Computing power device application technology for screening and early auxiliary diagnosis of liver cancer high-risk groups and postoperative treatment monitoring of patients

Abstract

The invention provides a computing power device application technology for screening of liver cancer high-risk groups, early auxiliary diagnosis and postoperative treatment monitoring of patients, and belongs to the fields of biotechnology, medical diagnosis and treatment and massive health. By collecting samples of human body liquid such as peripheral blood, plasma, serum, saliva and the like and using biological and medical detection technologies, molecular markers (nationally approved that each kit is equal to the molecular marker) of a target gene methylation level, miRNA expression quantity and protein level or immune index quantity are detected; the problems of low sensitivity and poor accuracy in liver cancer high-risk group screening, early auxiliary diagnosis and patient postoperative treatment monitoring (including kits) are solved by using a computing power device application technology according to the detection result of any combination of the molecular markers, the precision can be greatly improved to prevent excessive medical treatment, and the purposes of precise medical treatment and clinical application are achieved. The method has important practical significance on screening of high-risk groups of liver cancer, early auxiliary diagnosis and postoperative treatment monitoring of patients, and has wide social benefits and use values.

Description

Technical Field

[0001] The present invention relates to combining the computing power of large and small models with molecular bioinformatics, preventive medicine, biological detection technology, medical diagnostics and liver cancer treatment evaluation to conduct screening of high-risk groups for liver cancer, early auxiliary diagnosis and postoperative treatment monitoring and evaluation of patients, and belongs to the fields of big health, preventive medicine, clinical diagnostics, treatment evaluation and biotechnology. Background Art

[0002] Liver cancer is a major malignant tumor that people around the world are susceptible to. Currently, the screening effect of high-risk groups for liver cancer is far from ideal. Missing the best time for prevention and treatment has led to an increase in liver cancer deaths, which is a major global medical problem. Liver cancer biopsy and pathological examination are the gold standard for diagnosis. These are all damaging tests, and the patients suffer a lot before the operation. What kind of patients should undergo liver cancer biopsy and pathological examination is still a research challenge. Imaging diagnostic tests have the advantage of lesions occupying liver tumors and nodules, but they are very different in nature from the molecular biological level of the gold standard diagnosis of liver cancer biopsy and pathological examination, and are also affected by the subjective factors and experience of the radiologist. At present, the effective treatment effect of liver cancer is still far from ideal. There are still some new issues that need to be studied in treatment and rehabilitation evaluation. Clinical use The sensitivity and specificity of molecular markers such as AFP, DCP and kits are not high, especially the sensitivity and accuracy need to be significantly improved. There has been some progress in the study of molecular markers for liver cancer-related gene methylation levels, protein levels, miRNA expression levels and immune indexes. There are great difficulties in selecting molecular markers and optimizing a group of molecular markers. With the rapid development of big models, big data, artificial intelligence and the digitalization of the Internet of Things, many kits are still in the qualitative stage and lack quantitative processing. The results of a kit and multiple molecular markers cannot be accurately calculated using computing methods; and the prediction results of multiple indicators or more than two indicators lose specificity and sensitivity. Correct analysis. In order to overcome the above shortcomings, a series of integrated big model computing power-enabled calculations are performed on molecular markers for liver cancer-related gene methylation, protein expression, miRNA expression levels and immune indexes to solve the problems of low sensitivity and poor accuracy in screening of high-risk populations for liver cancer, early auxiliary diagnosis and postoperative treatment monitoring of patients (including kits). Such research on liver cancer application has not been officially reported at home and abroad. Such research greatly improves accuracy to prevent over-medicalization and achieve precision medicine. Non-destructive testing can also be used to achieve the goals of high-risk population screening, early auxiliary diagnosis and postoperative treatment monitoring of patients. Summary of the invention

[0003] The present invention aims to overcome the shortcomings in the application of comprehensive technologies in biotechnology, medical diagnosis and treatment, preventive medicine and health. By collecting samples of human fluids such as peripheral blood, plasma, serum, exosomes, saliva, etc., biological and medical detection technologies are used to detect the methylation level of target genes, the expression of miRNAs and the protein level or the molecular markers of immune indicators, as well as any combination of them or the molecular markers preferably combined by algorithms. The results are calculated using the technology of computing devices, thereby improving the accuracy of screening of high-risk groups for liver cancer, early auxiliary diagnosis and postoperative treatment monitoring of patients (including kits). Through non-destructive measurement, early screening and early diagnosis, the best time for prevention and treatment can be obtained, which is beneficial to improving the cure rate of liver cancer and the five-year survival rate of liver cancer patients.

