosis and assessment of liver cancer. Generally, the higher the AFP level, the more severe the liver cancer may be.

[0034] As a preferred embodiment of the present invention, relevant medical history information of the person to be tested can also be obtained, including: the time of hepatitis B infection and treatment, the time of hepatitis C infection and treatment, the time of diagnosis of cirrhosis and the degree of cirrhosis.

[0035] Liver disease history: No history of liver disease is 0; previous history of ordinary hepatitis is Grade I (duration of illness less than 1 year, cured and without complications); chronic hepatitis B or C infection is Grade II (low viral load, basically normal liver function, duration of illness 1-5 years); chronic hepatitis B or C infection is Grade III (high viral load, abnormal liver function, duration of illness 5-10 years); cirrhosis is Grade IV.

[0036] Basic information, examination results, and relevant medical history of the individuals to be tested can be obtained from the hospital's electronic medical record system and laboratory. Information on lifestyle habits can be obtained through questionnaires completed by the patients or their families. It should be noted that obtaining information from the electronic medical record system and laboratory is authorized by the users and does not violate any relevant regulations.

[0037] Based on the classification of different factors, and in the order of age, smoking status, alcohol consumption, and history of liver disease, a characteristic sequence for each patient is determined. For example, a subject who is a moderate smoker, moderate drinker, and has no history of liver disease might have a characteristic sequence represented as follows: .

[0038] Step S200: Based on the basic information, the test subjects are divided into feature clusters; the condition of historical patients in the feature cluster to which the test subjects belong and the fit between the test subjects and historical patients are analyzed to determine the condition prediction parameters of the test subjects.

[0039] Age, history of liver disease, and smoking / drinking history are all important risk factors for liver cancer. Older adults experience a decline in bodily functions, reducing the liver's ability to repair damage. Patients with a history of liver disease already have underlying liver lesions, increasing their risk of liver cancer. Smoking and drinking further damage liver cells, inducing liver cancer. However, some liver cancer patients may not be affected by these high-risk factors; instead, other factors may contribute to the development of liver cancer, such as genetic, metabolic, and autoimmune factors. Therefore, classifying individuals based on their baseline information helps analyze the characteristics of liver cancer incidence in different risk groups.

[0040] Based on the patients' basic information, lifestyle habits, and past medical history, the individual characteristics of all liver cancer patients in the database were first determined, and then all liver cancer patients were classified accordingly: First, the distribution of tumor features and AFP values ​​in the medical images are analyzed to determine the severity index of the patient's condition. Specifically: Obtain the average diameter and number of tumors in medical images.

[0041] The disease severity index of the test subjects was determined based on the average diameter, the number of tumors, and the real-time AFP value of the test subjects. The average diameter, the number of tumors, and the real-time AFP value of the test subjects were all positively correlated with the disease severity index.

[0042] In some embodiments, the disease severity index at the time of diagnosis for each patient being tested... It can be represented as: Where a represents the average diameter of the tumor extracted from the patient's medical images; b represents the number of tumors; and c represents the real-time AFP value detected in the patient being tested. It reflects the severity of the disease as shown in medical imaging; when the diameter and number of tumors in the patient's medical imaging are larger, and the AFP level in the laboratory test is also higher, it indicates that the disease is more severe at the time of diagnosis, and the corresponding disease severity index A value is larger.

[0043] The latest patients are those without a confirmed liver cancer tumor, but with chronic liver disease or cirrhosis. Liver cancer often begins with chronic liver disease, such as chronic hepatitis B, chronic hepatitis C, alcoholic liver disease, or non-alcoholic fatty liver disease. These diseases cause the liver to be in a state of chronic inflammation, triggering hepatocyte damage and repair processes. Further progression of chronic liver disease can lead to liver fibrosis, eventually resulting in cirrhosis. In cirrhosis, the liver structure is destroyed, forming pseudolobules, and hepatocyte regeneration is abnormal, providing a pathological basis for the development of liver cancer. Therefore, the following analysis needs to be conducted with reference to historically diagnosed patients.

