CN112331333A - Method for establishing liver cancer diagnosis model based on liver cancer triple-joint inspection - Google Patents
Method for establishing liver cancer diagnosis model based on liver cancer triple-joint inspection Download PDFInfo
- Publication number
- CN112331333A CN112331333A CN202011202929.4A CN202011202929A CN112331333A CN 112331333 A CN112331333 A CN 112331333A CN 202011202929 A CN202011202929 A CN 202011202929A CN 112331333 A CN112331333 A CN 112331333A
- Authority
- CN
- China
- Prior art keywords
- liver cancer
- model
- establishing
- diagnosis model
- cancer diagnosis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 208000014018 liver neoplasm Diseases 0.000 title claims abstract description 90
- 201000007270 liver cancer Diseases 0.000 title claims abstract description 89
- 238000003745 diagnosis Methods 0.000 title claims abstract description 71
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000007689 inspection Methods 0.000 title abstract description 10
- 206010073071 hepatocellular carcinoma Diseases 0.000 claims abstract description 58
- 231100000844 hepatocellular carcinoma Toxicity 0.000 claims abstract description 57
- 102000013529 alpha-Fetoproteins Human genes 0.000 claims abstract description 45
- 108010026331 alpha-Fetoproteins Proteins 0.000 claims abstract description 45
- 238000001514 detection method Methods 0.000 claims abstract description 32
- 108010094028 Prothrombin Proteins 0.000 claims abstract description 17
- 102100027378 Prothrombin Human genes 0.000 claims abstract description 17
- 230000002159 abnormal effect Effects 0.000 claims abstract description 17
- 229940039716 prothrombin Drugs 0.000 claims abstract description 17
- 238000007477 logistic regression Methods 0.000 claims abstract description 13
- 230000000405 serological effect Effects 0.000 claims abstract description 10
- 102100023635 Alpha-fetoprotein Human genes 0.000 claims 2
- 208000019423 liver disease Diseases 0.000 description 12
- 238000012549 training Methods 0.000 description 11
- 210000002966 serum Anatomy 0.000 description 10
- 238000010200 validation analysis Methods 0.000 description 10
- 230000035945 sensitivity Effects 0.000 description 8
- 208000019425 cirrhosis of liver Diseases 0.000 description 6
- 238000003018 immunoassay Methods 0.000 description 6
- 208000002672 hepatitis B Diseases 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000012795 verification Methods 0.000 description 5
- 102000009027 Albumins Human genes 0.000 description 4
- 108010088751 Albumins Proteins 0.000 description 4
- 206010016654 Fibrosis Diseases 0.000 description 4
- 238000008050 Total Bilirubin Reagent Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 230000007882 cirrhosis Effects 0.000 description 4
- 239000003550 marker Substances 0.000 description 4
- 102000004856 Lectins Human genes 0.000 description 3
- 108090001090 Lectins Proteins 0.000 description 3
- 206010028980 Neoplasm Diseases 0.000 description 3
- 108010063628 acarboxyprothrombin Proteins 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000002523 lectin Substances 0.000 description 3
- 210000004185 liver Anatomy 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 208000004057 Focal Nodular Hyperplasia Diseases 0.000 description 2
- 206010019646 Hepatic cyst Diseases 0.000 description 2
- 201000003803 Inflammatory myofibroblastic tumor Diseases 0.000 description 2
- 206010068331 Inflammatory pseudotumour Diseases 0.000 description 2
- 206010024652 Liver abscess Diseases 0.000 description 2
- 208000007720 Plasma Cell Granuloma Diseases 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 201000011510 cancer Diseases 0.000 description 2
- 238000011498 curative surgery Methods 0.000 description 2
- 201000011066 hemangioma Diseases 0.000 description 2
- 230000002440 hepatic effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000000056 organ Anatomy 0.000 description 2
- 238000010827 pathological analysis Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 208000035473 Communicable disease Diseases 0.000 description 1
- 208000027761 Hepatic autoimmune disease Diseases 0.000 description 1
- 241000700721 Hepatitis B virus Species 0.000 description 1
- 208000026594 alcoholic fatty liver disease Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000012333 histopathological diagnosis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 201000007450 intrahepatic cholangiocarcinoma Diseases 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000002638 palliative care Methods 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000035935 pregnancy Effects 0.000 description 1
- 230000001850 reproductive effect Effects 0.000 description 1
- 238000002271 resection Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000002054 transplantation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Pathology (AREA)
- Primary Health Care (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Biophysics (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Genetics & Genomics (AREA)
- Molecular Biology (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biotechnology (AREA)
- Evolutionary Biology (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Theoretical Computer Science (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention discloses a method for establishing a liver cancer diagnosis model based on liver cancer triple joint inspection, which comprises the following steps: establishing a primary hepatocellular carcinoma clinical representation and a primary hepatocellular carcinoma database of laboratory data, and collecting laboratory indexes of a patient; step two, collating laboratory indexes, and establishing a multi-factor Logistic regression model to obtain a liver cancer diagnosis model; wherein, the laboratory indexes of the liver cancer diagnosis model C-GALAD comprise basic information sex, age, liver cancer serological markers alpha-fetoprotein heteroplasmon, alpha-fetoprotein and abnormal prothrombin of the patient. Laboratory indexes of the liver cancer diagnosis model LAD comprise liver cancer serology markers including alpha-fetoprotein heteroplasmon, alpha-fetoprotein and abnormal prothrombin. The parameters of the two liver cancer diagnosis models are easy to obtain, and the parameters are on the same detection sub-discipline detection platform, so that the single report unit is less, the influence factors are less, and the clinical applicability and the feasibility are strong.