[0004] The present invention is to use existing biological and medical detection technologies to detect the data of any molecular marker indicator combination in the basic molecular marker indicator system (also including related kit detection data); use the basic large model or small model algorithm to calculate the data of any molecular marker indicator combination to achieve the basic target classification and identification problem, thereby improving the accuracy of screening high-risk groups for liver cancer, early auxiliary diagnosis and postoperative treatment monitoring of liver cancer patients.

[0005] The content of the present invention is to expand the basic molecular marker index system to the imaging indicators in the imaging equipment, such as nodule conditions, mass conditions, vascular flow status, tumor imaging information, tumor space and tumor time portal vein (or hepatic vein, etc.) status or delayed liver tumor status, etc., which all constitute the marker index system of imaging characteristics. For the basic molecular marker index system and the marker index system of imaging characteristics and any combination of one or two or more markers in the marker index system of imaging characteristics, the basic target classification and recognition problem is achieved through the computing device technology.

[0006] The present invention is to solve the basic target classification and recognition problem by establishing and constructing a linear function or a nonlinear function, and using the value of the linear function, or the value of the nonlinear function, or combining the linear function with any one or any combination of two or more basic large model or small model algorithms, or combining the nonlinear function with any one or any combination of two or more basic large model or small model algorithms, using similar algorithms of computing device technology and similar methods.

[0007] The present invention is to define positive and negative by the proportion of negative and positive (or frequency or probability or objective weighting or weighted accumulation process) in the ROC curve sensitivity and specificity analysis method, comprehensive evaluation method and the method of finally determining negative and positive in the kit research for various liver cancer markers including the nationally approved kit and any combination of liver cancer markers. To achieve the purpose of screening high-risk population for liver cancer, early auxiliary diagnosis and postoperative treatment monitoring of patients.

[0008] The present invention is to select or optimize the basic molecular marker index system and the new basic molecular marker index system index (kit) detection, and make a table for comparison of the data range of any one or more marker combination detection of miRNA expression, protein expression, and DNA methylation level; and to achieve the comparison method by simple function calculation such as calculation of various indexes or comprehensive indexes, and to achieve the basic target classification and identification problems respectively through the detection data of any person and the comparison.

[0009] The present invention is to select molecular markers or a combination of markers or a kit in advance by an algorithm or an optimization algorithm, load the calculation process in the computing device application technology selected in advance and the software for compiling specific basic target classification and recognition problems into a portable device or a microcomputer or a mobile phone or an online background and platform or a proprietary device, etc., input the individual detection experiment results (data) into the portable device or a microcomputer or a mobile phone or an online background and platform or a proprietary device, etc., and after calculation, output the results of individual discrimination belonging to the specific basic target classification and recognition problem. DETAILED DESCRIPTION