[0044] Secondly, based on the severity indicators of historical patients in the characteristic cluster to which the test subject belongs, and the fit between the test subject and historical patients, the predictive parameters for the test subject's condition are determined: For the individuals to be tested, they are categorized into feature clusters based on their basic information: First, determine the characteristic sequence of the person to be tested. Then, cluster the characteristic sequence of the person to be tested with the characteristic sequences of all historical patients in the historical database to determine the cluster to which the person to be tested belongs. This cluster can be denoted as the characteristic cluster.

[0045] Calculate the Euclidean distance between the test subject and the feature sequences of each historical patient in its feature cluster. This Euclidean distance is used to represent the clustering distance between the test subject and the feature sequences of other historical patients in the feature cluster. The smaller the clustering distance, the more concentrated the data distribution in the feature cluster and the better the clustering quality.

[0046] The inverse of the cluster distance between the latest patient and other feature sequences within the cluster is used as the weight of each feature sequence within the cluster; The formula for calculating the predictive parameter A1 of the patient's condition is as follows: Where q represents the number of historical patients contained in the feature cluster to which the person being tested belongs. This represents the Euclidean distance between the person being tested and the feature sequence of the i-th historical patient in the feature cluster; This represents the severity index of the i-th historical patient within the feature cluster to which the person being tested belongs.

[0047] The smaller the cluster distance between the person being tested and the historical patients within their respective feature clusters, and the larger the severity index of the historical patients' conditions, the larger the condition prediction parameter A1 for the person being tested will be.

[0048] Step S300: When the disease prediction parameters meet the monitoring conditions, the risk prediction parameters are obtained by predicting the individuals to be tested using a preset prediction method.

[0049] The monitoring condition is that the disease prediction parameter is greater than a preset risk threshold. In this embodiment of the invention, the preset risk threshold is 0.6. In other specifications, the implementer may adjust this value according to the actual situation. When the disease prediction parameter of the subject is greater than 0.6, it indicates that the subject has a high risk of liver cancer based on historical data and needs further comprehensive examination and close monitoring. When the disease prediction parameter is less than or equal to 0.6, it indicates that the patient's risk of liver cancer may be low, or has not yet seriously affected liver function and surrounding tissues, and the possibility of metastasis is small. In this case, guidance on healthy lifestyle can be provided, and regular check-ups can be conducted.

[0050] When the disease prediction parameter is greater than 0.6, the prediction of the test subject is carried out by the preset prediction method.

[0051] The preset prediction method is as follows: analyze the progression characteristics of the test indicators before and after the threshold of the test subject's examination indicators to determine the lesion progression parameters; adjust the acquisition time of the examination indicators through the lesion progression parameters, acquire the test subject's examination indicator sequence in segments, analyze the upward trend of adjacent examination indicator sequences to obtain the degree of hepatocyte damage; combine the lesion progression parameters, the degree of hepatocyte damage and the disease prediction parameters to determine the risk prediction parameters of the test subject.

[0052] In some embodiments, the preset prediction method can be implemented through the following steps S310 to S330. (See also...) Figure 2 , Figure 2 Flowchart of the preset prediction method: Step S310: Analyze the progression characteristics of the examination indicators of the test subjects before and after the threshold is exceeded, and determine the disease progression parameters.

[0053] Using the severity of liver cancer in patients with similar histories as a predictive reference for the disease in prospective patients is flawed. Histories only provide a preliminary reference and may overlook individual-specific factors such as genetics and individual immune function. Furthermore, predictions based on histories are merely grouped and predicted based on characteristic sequences, failing to consider the dynamic nature of liver disease itself. Therefore, further analysis combining changes in relevant indicators from multiple tests of prospective patients is needed for early prediction of liver cancer incidence. Continuous monitoring of tumor markers was conducted on the subjects every two months; an AFP sequence was obtained in chronological order, consisting of the most recent test and multiple historical AFP values. The progression characteristics of AFP sequences before and after threshold breakthrough in the subjects were analyzed to determine the initial and subsequent progression parameters of AFP values.

[0054] First, the progression characteristics of the AFP sequence after threshold breakthrough in the subjects were analyzed to determine the subsequent progression parameters of the AFP value.

[0055] All AFP values ​​in the AFP sequence that are greater than a preset threshold are obtained and used as the target AFP value. In this invention, the preset threshold is 400.