Description
Technical Field
The invention belongs to the technical field of clinical examination and diagnosis, and particularly relates to a method for establishing a liver cancer diagnosis model based on liver cancer triple-joint detection.
Background
Worldwide, liver cancer accounts for the sixth incidence of all cancers, with the third highest mortality rate, mainly including hepatocellular carcinoma (HCC) (about 75% -85%), intrahepatic cholangiocellular carcinoma (about 10% -15%) and other rare pathological types. Although significant advances have been made in the prevention, diagnosis, management and treatment of liver disease by hepatologists and researchers in our country over the last two decades, the overall 5-year survival rate remains below 40%, and about two-thirds of liver cancers are in the middle and late stages when clinically discovered, losing the best opportunity for liver resection and liver transplantation, and even palliative treatment cannot be performed. Therefore, the search for a good liver cancer diagnosis index is a critical problem to be solved urgently at the present stage. Serum alpha-fetoprotein (AFP), alpha-fetoprotein heteroplasmon (AFP-L3%) and abnormal prothrombin (DCP, also called PIVKAII) are all commonly used serological markers for diagnosing HCC at the present stage, but the diagnostic sensitivity and specificity of a single serological marker for early liver cancer are insufficient.
The artificial intelligence is a new technology which is rapidly developed in many fields in recent years, and the HCC early warning model is applied to the early diagnosis of HCC, so that a new way can be opened up for the diagnosis of HCC. The international multicenter and the previous research of the team have proved that the diagnosis model can greatly improve the clinical diagnosis efficiency of HCC. The international general GALAD model is established by Johnson et al in 2014, is based on a diagnosis model of sex (G), age (A), three liver cancer serum markers, namely alpha-fetoprotein (A), alpha-fetoprotein heteroplasmon (L) and abnormal prothrombin (D), can obviously improve the diagnosis efficiency of HCC, supplements the limitation of ultrasonic to noninvasive diagnosis of liver cancer, and is verified in international queues. The GAPTALA model established in the early stage of the team is a diagnosis model combining laboratory indexes of sex (G), age (A), platelet (P), total bilirubin (T), albumin (A), alpha fetoprotein heteroplasmon (L), alpha fetoprotein (A) and abnormal prothrombin (D), the diagnosis efficiency of the model on HCC is far higher than that of a single serological marker, and the model is verified in independent clinical cases.
However, the above international model GALAD and international multi-center research do not focus on hepatitis B virus-related liver cancer, and more than 80% of HCCs in China are related to hepatitis B virus infection. Although the GAPTALAD model established in the early stage of the team has higher diagnosis efficiency on HBV-related HCC, laboratory indexes contained in the model relate to detection systems for multiple clinical examinations of multiple sub-disciplines, the number of report units is more, and the model is limited by various factors such as different examination sub-disciplines, different detection principles, different instruments and equipment and the like when being used: firstly, the test reports of each test sub-discipline professional group have different issuing time, the final model result needs the coordination and cooperation of a plurality of professional groups, and the clinical suitability is limited; secondly, the report unit excessively performs automatic calculation on the model numerical value, so that certain maintenance difficulty and result variation calculation risk exist; third, the final calculation results of the model may be biased due to differences in instrumentation, reagents, methodologies, etc. between the detection targets, particularly clinical biochemical targets laboratories.
Disclosure of Invention
The invention aims to provide a method for establishing a diagnosis model of primary hepatocellular carcinoma based on more than 80% of Chinese people with hepatitis B virus infection as a background, and the model is used for identifying primary hepatocellular carcinoma (HCC), Benign Liver Diseases (BLDs) and/or healthy contrast so as to effectively improve the clinical application effect of triple joint inspection. Therefore, the applicant tries to reduce the parameters of the model on the basis of the original modeling type GAPTALD under the condition of ensuring that the effectiveness of the liver cancer diagnosis model is not influenced, so that the parameters of the model are obtained on the same detection platform of the sub-disciplines of examination as far as possible, and errors and clinical operation complexity caused by different sub-disciplines of examination, different detection principles and the like are avoided.