[0010] Example: By reasonably selecting five relevant gene markers (X, Y, Z, W and U) involved in the occurrence and development of liver cancer, a high-risk group A1 for liver cancer, a medium-risk group A2 for precancerous lesions and a healthy group A3 without risk were set. A1 and A2 were previously proved to have liver cancer or precancerous lesions by pathological biopsy or other methods, and the healthy group A3 was a healthy group proved by physical examination center. 101 people in A1, 70 people in A2 and 3 were tested. 68 people conducted experiments using biotechnology and medical tests and obtained experimental results data. Various event probability libraries were derived from the experimental results data using statistics, such as: p(A1)=0.4226, p(A2)=0.2929, p(A3)=0.2848, p(×11|A1)=0.19802, p(Y23|A1)=0.81188, p(Z33|A1)=0.79208, p(W43|A1)=0.83168, p(U52|A1)=0.30693 and so on.Now there is a 43-year-old male who is willing to undergo liver cancer risk screening (or willing to undergo liver cancer auxiliary diagnosis in the hospital). Through peripheral blood extraction, biotechnology and medical testing, the first biomarker is equal to 2.42, the second biomarker is equal to 9.3, the third biomarker is equal to 242, the fourth biomarker is equal to 70, and the fifth biomarker is equal to 38. Assume that the five marker data are represented by X11, Y23, Z33, W43 and U52 respectively, and the individual symptom characteristics of this man are represented by B, symbol B = X11Y23Z3 3W43U52, calculated by the Beyes model method in the computing device of the present invention, p(B|A1)=p(X11|A1)×p(Y23|A1)×p(Z33|A1)×p(W43|A1)×p(U52|A1)=0.032506, similarly, p(B|A2)=p(X11|A2)×p(Y23|A2)×p(Z33|A2)×p(W43|A2)×p(U52|A2)=0.000846, p(B|A3)=p(X11|A2)×p(Y2 3|A3)×p(Z33|A3)×p(W43|A3)×p(U52|A3)=0.000000, P(A1|B)=(p(A1)×p(B|A1)) / (p(A1)×p(B|A1)+p(A2)×p(B|A2 )+p(A3)×p(B|A3))=0.98226P(A2|B)=(p(A2)×p(B|A2)) / (p(A1)×p(B|A1)+p(A2)×p(B|A2)+p(A3)×p(B|A3))=0.01 773P(A3|B)=(p(A3)×p(B|A3)) / (p(A1)×p(B|A1)+p(A2)×p(B|A2)+p(A3)×p(B|A3))=0.00000Max{0.98226,0.01773,0.00000}=0.98226, the result belongs to the high-risk group A1 for liver cancer. From the perspective of liver cancer risk screening, please ask this man to go to a tertiary hospital for further clinical diagnosis and confirmation; from the perspective of clinical auxiliary diagnosis, please ask this man to do a pathological biopsy immediately. The final result of the tissue pathological biopsy during surgery in a tertiary hospital proved that the man had liver cancer. It is consistent with the result calculated by the application technology of the computing device.