[0056] If the AFP level in the sequence consistently rises in each test, this is a significant warning sign of liver cancer in the individual being tested. Furthermore, when AFP exceeds 400 μg / L, and other factors that can cause elevated AFP levels, such as pregnancy and germ cell tumors, are excluded, the correlation with liver cancer significantly increases. If AFP remains at a high level or continues to rise after exceeding this critical threshold, it indicates a more severe liver cancer condition. Values ​​greater than 400 μg / L in the AFP sequence are identified as target AFP values. The first target AFP value appearing in the time series is designated as the first target AFP value. The difference between all target AFP values ​​and the preset threshold (400 μg / L) is determined. Other AFP values ​​less than 400 μg / L are considered non-target values.

[0057] Determine the frequency and maximum consecutive length of the target AFP value: Obtain the length of the time series after the first appearance of the target AFP, as the post-breakthrough sequence length; take the proportion of the maximum continuous length in the post-breakthrough sequence length as the breakthrough progress value; count the number of times the target AFP value appears, denoted as d; If there are no non-target values ​​between two adjacent target AFP values, then the two adjacent target AFP values ​​are considered to be continuous. The maximum continuous length of all target AFP values ​​after the first target AFP value is determined and denoted as e. Analyze the overall magnitude of the target AFP value exceeding the preset breakthrough threshold to determine the reference value after breakthrough. The method for obtaining the reference value after breakthrough is to calculate the average of the differences between all target AFP values ​​and the preset breakthrough threshold, and use this as the reference value after breakthrough for the AFP sequence.

[0058] By combining the number of occurrences of the target AFP value, the breakthrough progress value, and the reference value after the breakthrough, the subsequent progress parameters of the AFP value are obtained.

[0059] In some embodiments, the subsequent progression parameters after the subject's AFP threshold is exceeded. The calculation formula is: Where e0 represents the length of the sequence after the breakthrough; This is a reference value for AFP sequences after a breakout.

[0060] The more continuous the target AFP values ​​are after the first target AFP value, and the greater the difference between the target AFP value and the preset breakthrough threshold, the more it indicates that the AFP value continues to increase after breaking the threshold. The larger the value of the subsequent progression parameter B, the more severe the liver cancer lesion in the subject.

[0061] Before AFP exceeds the threshold, it shows a continuous and significant increase. The earlier the threshold is exceeded, the faster the AFP changes, and the more severe the liver cancer lesions in the tested individuals.

[0062] Furthermore, the progression characteristics of the AFP sequence of the subjects before the threshold breakthrough were analyzed to determine the early progression parameters of the AFP value.

[0063] The segment preceding the first target AFP value in the AFP sequence is designated as the pre-breakthrough sequence, and its length is determined and denoted as the pre-breakthrough sequence length g. A smaller pre-breakthrough sequence length g indicates that the subject's AFP threshold was reached earlier and the lesion was more severe. In other words, the length of the time sequence preceding the first appearance of the target AFP value is used as the pre-breakthrough sequence length.

[0064] Since the time interval between each check is consistent, the average difference between any two adjacent values ​​in the pre-breakout sequence is calculated and denoted as the pre-breakout reference value f1. In other words, the average difference between all adjacent AFP values ​​in the time series before the target AFP value first appears is calculated and used as the pre-breakout reference value.

[0065] The pre-breakthrough reference value is used to indicate changes in AFP levels. The higher the pre-breakthrough reference value, the higher the AFP level in each test, indicating that liver cancer cells may be proliferating, leading to the secretion of more AFP, suggesting that the lesion is progressing.

[0066] Based on the pre-breakthrough sequence length and the pre-breakthrough reference value, the preceding progression parameters of the AFP value are obtained; wherein, the pre-breakthrough sequence length is negatively correlated with the preceding progression parameters, and the pre-breakthrough reference value is positively correlated with the preceding progression parameters.

[0067] In some embodiments, the frontal progression parameter of AFP values The calculation formula is: Among them, the larger the reference value f1 before the breakthrough, the more the AFP value is continuously rising before the threshold is broken. The earlier the threshold is broken, the higher the activity of tumor cells, indicating faster liver cancer lesions, and the larger the corresponding C value.