The applicant finds that the model diagnosis efficiency is not influenced after three parameters (P, T and A) with different detection methodologies are removed on the basis of the original self-modeling GAPTALD through long-term research. Therefore, the applicant innovates and establishes a personalized liver cancer Chinese model (C-GALAD model) and an independent triple inspection model (LAD model) which are simple and suitable for the Chinese situation. Through verification, the establishment of the LAD model can provide comprehensive quantitative interpretation standard for interpretation of the detection result of the liver cancer triple joint inspection, and is convenient for clinicians and patients to intuitively obtain the comprehensive interpretation result; because liver cancer is related to sex and age, the individual information of the patient which is convenient and easy to obtain is fused: sex and age, the obtained C-GALAD model realizes individual application, and effectively improves the clinical application effect of the triple joint inspection. Therefore, the first objective of the invention is to provide a method for establishing a personalized liver cancer diagnosis model C-GALAD based on liver cancer triple-joint detection. The second purpose of the invention is to provide a method for establishing a liver cancer diagnosis model LAD based on liver cancer triple-joint detection, which is used when gender and age information cannot be acquired.
In order to achieve the purpose, the invention provides the following technical scheme:
as a first aspect of the invention, a method for establishing a liver cancer diagnosis model C-GALAD based on liver cancer triple-joint detection comprises the following steps:
establishing a primary hepatocellular carcinoma clinical representation and a primary hepatocellular carcinoma database of laboratory data based on medical records such as a laboratory information management system and an electronic medical record, and collecting laboratory indexes of a patient;
and step two, collating laboratory indexes, establishing a multi-factor Logistic regression model, and obtaining a liver cancer diagnosis model C-GALAD. The obtained liver cancer diagnosis model C-GALAD can distinguish primary hepatocellular carcinoma from benign liver disease and/or healthy control, and assist in judging the risk of the benign liver disease or/and the healthy control of the primary hepatocellular carcinoma, and accurate individual application is realized due to fusion and individualized factors such as age, sex and the like.
Wherein the laboratory indexes comprise basic information sex (G), age (A), liver cancer serological markers alpha-fetoprotein heteroplasmon (L), alpha-fetoprotein (A) and abnormal prothrombin (D) of the patient.
According to the invention, the model formula of the liver cancer diagnosis model C-GALAD is as follows:
C-GALAD＝-8.654+A×Gender+B×Age+C×AFP-L3+D×Log10(AFP)+E×Log10(DCP)；
wherein the coefficient A is 1.142-1.516; the coefficient B is 0.037-0.052; the coefficient C is 0.045-0.082; the coefficient D is 0.750-1.020; the coefficient E is 2.894-3.382.
Preferably, the factor a is 1.329; the coefficient B is 0.044; the coefficient C is 0.063; the coefficient D is 0.885; the coefficient E is 3.138.
According to the invention, serum alpha-fetoprotein is quantitatively detected by an (electro) chemiluminescence full-automatic immunoassay analyzer, a liver cancer serological marker alpha-fetoprotein heteroplasmon is quantitatively detected by (electro) chemiluminescence of a lectin LCA + antibody, and abnormal serum prothrombin is quantitatively detected by a chemiluminescence full-automatic immunoassay analyzer.
According to the present invention, the liver cancer diagnosis model C-GALAD has a value in the range of 0.5268-1.1734 as a cutoff value (cut-off value) of the judgment result.
Preferably, the liver cancer diagnostic model C-GALAD has a cut-off value of 0.9382.
As a second aspect of the present invention, a method for establishing a liver cancer diagnosis model LAD based on triple-link detection of liver cancer comprises the following steps:
establishing a primary hepatocellular carcinoma clinical representation and a primary hepatocellular carcinoma database of laboratory data based on medical records such as a laboratory information management system and an electronic medical record, and collecting laboratory indexes of a patient;
and step two, collating the laboratory indexes, establishing a multi-factor Logistic regression model, and obtaining a liver cancer diagnosis model LAD. The model provides a comprehensive quantitative interpretation standard for interpretation of the triple detection result of the liver cancer; when the model is applied, primary hepatocellular carcinoma can be distinguished from benign liver diseases and/or healthy controls, and the risk of primary hepatocellular carcinoma of the benign liver diseases and/or the healthy controls is judged in an auxiliary mode.
Wherein the laboratory indexes comprise liver cancer serological markers of alpha-fetoprotein heteroplasmon (L), alpha-fetoprotein (A) and abnormal prothrombin (D).
According to the invention, the model formula of the liver cancer diagnosis model LAD is as follows:
LAD＝-5.930+H×AFP-L3+Y*Log10(AFP)+Z×Log10(DCP)；
wherein the coefficient H is 0.039-0.075, the coefficient Y is 0.759-1.025, and the coefficient Z is 3.277-3.772.