Claims

1. A computing device application technology for liver cancer high-risk population screening, early auxiliary diagnosis and patient postoperative treatment monitoring, characterized in that: Use existing biological and medical testing technologies to detect the data of any molecular marker combination in the following basic molecular marker indicator system (also including related kit test data); use the following basic large model or small model algorithm to calculate the data of any molecular marker combination to achieve the following basic target classification and identification problem, thereby improving the accuracy of screening for high-risk groups for liver cancer, early auxiliary diagnosis and postoperative treatment monitoring of liver cancer patients. The basic target classification and recognition problem involved in its characteristics is: the basic target classification and recognition problem is composed of the following two, three, four or more classification and recognition problems. The two classification (A and B) recognition problems are: that is, the positive and negative of the mass (or nodule), the malignant and benign of the mass (or nodule), liver cancer (or high risk) and liver precancerous lesions (including cirrhosis, HBV, HCV, hepatitis, and any of their related hepatitis B virus, hepatitis, aflatoxin and other liver diseases (all not HCC) are called risk class or medium risk class; these liver diseases can also be Any one or any two or more liver diseases are considered as precancerous lesions of liver cancer and are called risk class or medium-risk class), clinical stage I liver cancer and risk class, clinical stage I and II liver cancer and risk class, early stage liver cancer (or small liver cancer) and risk class, postoperative recurrence and non-recurrence of liver cancer, improvement and deterioration of treatment of patients after liver cancer surgery (improvement and deterioration are determined by comparing the development trends of one or more combined molecular marker indicators monitored), preoperative metastasis and non-metastasis of liver cancer, HCC and other cancers, and two types of classification and identification problems of HCC and ICC. The three-category (A, B, and C) identification problem: Class A and Class B represent the above-mentioned meanings respectively, Class C represents normal healthy people's liver distension or normal healthy people's liver distension and mild liver disease such as early hepatitis disease (respectively called risk-free or low-risk), and the three-category identification problem is a high-risk, medium-risk, and low-risk (or risk-free) identification problem (it can also be three types of auxiliary diagnosis identification); it also includes Class A, Class B, and Class C identification problems with arbitrary meanings; it also includes two-category identification problems of Class A and Class B, Class A and Class C, Class B and Class C, or Class A and Class B and Class C, or Class A and Class B and Class C, or Class A and Class B and Class C. The identification problems of four categories or above (category A, category B, category C, category D, etc.) are as follows: category A, category B and category C represent the above meanings respectively, category D refers to liver diseases or non-liver diseases involved in or developed by category A and category B or related to research needs, such as category A, B, C and D identification problems of cancers other than liver cancer; category D refers to clinical stages 3 and 4 of liver cancer, and category A refers to clinical stages 2 and 1 of liver cancer, category A, B, C and D identification problems; if A, B, C represent high, medium and low differentiation, and D represents precancerous lesions, category A, B, C and D identification problems; in short, the meanings of category A, category B, category C and category D can be arbitrarily changed, and their arbitrary meanings, such as metastasis, non-metastasis, ICC, non-liver cancer cancer names or hepatitis C virus, etc., as long as they are not confused, they are all identification problems of four categories or above, and classification identification problems of four categories or above also include classification identification problems of any combination of two categories, any combination of three categories, and any combination of four categories. The basic molecular marker index system involved in its characteristics is that the molecular markers are composed of the following target gene methylation levels, miRNA (microRNA) expression (expression level) and protein levels (protein expression level) or immune indicators. Among the molecular markers, any two or more molecular markers bundled together, any selected related kit is also a molecular marker, any molecular marker (or a preferred group of molecular markers) selected from the sensitivity and specificity analysis of the ROC curve, or any other method. Preferred molecular markers (or a suitable group of molecular markers), the above molecular markers and any combination of two or more of them constitute the basic molecular marker index system; this basic molecular marker index system also includes, using existing biological and medical detection technologies, respectively extracting plasma or serum or peripheral blood from human blood, free tfDNA in human blood, free ctDNA in human tumor blood, urine or pulmonary fluid or exosomes or saliva Liquid samples include mononuclear cells in the blood, circulating free DNA, circulating tumor DNA or other human fluids such as urine. After the liquid samples are pretreated, quantitative PCR, fluorescence quantitative PCR, quantitative methylation-specific PCR, real-time fluorescence quantitative PCR, methylation chip, next-generation sequencing, digital PCR, nucleic acid mass spectrometry, chemiluminescence, ELISA (elisa), immunochemiluminescence, chemiluminescence, second-generation sequencing technology, miRNA expression profile sequencing, fluorescence quantitative PCR detection and analysis after synthesizing cDNA, and detection technology according to kits (nationally approved and unapproved, as well as foreign kits), etc., are used to detect the methylation level (expression level) of the target gene, the expression amount (expression level) of miRNA (microRNA), the protein level or the amount of immune indicators, and obtain the detection results (it is noted here that the data of the basic molecular marker indicator system are not only derived from experimental detection data, but also from any other data or any combination of the two source data).), the target gene methylation levels refer to the DNA methylation of SCN4B, HOXA10, BDH1, GNB4, TCF24, Riplet, ACP1, SPP1, LCAT, P15, P16, BMPR1, PLAC8, ATXN1, KLF3, DBX2, THY1, TGR5, MT1M, MT1G, INK4A, VIM, RASSF1A, APC, COX2, DAB2IP, ZNF783 and TSPYL5 molecular markers (one or more The expression levels of miRNAs include the following: miR-122, miR-223, miR-26a, miR-27a, miR-801, miR-200a, miR-221, miR-16, miR-195, miR-206, miR-15a, miR-16a, miR-224, miR-21, h The expression levels of sa-let-7a, miRNA-301, hsa-miR-125a, miR-101-3p, miR-29a, miR-29c, miRNA-26a, miRNA-485-3p, miR-15a, miR-133a, miR-143, miR-145, miR-192, and miR-505 markers, and also include the miRNA expression levels (or positive and negative) measured by any molecular marker combination kit; protein level (expression level) or immune index The amounts are, respectively, the protein levels or blood marker amounts of PRKD3, AFP, PIVKA-II (DCP), AFP-L3, IFN-γ, TNF-α, CK7, CK9, CKIT, GS, HSP70, GPC-3, γ-GT, GP73, AST, GGT, GGT-II, GPDA-F, ALT, AFU, TK1, CHI3L1, CO3, ALP, and FER, and also include the protein levels or blood marker amounts (or positive and negative) measured by a kit of any combination of markers. The basic large model or small model algorithm involved in its characteristics is that the classification and recognition algorithms include: random deep forest algorithm, decision tree algorithm, logistic regression (Logistic), Bayesian algorithm, support vector machine algorithm, neural network algorithm, machine learning classification algorithm, deep learning classification algorithm, cluster analysis algorithm, discriminant analysis algorithm, artificial intelligence classification algorithm, K-nearest neighbor algorithm, association rule classification algorithm, kernel-based algorithm, K-meas algorithm, fuzzy classification recognition algorithm, other classification algorithm recognition algorithm, etc., and a comprehensive evaluation is performed by one of their classification and recognition algorithms or a combination of any two or more classification and recognition algorithms, or for the classification and recognition problems of specific basic targets selected in advance and the detection data (or any other data from any source or data from both sources) after arbitrarily combining into a group of specific molecular markers in the basic molecular marker indicator system, any algorithm after optimization or any selected algorithm, or any algorithm in the classification and recognition algorithm, the above algorithms and any combination of them constitute the basic large model or small model algorithm.