[0068] Finally, by combining the early and late progression parameters, the disease progression parameters of the subjects were determined: The proportions of the lengths of the pre-breakthrough and post-breakthrough sequences relative to the total length of the AFP sequence are used as weights for the pre-breakthrough and post-breakthrough progression parameters, respectively, to obtain the lesion progression parameter P reflected by the AFP value. Wherein, AFP sequence represents the length of the entire sequence; g is the length of the pre-breakthrough sequence; C is the anterior progression parameter; B is the posterior progression parameter; and sorm is the positive correlation mapping function. In this embodiment of the invention, the numerical values ​​of the lesion progression parameters are positively correlated and mapped to the interval [-0.5, 0.5]. The positive correlation mapping function can be obtained through... In this implementation, norm is the normalization function.

[0069] When the AFP value obtained above reflects the disease progression parameter P, the larger the disease progression parameter is, in addition to the fact that liver cancer is a factor that causes the disease progression parameter to be large, it may also be due to benign liver diseases, such as liver abscess, viral hepatitis, and other drug factors, which can also cause an increase in AFP.

[0070] In addition to tumor markers, liver function indicators are also important indicators for evaluating liver cancer lesions. However, if the detection interval of tumor markers is too long, some liver cancer patients may not have elevated tumor markers, or there may not be obvious changes in the early stages of liver cancer. Liver cancer lesions are characterized by dynamic changes, especially at certain stages where they may progress rapidly. If the detection interval of tumor markers is too long, the critical period of disease development may be missed. Therefore, it is also necessary to combine changes in liver function indicators within a short time interval with tumor markers to cross-reference and more accurately assess the liver cancer lesion.

[0071] Step S320: Using the lesion progression parameters, adjust the acquisition time of the examination indicators, acquire the examination indicator sequence of the test subject in segments, analyze the upward trend of adjacent examination indicator sequences, and obtain the degree of hepatocyte damage.

[0072] Liver function indicators are tested every two weeks; the preset time for obtaining liver function indicators is one month prior to the current latest time.

[0073] Based on the magnitude of the disease progression parameters obtained from the AFP value analysis, the preset time for acquiring liver function indicators is adjusted. When the value of the disease progression parameter is larger, the data collection time for liver function indicators should be longer, as more liver function indicator data is needed to confirm the patient's liver cancer incidence. Conversely, when the value of the disease progression parameter is smaller, the data collection time for liver function indicators can be shorter.

[0074] Therefore, by adjusting the acquisition time of liver function indicators using the aforementioned disease progression parameters, the adjustment time can be obtained.

[0075] In some embodiments, the method for obtaining the adjustment duration is as follows: Where S is the adjustment duration; Parameters for disease progression; The duration for which the indicator is acquired.

[0076] After determining the adjustment duration of liver function indicators, ALT and AST sequences from multiple tests within that duration are obtained. In other words, based on the adjustment duration, the ALT and AST sequences of the subjects are obtained in segments, and the time length of the ALT and AST sequences is the adjustment duration.

[0077] ALT and AST are mainly found inside hepatocytes. When liver cancer damages hepatocytes, they are released into the blood, causing elevated levels in the blood. In multiple tests, if liver cancer is in the development stage and the lesion is progressing rapidly, ALT and AST may rise sharply, showing a rapid upward trend, even increasing several times within weeks. At the same time, the ratio of the two may also change. In severe hepatocyte damage, mitochondria are damaged, AST release increases, and the AST / ALT ratio may be greater than 1, or even greater than 2 in some cases.

[0078] Analyzing the ascending trends of ALT and AST sequences reveals the degree of hepatocyte damage, specifically: Analyze the relative upward trend of ALT values ​​in the ALT sequence to obtain the first upward trend; By analyzing the relative upward trend of AST values ​​in the AST sequence, a second upward trend is obtained; By comparing the real-time ALT and real-time AST values ​​of the subjects, the relative changes in the indicators can be obtained; The degree of hepatocyte damage is obtained by combining the first upward trend, the second upward trend, and the relative change of the indicator; wherein the first upward trend, the second upward trend, and the relative change of the indicator are all positively correlated with the degree of hepatocyte damage.