Preferably, the coefficient H is 0.057; the coefficient Y is 0.892; the factor Z is 3.524.
According to the invention, serum alpha-fetoprotein is quantitatively detected by an (electro) chemiluminescence full-automatic immunoassay analyzer, a liver cancer serological marker alpha-fetoprotein heteroplasmon is quantitatively detected by (electro) chemiluminescence of a lectin LCA + antibody, and abnormal serum prothrombin is quantitatively detected by a chemiluminescence full-automatic immunoassay analyzer.
According to the present invention, the LAD of the liver cancer diagnosis model has a cut-off value in a range of 0.1455 to 0.9901.
Preferably, the LAD of the liver cancer diagnosis model has a cut-off value of 0.3320.
The liver cancer diagnosis model based on the liver cancer triple-joint detection has the beneficial effects that:
1. the parameters of the liver cancer diagnosis model C-GALAD and the liver cancer diagnosis model LAD are easy to obtain, the parameters are obtained on the same test sub-discipline platform, a single report unit is adopted, the error of the experimental result is small, and the influence of different detection platforms on the comprehensive calculation result is avoided, so that the clinical applicability and the feasibility of the model are stronger.
2. The establishment of the C-GALAD model and the LAD model can provide comprehensive quantitative interpretation standards for the interpretation of the detection result of the liver cancer triple joint inspection, and facilitates clinicians and patients to intuitively obtain the comprehensive interpretation result; when the model is used, the model can be used in a laboratory detection system or an information system LIS, the model value is automatically calculated through the detection index value in the model, and the risk of primary hepatocellular carcinoma caused by benign liver diseases or/and healthy contrast is judged through the cut-off value of the model.
3. Because the liver cancer is related to the sex and the age, the sex and the age of the patient which are convenient and easy to obtain are combined, the obtained C-GALAD model realizes the individual application, and the clinical application effect of the triple joint inspection is effectively improved. The LAD model may be selected when gender and age information cannot be obtained.
Drawings
FIG. 1 is a graph of the diagnostic efficiency of the training set for diagnosing primary hepatocellular carcinoma using the C-GALAD model.
FIG. 2 is a graph showing the diagnostic efficiency of the C-GALAD model for diagnosing primary hepatocellular carcinoma in the validation group.
Fig. 3 is a graph of the diagnostic efficiency of the training set for diagnosing primary hepatocellular carcinoma using the LAD model.
Fig. 4 is a graph of the diagnostic efficiency of the validation group for diagnosing primary hepatocellular carcinoma using the LAD model.
Fig. 5 shows the GAPTALAD model and the efficiency of diagnosis of primary hepatocellular carcinoma in the training set after the removal of platelets (P), total bilirubin (T) and albumin (a), respectively.
Fig. 6 shows the GAPTALAD model and the efficiency of diagnosis of primary hepatocellular carcinoma in the validation set after the removal of platelets (P), total bilirubin (T) and albumin (a), respectively.
FIG. 7 shows the efficiency of diagnosis of primary hepatocellular carcinoma in the training set following the C-GALAD model and gender (G), age (A) knockout.
FIG. 8 is a graph of the diagnostic efficiency of primary hepatocellular carcinoma in a validation set following the C-GALAD model and gender (G), age (A) knockout.
Detailed Description
The present invention will be further described with reference to the following specific examples. It should be understood that the following examples are illustrative only and are not intended to limit the scope of the present invention. The experimental procedures, in which specific conditions are not specified, in the following examples are generally conducted under conventional conditions, or under conditions provided by the manufacturers.
The following examples are based on the establishment and application of regression models of national clinical test indices for differentiating primary hepatocellular carcinoma from benign liver disease and/or healthy controls. Therefore, the invention firstly establishes a primary hepatocellular carcinoma clinical representation and a primary hepatocellular carcinoma database of laboratory data for data acquisition, and the acquisition source is a laboratory information management system LIS of a detection unit. The experimental group is patients who receive surgical excision and have pathological diagnosis of primary hepatocellular carcinoma (HCC) after operation, and the control group is benign liver disease patients (BLDs) which are hospitalized in the same period and have clinical positive exclusion of the primary hepatocellular carcinoma, and comprises liver cirrhosis, hepatic hemangioma, liver abscess, hepatic cyst, inflammatory pseudotumor, focal nodular hyperplasia and the like.
1. Sample collection
(1) Collection of serum samples from patients with primary hepatocellular carcinoma
Using retrospective analysis, the training set was from 5998 cases of primary hepatocellular carcinoma patients receiving curative surgery treatment at the hospital from 1 month 2015 to 12 months 2017, with 85% of hepatitis b virus infection cases. The validation group was 2064 cases of primary hepatocellular carcinoma patients who received curative surgery treatment in the hospital from 2018 month 1 to 2018 month 12, with 82.7% of hepatitis b virus infection cases.