2. A computing device application technology for liver cancer high-risk population screening, early auxiliary diagnosis and patient postoperative treatment monitoring, characterized in that: The basic molecular marker index system in 1 above is expanded to imaging equipment including medical X-ray machines, digital imaging equipment, X-ray computed tomography equipment, magnetic resonance imaging equipment, ultrasound imaging (imaging), nuclear medicine imaging equipment, color Doppler blood flow imaging, enhanced (or non-enhanced or dynamically enhanced) CT, MRI, and imaging indicators in digital subtraction angiography such as nodule conditions, mass conditions, vascular flow status, tumor imaging information, tumor space and tumor time portal vein (or hepatic vein) status or delayed liver tumor status, etc. (including indicators with similar names here) all constitute the marker index system of imaging characteristics (also including a large number of tumor characteristic marker indicators measured by imaging equipment, and also including quantitative or standardized indicators of imaging tumor markers). Any one or any combination of two or more markers in the basic molecular marker index system and the marker index system of imaging characteristics, and any combination of one or two or more markers in the marker index system of imaging characteristics, are processed in a similar way as in 1 above to achieve the purpose of the classification and recognition problem in 1 above.

3. A computing device application technology for liver cancer high-risk population screening, early auxiliary diagnosis and patient postoperative treatment monitoring, characterized in that: The basic molecular marker index system of 1 above is expanded to a new basic molecular marker index system. The new basic molecular marker index system refers to molecular markers about target gene methylation level, miRNA (microRNA) expression (expression level) and protein level or immune index. Such molecular markers are molecular markers not included in the previous basic molecular marker index system and molecular markers discovered now and in the future (including the test results of the kits that will be newly approved by the state). For the basic molecular marker index system and one marker or any combination of two or more markers in the new basic molecular marker index system, or for one marker or any combination of two or more markers in the new basic molecular marker index system, a similar method as 1 above is used to achieve the purpose of the classification and identification problem of 1 above.