[0079] In this embodiment of the invention, taking the ALT value and its corresponding ALT sequence as an example, the relative change value j of the ALT value between two adjacent detections in the ALT sequence is determined. ,in, and These represent the ALT values ​​of the later and earlier detections, respectively; the larger the value of j, the faster the ALT increases.

[0080] The mean of the relative changes in ALT values ​​between all two adjacent detections in the ALT sequence is calculated and denoted as the first upward trend j1, which is used to represent the upward trend of ALT values ​​within this time range. If the relative change rate of ALT values ​​in multiple detections within this period remains at a high level, it indicates that the ALT values ​​have maintained a rapid increase within this period.

[0081] Taking the AST value and its corresponding AST sequence as an example, determine the relative change value k of the AST value between two adjacent detections in the AST sequence. ,in, and These represent the AST values ​​of the second and first detections, respectively. The larger the value of k, the faster the AST increases.

[0082] The mean of the relative changes in AST values ​​between all two adjacent detections in the AST sequence is calculated and denoted as the second upward trend k1. This is used to represent the upward trend of AST values ​​within this time range. If the relative change rate of AST values ​​in multiple detections within this period remains at a high level, it indicates that the AST values ​​have maintained a rapid increase within this period.

[0083] The formula for calculating the degree of liver cell damage in the test subject is as follows: Where j1 represents the first upward trend and k1 represents the second upward trend. and These represent the real-time AST value and real-time ALT value of the last detection, respectively; This represents the relative change of indicators. When both AST and ALT values ​​rise rapidly, and the ratio of the last measured AST to ALT values ​​is larger, the dialysate reflects a more severe degree of hepatocellular damage in the patient; the corresponding value indicates a greater degree of hepatocellular damage.

[0084] Step S330: Combine the lesion progression parameters, the degree of hepatocyte damage, and the disease prediction parameters to determine the risk prediction parameters for the person to be tested.

[0085] The AFP value, representing the disease progression parameter, is adjusted based on the degree of hepatocellular damage in the patient to obtain the patient's predictive adjustment weight. This predictive adjustment weight characterizes the multidimensional disease features of liver cancer; that is, the predictive adjustment weight is determined based on the disease progression parameter and the degree of hepatocellular damage. The formula for calculating this predictive adjustment weight Y is: The greater the progression of liver cancer lesions indicated by AFP, and the greater the degree of liver cell damage, the more pronounced the liver cancer lesion characteristics in the patient.

[0086] By combining disease prediction parameters obtained from historical liver cancer patients with similar characteristics with prediction adjustment weights derived from changes in continuous monitoring indicators, risk prediction parameters for early-stage liver cancer in patients are obtained. In other words, the disease prediction parameters are weighted based on the prediction adjustment weights to obtain the risk prediction parameters.

[0087] In some embodiments, the product of the prediction adjustment weight and the disease prediction parameter is used as the risk prediction parameter.

[0088] The higher the disease prediction parameter and the greater the predictive adjustment weight corresponding to the multidimensional lesion characteristics, the higher the risk of liver cancer in the individual tested. The liver cancer lesion may be in a rapid progression stage. In this case, the monitoring interval of various indicators should be shortened, the examination frequency should be increased, and the patient's physical symptoms should be closely monitored. Conversely, the lower the value of the risk prediction parameter, the lower the risk of liver cancer in the current test. The liver cancer lesion may be in a relatively stable or slow development stage. The current monitoring frequency can be maintained to continuously track the changes in the condition, and the patient should be encouraged to maintain a healthy lifestyle, such as quitting smoking and alcohol, and maintaining a balanced diet.

[0089] By integrating multi-source data, including information on the lifestyle habits and disease characteristics of historically diagnosed patients, as well as the changes in multidimensional indicators during continuous monitoring of patients, the degree of liver cancer lesions reflected by each indicator is quantified, which improves the accuracy of predicting the incidence of liver cancer in patients, more accurately identifies patients at risk of liver cancer, and provides strong support for early prevention and diagnosis.