The patients with primary hepatocellular carcinoma were enrolled to meet the following criteria: the early cases which can be treated by radical operation are mainly treated; pathological diagnosis definitively grade its tumor pathology (Edmondson); all the relevant information of the cases is complete; eliminating chronic liver diseases caused by other reasons, such as alcoholic fatty liver disease, autoimmune liver disease and the like; excluding pregnancy, reproductive embryonic tumors, malignant tumors of other organs, serious infectious diseases, other important organ diseases and the like.
(2) Serum specimen collection for patients with benign liver disease
Using retrospective analysis, the training set was derived from 1731 cases of benign liver diseases (including cirrhosis, hepatic hemangioma, liver abscess, hepatic cyst, inflammatory pseudotumor, focal nodular hyperplasia, etc.) patients receiving treatment at the hospital 2015 1-2018 12. The validation group was from 253 patients with cirrhosis who received treatment at the hospital between 2018 and 2019, months 8.
Wherein the diagnosis of the selected cirrhosis patients meets the following criteria: referring to "guidelines for prevention and treatment of hepatitis B", cirrhosis is diagnosed as clinical features or imaging; liver cirrhosis is suggested by liver penetrating histopathological diagnosis standards; the relevant information of all cases is complete; eliminating liver cancer and other malignant tumors.
2. Detection instrument and statistical software
The serum alpha-fetoprotein adopts an (electro) chemiluminescence full-automatic immunoassay analyzer to carry out quantitative detection; the alpha-fetoprotein heteroplasmon is quantitatively detected by (electro) chemiluminescence of a lectin LCA + antibody; the abnormal prothrombin in serum is quantitatively detected by a chemiluminescence full-automatic immunoassay analyzer, and all data are statistically analyzed by SPSS 22.0 analysis software.
3. The basic information of the selected cases is shown in table 1.
TABLE 1 basic information of the selected cases
Example 1 establishment of liver cancer diagnostic model C-GALAD
The model parameters selected in this example are patient basic information sex (G), age (A), liver cancer serological markers alpha-fetoprotein heteroplasmon (L), alpha-fetoprotein (A), and abnormal prothrombin (D), then a multi-factor binary Logistic regression analysis is performed on all inclusion indexes by using a binary Logistic regression, and on the basis, a multi-factor Logistic regression model C-GALAD based on sex, age and liver cancer triple joint inspection (AFP, AFP-L3, DCP) is established, and the results are shown in Table 2.
TABLE 2 multifactor Logistic regression model C-GALAD
| Variable (Variable) | β(95％CI) | OR(95％CI) | P value |
| Gender(G) | 1.329(1.142-1.516) | 3.778(3.134-4.554) | <0.001 |
| Age(A) | 0.044(0.037-0.052) | 1.045(1.038-1.053) | <0.001 |
| Log10AFP,ng/mL(A) | 0.885(0.750-1.020) | 2.423(2.117-2.773) | <0.001 |
| Log10DCP,mAU/mL(D) | 3.138(2.894-3.382) | 23.054(18.058-29.431) | <0.001 |
| AFP-L3,％(L) | 0.063(0.045-0.082) | 1.065(1.046-1.085) | <0.001 |
| Constant | -8.654 |
Wherein, the corresponding Chinese names and abbreviations in English in Table 2 are as follows:
(1) gender, G; (2) age, A; (3) AFP-L3, alpha fetoprotein heteroplasmon, L; (4) AFP, alpha-fetoprotein, a; (5) DCP, also known as PIVKA-II, abnormal prothrombin, D.
The results show that: the model formula of the liver cancer diagnosis model C-GALAD is as follows:
C-GALAD＝-8.654+1.329×Gender+0.044×Age+0.063×AFP-L3+0.885×Log10(AFP)+3.138×Log10(DCP). Wherein the coefficient 95% confidence intervals of the parameters are sex (Gender) G:1.142-1.516, age (age) A:0.037-0.052, alpha-fetoprotein heteroplasmon (AFP-L3) L: 0.045-0.082, Alpha Fetoprotein (AFP) A:0.750-1.020, abnormal prothrombin (DCP or PIVKA-II) D: 2.894-3.382, Gender (Gender) 1 for men and 0 for women. It should be noted that the coefficient of variation of the model formula is a range of values.
Example 2 establishment of liver cancer diagnostic model LAD
The model parameters selected in this example are liver cancer serology markers alpha fetoprotein heteroplasmon (L), alpha fetoprotein (a), and abnormal prothrombin (D), and then a multi-factor binary Logistic regression analysis is performed on all inclusion indexes by using a binary Logistic regression, and on this basis, a multi-factor Logistic regression model LAD based on liver cancer triple joint test (AFP, AFP-L3, DCP) is established, and the results are shown in table 3.