4. A computing device application technology for liver cancer high-risk population screening, early auxiliary diagnosis and patient postoperative treatment monitoring, characterized in that: The above 1, 2 and 3 are expanded to construct linear functions or nonlinear functions, and the values ​​of linear functions, or the values ​​of nonlinear functions, or the linear function is combined with any one or any combination of two or more basic large model or small model algorithms, or the nonlinear function is combined with any one or any combination of two or more basic large model or small model algorithms, and similar methods to the above 1, 2 and 3 are used to solve the classification and recognition problems of the above 1, 2 and 3.

5. A computing device application technology for liver cancer high-risk population screening, early auxiliary diagnosis and patient postoperative treatment monitoring, characterized in that These three methods, method 1, for the various liver cancer markers involved in 1, 2 and 3 above (including kits approved by the state), any combination of two or more liver cancer markers, using the threshold of the experiment itself or the threshold given by the kit or the threshold given in other articles, using the AUC analysis method of ROC curve sensitivity and specificity, for two or more liver cancer markers, and when the number of liver cancer markers is an even number, the diagnosis is positive if the number of positives is greater than the number of negatives, and the diagnosis is negative if the number of positives is less than the number of negatives. Method 2, for the various liver cancer markers involved in 1, 2 and 3 above (including kits approved by the state), any algorithm in the basic large model or small model algorithm or any combination of algorithms is used to determine the negative and positive methods (or the negative and positive methods are determined using the AUC of the ROC curve sensitivity and specificity analysis; Or the method of determining negative and positive in the research kit; other calculation methods to determine negative and positive) and any combination of them to determine negative and positive methods are all collectively referred to as comprehensive evaluation methods. In the comprehensive evaluation method, more than 2 / 3 of the positive numbers are diagnosed as strong positive, 1 / 3 to less than 2 / 3 of the positive numbers are determined as positive, less than 1 / 3 of the positive numbers are determined as weak positive, and no positive is determined as negative. Method 3, according to the similar comprehensive evaluation method of the above two methods in 5, in the kit, the method of finally determining positive and negative according to the proportion of negative and positive (or frequency or probability); or in the kit, the number of molecular markers selected in advance, the weight number is specified according to the contribution size of each molecular marker (or the weight number is specified according to the weight size of each molecular marker, or the weight number of each molecular marker is equal), and finally the number after the sum of the marker and the weight number is greater than or equal to a fixed number or less than a fixed number, the method of determining positive and negative; all belong to the scope of this claim.

6. A computing device application technology for liver cancer high-risk population screening, early auxiliary diagnosis and patient postoperative treatment monitoring, characterized in that: Through the detection of any marker of the basic molecular marker index system 1 above and the new basic molecular marker index system 3 above, the specific two or three or four categories of meaning representation of the basic target classification and identification problem 1 above are tabulated corresponding to the data range of any one or more marker combination detection of miRNA expression, protein expression, and DNA methylation level in the basic molecular marker index system and the new basic molecular marker index system for comparison; there is also a simple function to calculate various indexes or comprehensive indexes to achieve the comparison method; there is also a drawing software containing different calculation methods, which is loaded into a portable device (or microcomputer, mobile phone, online platform, proprietary equipment, etc.) for calculation, and through the detection data of any person, the specific two or three or four categories of targets of the specific basic target classification and identification problem can be achieved respectively.

7. A computing device application technology for liver cancer high-risk population screening, early auxiliary diagnosis and patient postoperative treatment monitoring, characterized in that: In advance, similar methods and ideas as in 1, 2, 3, 4, 5, 6 above are used to compile software for the research data, results, implementation methods and calculation processes and load them into portable devices, microcomputers, mobile phones, online backends and platforms, or proprietary devices, etc. The experimental results of an individual test (or data quantified by the results) or the experimental results of the markers of 1, 2, 3, 4, 5, 6 above (data or data quantified by the results) are input into portable devices, microcomputers, mobile phones, online backends and platforms, or proprietary devices, etc. After calculation, the basic target classification and recognition problem (or the purpose of a similar basic target classification and recognition problem) of determining whether an individual belongs to the specifically specified 1 above is output.