[0090] Please see Figure 3 , Figure 3 A system block diagram of an artificial intelligence-based liver cancer incidence pattern analysis and early prediction system is provided for embodiments of the present invention. The system includes the following modules: The information acquisition module is used to acquire the basic information and examination indicators of the person to be tested; The preliminary prediction module is used to classify the test subjects into feature clusters based on basic information; analyze the condition of historical patients in the feature cluster to which the test subject belongs, and the correlation between the test subject and historical patients, to determine the condition prediction parameters of the test subject; The risk prediction module is used to predict the risk of individuals under test by using a preset prediction method when the disease prediction parameters meet the monitoring conditions. The preset prediction method is as follows: analyze the progression characteristics of the test indicators before and after the threshold of the test subject's examination indicators to determine the lesion progression parameters; adjust the acquisition time of the examination indicators through the lesion progression parameters, acquire the test subject's examination indicator sequence in segments, analyze the upward trend of adjacent examination indicator sequences to obtain the degree of hepatocyte damage; combine the lesion progression parameters, the degree of hepatocyte damage and the disease prediction parameters to determine the risk prediction parameters of the test subject.

[0091] Alternatively, the transmission medium may be a wired link, such as, but not limited to, coaxial cable, fiber optic cable and digital subscriber line, or a wireless link, such as, but not limited to, wireless Fidelity (WIFI), Bluetooth and mobile device networks.

[0092] It should be noted that the device provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0093] This invention provides a computer device. Exemplarily, the computer device includes: a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the computer device can perform any of the aforementioned artificial intelligence-based methods for analyzing the incidence patterns and early prediction of liver cancer.

[0094] Furthermore, embodiments of the present invention also protect an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to perform the artificial intelligence-based liver cancer incidence pattern analysis and early prediction method provided in embodiments of the present invention.

[0095] In this embodiment of the invention, the device can be divided into functional modules according to the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and is only a logical functional division. In actual implementation, there may be other division methods.

[0096] When each module is divided according to its function, the device may also include a signal uploading module, a determination module, and an adjustment module. It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here.

[0097] It should be understood that the apparatus provided in this embodiment of the invention is used to perform the above-described artificial intelligence-based liver cancer incidence pattern analysis and early prediction method, and thus can achieve the same effect as the above-described implementation method.

[0098] When using integrated units, the device may include a processing module and a storage module. When applied to a device, the processing module can be used to control and manage the device's operations. The storage module can be used to support the device in executing program code, etc. The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as described in this disclosure. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of Digital Signal Processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0099] In addition, the device provided in the embodiments of the present invention may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the artificial intelligence-based liver cancer incidence pattern analysis and early prediction method provided in the above embodiments.

[0100] This invention also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the aforementioned method steps to implement the artificial intelligence-based liver cancer incidence pattern analysis and early prediction method provided in the above embodiments.

[0101] This invention also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to realize the artificial intelligence-based liver cancer incidence pattern analysis and early prediction method provided in the above embodiments.

[0102] In this invention, the apparatus, computer-readable storage medium, computer program product, or chip provided in the embodiments are all used to execute the corresponding methods described above. Therefore, the beneficial effects they achieve can be referred to the beneficial effects in the corresponding methods described above, and will not be repeated here. Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by this invention, it should be understood that the disclosed apparatus and method can be implemented in other ways.