TABLE 3 Multi-factor Logistic regression model LAD
| Variable (Variable) | β(95％CI) | OR(95％CI) | P value |
| Log10AFP,ng/mL(A) | 0.892(0.759-1.025) | 2.439(2.136-2.786) | <0.001 |
| Log10DCP,mAU/mL(D) | 3.524(3.277-3.772) | 33.931(26.488-43.464) | <0.001 |
| AFP-L3,％(L) | 0.057(0.039-0.075) | 1.059(1.040-1.078) | <0.001 |
| Constant | -5.930 |
Wherein, the corresponding Chinese names and abbreviations in English in Table 3 are as follows:
(1) AFP-L3, alpha fetoprotein heteroplasmon, L; (2) AFP, alpha-fetoprotein, a; (3) DCP, also known as PIVKA-II, abnormal prothrombin, D.
The results show that: the model formula of the liver cancer diagnosis model LAD is as follows:
LAD＝-5.930+0.057×AFP-L3+0.892*Log10(AFP)+3.524×Log10(DCP). Wherein the coefficient 95% confidence intervals of the parameters are AFP-L3: 0.039-0.075, AFP: 0.759-1.025, DCP: 3.277-3.772. It should be noted that the coefficient of variation of the model formula is a range of values.
Example 3 diagnostic efficiency validation
(1) Diagnosis efficiency verification of liver cancer diagnosis model C-GALAD on primary hepatocellular carcinoma patient
According to the maximum principle of Youden index (Sensitivity + Specificity-1), 0.9382 is taken as cut-off value. In the training set, the diagnostic sensitivity of the C-GALAD model for primary hepatocellular carcinoma patients was 86.9%, the specificity was 90.0%, the accuracy was 87.6%, and the area under the curve was 0.952[ 95% CI (0.947-0.957) ]; in the validation group, the diagnosis sensitivity of the C-GALAD model to the primary hepatocellular carcinoma patients is 86.9%, the specificity is 80.2%, the accuracy is 85.2%, the area under the curve is 0.908[ 95% CI (0.889-0.926) ] (see Table 4 and FIG. 1 and FIG. 2), and the model cut-off value is a series of values between 0.5268 and 1.1734, the Youden index is almost the same, the principle of the maximum Youden index is met, and the sensitivity and the specificity of the model for diagnosing HCC are close, so that the conclusion is drawn that the model cut-off value is between 0.5268 and 1.1734, and the model diagnosis efficiency is equivalent.
(2) Verification of diagnosis efficiency of liver cancer diagnosis model LAD on primary hepatocellular carcinoma patient
With 0.3320 as cut-off value, in the training group, the diagnosis sensitivity of the LAD model to primary hepatocellular carcinoma patients was 89.5%, the specificity was 86.7%, the accuracy was 88.9%, and the area under the curve was 0.943[ 95% CI (0.938-0.948) ]; in the validation group, the diagnosis sensitivity of the LAD model to the primary hepatocellular carcinoma patients is 87.9%, the specificity is 73.9%, the accuracy is 86.4%, the area under the curve is 0.905[ 95% CI (0.886-0.924) ] (see Table 4 and FIG. 3 and FIG. 4), and the model cut-off value is a series of values between 0.1455 and 0.9901, the Youden index is almost the same, the principle of Youden index maximum is met, and the sensitivity and specificity of the model for diagnosing HCC are close, so that the conclusion is drawn that the model cut-off value is between 0.1455-0.9901, and the model diagnosis efficiency is equivalent.
(3) According to different cut-off values, the diagnosis efficiency of the primary hepatocellular carcinoma in a training group and a verification group after platelets (P), total bilirubin (T) and albumin (A) are respectively eliminated, and the diagnosis efficiency of the primary hepatocellular carcinoma in the training group and the verification group after the gender (G) and the age (A) are respectively eliminated in a C-GALAD model is analyzed. (see FIGS. 1-8 and Table 4).
The results show that the C-GALAD and LAD models are simpler and more easily obtained in contained parameters and higher in diagnosis efficiency, and the contained parameters can be obtained by a single detection platform, so that the clinical accessibility is strong, the interference factors are few, and the application prospect is wider.
TABLE 4 comparison of diagnostic value of GAPTALAD, C-GALAD, LAD models to HCC in training and validation groups after eliminating different source parameters respectively
The foregoing is merely an example of the embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (10)
1. A method for establishing a liver cancer diagnosis model C-GALAD based on liver cancer triple-joint detection is characterized by comprising the following steps:
establishing a primary hepatocellular carcinoma clinical representation and a primary hepatocellular carcinoma database of laboratory data, and collecting laboratory indexes of a patient;
step two, collating laboratory indexes, establishing a multi-factor Logistic regression model, and obtaining a liver cancer diagnosis model C-GALAD;
wherein the laboratory indexes comprise basic information sex G, age A, liver cancer serology markers alpha-fetoprotein heteroplasmon L, alpha-fetoprotein A and abnormal prothrombin D of the patient.
2. The method for establishing the liver cancer diagnosis model C-GALAD based on liver cancer triple-detection as claimed in claim 1, wherein the model formula of the liver cancer diagnosis model C-GALAD is as follows:
C-GALAD＝-8.654+A×Gender+B×Age+C×AFP-L3+D×Log10(AFP)+E×Log10(DCP)；
wherein the coefficient A is 1.142-1.516; the coefficient B is 0.037-0.052; the coefficient C is 0.045-0.082; the coefficient D is 0.750-1.020; the coefficient E is 2.894-3.382.
3. The method for establishing a liver cancer diagnosis model C-GALAD based on triple-link detection of liver cancer according to claim 2, wherein the coefficient a is 1.329; the coefficient B is 0.044; the coefficient C is 0.063; the coefficient D is 0.885; the coefficient E is 3.138.
4. The method of claim 2, wherein the liver cancer diagnosis model C-GALAD has a cut-off value in the range of 0.5268-1.1734.
5. The method of claim 4, wherein the liver cancer diagnosis model C-GALAD has a cut-off value of 0.9382.
6. A method for establishing a liver cancer diagnosis model LAD based on liver cancer triple-joint detection is characterized by comprising the following steps:
establishing a primary hepatocellular carcinoma clinical representation and a primary hepatocellular carcinoma database of laboratory data, and collecting laboratory indexes of a patient;
step two, collating laboratory indexes, establishing a multi-factor Logistic regression model, and obtaining a liver cancer diagnosis model LAD;
wherein the laboratory indexes are liver cancer serological markers alpha fetoprotein heteroplasmon L, alpha fetoprotein A and abnormal prothrombin D.
7. The method for establishing the liver cancer diagnosis model LAD based on liver cancer triple-joint detection as claimed in claim 6, wherein the model formula of the liver cancer diagnosis model LAD is as follows:
LAD＝-5.930+H×AFP-L3+Y*Log10(AFP)+Z×Log10(DCP)；
wherein the coefficient H is 0.039-0.075, the coefficient Y is 0.759-1.025, and the coefficient Z is 3.277-3.772.
8. The method for establishing a liver cancer diagnosis model LAD based on liver cancer triple-joint detection according to claim 7, wherein the coefficient H is 0.057; the coefficient Y is 0.892; the factor Z is 3.524.
9. The method for establishing a liver cancer diagnosis model LAD based on triple-joint detection of liver cancer according to claim 6, wherein the value of the liver cancer diagnosis model LAD is cut-off value in the range of 0.1455-0.9901.
10. The method for establishing a liver cancer diagnosis model LAD based on liver cancer triple-joint detection according to claim 9, wherein the liver cancer diagnosis model LAD has a cut-off value of 0.3320.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011202929.4A CN112331333A (en) | 2020-11-02 | 2020-11-02 | Method for establishing liver cancer diagnosis model based on liver cancer triple-joint inspection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011202929.4A CN112331333A (en) | 2020-11-02 | 2020-11-02 | Method for establishing liver cancer diagnosis model based on liver cancer triple-joint inspection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112331333A true CN112331333A (en) | 2021-02-05 |
Family
ID=74324129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011202929.4A Pending CN112331333A (en) | 2020-11-02 | 2020-11-02 | Method for establishing liver cancer diagnosis model based on liver cancer triple-joint inspection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112331333A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113436677A (en) * | 2021-08-13 | 2021-09-24 | 北京大学 | System and method for testing possibility of liver cancer of subject |
| CN114121147A (en) * | 2021-12-10 | 2022-03-01 | 四川大学华西医院 | Nerve cell adhesion molecule detection system |
| CN115792234A (en) * | 2022-08-30 | 2023-03-14 | 杭州普望生物技术有限公司 | Marker combination for liver disease detection and application thereof |
| WO2023082139A1 (en) * | 2021-11-11 | 2023-05-19 | 华大数极生物科技(深圳)有限公司 | Nucleic acid for diagnosing liver cancer and protein detection kit |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109797222A (en) * | 2019-03-18 | 2019-05-24 | 刘卫辉 | A kind of diagnosing cancer of liver marker, detection primer, the purposes of kit and diagnosing cancer of liver marker |
| CN110993098A (en) * | 2019-12-06 | 2020-04-10 | 高春芳 | Establishment and application of novel blood multi-index liver cancer diagnosis model (GAP-TALAD) |
-
2020
- 2020-11-02 CN CN202011202929.4A patent/CN112331333A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109797222A (en) * | 2019-03-18 | 2019-05-24 | 刘卫辉 | A kind of diagnosing cancer of liver marker, detection primer, the purposes of kit and diagnosing cancer of liver marker |