[0103] The device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0104] It should also be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0105] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0106] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0107] The above content is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for analyzing the incidence patterns and early prediction of liver cancer based on artificial intelligence, characterized in that, The method includes the following steps: Obtain basic information and examination indicators of the individuals to be tested; Based on basic information, the individuals to be tested are divided into feature clusters; the condition of historical patients in the feature cluster to which the individual to be tested belongs and the correlation between the individual to be tested and historical patients are analyzed to determine the predictive parameters of the individual's condition. When the disease prediction parameters meet the monitoring conditions, the risk prediction parameters are obtained by predicting the risk of the person to be tested through a preset prediction method. The preset prediction method is as follows: analyze the progression characteristics of the test indicators before and after the threshold of the test subjects to determine the disease progression parameters; adjust the acquisition time of the test indicators through the disease progression parameters, acquire the test indicator sequence of the test subjects in segments, analyze the upward trend of adjacent test indicator sequences to obtain the degree of hepatocellular damage; combine the disease progression parameters, the degree of hepatocellular damage and the disease prediction parameters to determine the risk prediction parameters of the test subjects. Specifically, the progression characteristics of AFP sequences before and after threshold breakthrough in the subjects were analyzed to determine the pre- and post-peak progression parameters of AFP values; and the lesion progression parameters of the subjects were determined by combining the pre- and post-peak progression parameters. Specifically, all AFP values ​​in the AFP sequence that are greater than a preset threshold are obtained and used as the target AFP value; Determine the frequency and maximum consecutive length of the target AFP value; Obtain the length of the time series sequence after the first occurrence of the target AFP value, and use it as the length of the sequence after the breakthrough. The proportion of the maximum continuous length to the length of the sequence after the breakthrough is used as the breakthrough progress value; Analyze the overall magnitude of the target AFP value exceeding the preset threshold to determine the reference value after the breakthrough. By combining the number of occurrences of the target AFP value, the breakthrough progress value, and the reference value after the breakthrough, the subsequent progress parameters of the AFP value are obtained; Among them, the length of the time series before the first appearance of the target AFP value is obtained as the length of the pre-breakthrough sequence; Calculate the average difference between all adjacent AFP values ​​in the time series before the target AFP value first appears, and use it as a reference value before the breakthrough. Based on the pre-breakout sequence length and the pre-breakout reference value, the preceding progression parameters of the AFP value are obtained; among them, the pre-breakout sequence length is negatively correlated with the preceding progression parameters, and the pre-breakout reference value is positively correlated with the preceding progression parameters. The inspection indicators include: ALT value and AST value; The adjustment time is obtained by adjusting the acquisition time of liver function indicators through disease progression parameters; Based on the adjustment time, the ALT and AST sequences of the subjects were obtained in segments, and the upward trend of the ALT and AST sequences was analyzed to obtain the degree of hepatocyte damage.

2. The method for analyzing the incidence patterns and early prediction of liver cancer based on artificial intelligence according to claim 1, characterized in that, The analysis examines the condition of historical patients within the feature cluster to which the test subject belongs, and the correlation between the test subject and historical patients, to determine the predictive parameters for the test subject's condition, including: Acquire medical images of the person to be tested; the examination indicators include: AFP value; The distribution of tumor features and the magnitude of AFP values ​​in the medical images were analyzed to determine the severity index of the patient's condition. Based on the severity indicators of historical patients in the characteristic cluster to which the test subject belongs, and the correlation between the test subject and historical patients, the predictive parameters of the test subject's condition are determined.

3. The method for analyzing the incidence patterns and early prediction of liver cancer based on artificial intelligence according to claim 2, characterized in that, The analysis of the distribution of tumor features and AFP values ​​in the medical images determines the severity index of the patient's condition, including: Obtain the average diameter and number of tumors in medical images; The severity index of the patient's condition is determined based on the average diameter, the number of tumors, and the real-time AFP value of the patient; the average diameter, the number of tumors, and the real-time AFP value of the patient are all positively correlated with the severity index.

4. The method for analyzing the incidence patterns and early prediction of liver cancer based on artificial intelligence according to claim 1, characterized in that, The monitoring condition is that the disease prediction parameter is greater than the preset risk threshold.

5. The method for analyzing the incidence patterns and early prediction of liver cancer based on artificial intelligence according to claim 1, characterized in that, The analysis of the ascending trends of ALT and AST sequences yields the degree of hepatocyte damage, including: Analyze the relative upward trend of ALT values ​​in the ALT sequence to obtain the first upward trend; By analyzing the relative upward trend of AST values ​​in the AST sequence, a second upward trend is obtained; By comparing the real-time ALT and real-time AST values ​​of the subjects, the relative changes in the indicators can be obtained; The degree of hepatocyte damage is obtained by combining the first upward trend, the second upward trend, and the relative change of the indicator; wherein the first upward trend, the second upward trend, and the relative change of the indicator are all positively correlated with the degree of hepatocyte damage.

6. The method for analyzing the incidence patterns and early prediction of liver cancer based on artificial intelligence according to claim 1, characterized in that, The method of combining the disease progression parameters, the degree of hepatocyte damage, and the disease prediction parameters to determine the risk prediction parameters for the individual being tested includes: Based on the disease progression parameters and the degree of hepatocyte damage, predictive adjustment weights are determined; based on the predictive adjustment weights, the disease prediction parameters are weighted to obtain risk prediction parameters.

Images