| CN110993098A (en) * | 2019-12-06 | 2020-04-10 | 高春芳 | Establishment and application of novel blood multi-index liver cancer diagnosis model (GAP-TALAD) |
Non-Patent Citations (2)
| Title |
|---|
| MIAOXIA LIU 等: "Validation of the GALAD Model and Establishment of GAAP Model for Diagnosis of Hepatocellular Carcinoma in Chinese Patients", 《JOURNAL OF HEPATOCELLULAR CARCINOMA》 * |
| 林清标等: "GALAD模型在肝细胞患者中的诊断价值验证", 《中华肝脏病杂志》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113436677A (en) * | 2021-08-13 | 2021-09-24 | 北京大学 | System and method for testing possibility of liver cancer of subject |
| WO2023082139A1 (en) * | 2021-11-11 | 2023-05-19 | 华大数极生物科技(深圳)有限公司 | Nucleic acid for diagnosing liver cancer and protein detection kit |
| CN114121147A (en) * | 2021-12-10 | 2022-03-01 | 四川大学华西医院 | Nerve cell adhesion molecule detection system |
| CN114121147B (en) * | 2021-12-10 | 2023-06-30 | 四川大学华西医院 | Neural cell adhesion molecule detection system |
| CN115792234A (en) * | 2022-08-30 | 2023-03-14 | 杭州普望生物技术有限公司 | Marker combination for liver disease detection and application thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112331333A (en) | Method for establishing liver cancer diagnosis model based on liver cancer triple-joint inspection | |
| Di Tommaso et al. | Diagnostic accuracy of clathrin heavy chain staining in a marker panel for the diagnosis of small hepatocellular carcinoma | |
| Gan et al. | Diagnostic value of alpha-L-fucosidase for hepatocellular carcinoma: a meta-analysis | |
| Yoo et al. | Korean multi-center cancer cohort study including a biological materials bank (KMCC-I) | |
| CN104094116A (en) | Methods for detecting 5t4-positive circulating tumor cells and methods of diagnosis of 5t4-positive cancer in a mammalian subject | |
| CN111584082A (en) | Establishment and application of primary hepatocellular carcinoma microvascular invasion regression prediction model based on clinical examination multidimensional data | |
| CN110993098A (en) | Establishment and application of novel blood multi-index liver cancer diagnosis model (GAP-TALAD) | |
| CN106461664A (en) | Circulating tumor cell diagnostics for lung cancer | |
| CN107024586A (en) | Method based on artificial neural network tumor-marker joint-detection auxiliary diagnosis liver cancer | |
| CN109457032A (en) | Thyroid cancer molecule diagnosis kit | |
| CN113539498A (en) | Decision tree model-based system for predicting malignant risk of isolated pulmonary nodules | |
| Yan et al. | Total bile acid-to-cholesterol ratio as a novel noninvasive marker for significant liver fibrosis and cirrhosis in patients with non-cholestatic chronic hepatitis B virus infection | |
| Kocjan | Evaluation of the cost effectiveness of establishing a fine needle aspiration cytology clinic in a hospital out‐patient department | |
| CN114864080A (en) | Method, system, equipment and medium for establishing liver cancer diagnosis model C-GALAD II | |
| JP2010243406A (en) | Method for detecting extent of clinical condition of liver cancer and chronic liver disease, using discriminant function taking measurement values of afp and pivka-ii as characteristic values | |
| CN114155956A (en) | System for predicting blood vessel invasion probability of primary liver cancer patient incapable of being resected by surgery | |
| CN110148471A (en) | Liver and gallbladder tumor patient survival rate prediction data collection system and model building method | |
| Ma et al. | Comparison of four markers of hepatic fibrosis and hepatic function indices in patients with liver cirrhosis and hepatoma | |
| CN107850526A (en) | Assess the method for cell breast samples and the composition for putting into practice methods described | |
| Lee et al. | Complete blood count reflects the degree of oesophageal varices and liver fibrosis in virus‐related chronic liver disease patients | |
| CN108982874A (en) | It is a kind of detect human prostate cancer antigen PSA, AR-V7 immunofluorescent reagent box and application | |
| El-Mezayen et al. | Diagnostic Performance of Circulating Tumor Cells for Predicting of Hepatocellular Carcinoma in Hepatitis C Virus-High Risk Patients: Role of Liquid Biopsy | |
| CN110954701B (en) | Diagnostic kit for hepatic fibrosis or cirrhosis | |
| Ponce-Zepeda et al. | Touch preparation for rapid onsite evaluations of renal mass biopsies: concordance rate, pearls, and pitfalls | |
| Shen et al. | Reliability of combined fine needle aspiration and core needle biopsies in the diagnosis of liver lesions: An 8‐year institutional experience |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210205 |