Method for auxiliary diagnosis of liver cancer by means of combined detection of tumor markers based on artificial neural network

Abstract

The invention discloses a method for auxiliary diagnosis of liver cancer by means of combined detection of tumor markers based on an artificial neural network. The method comprises the following steps: collecting a blood sample of a patient; adopting a chemiluminesent immunoassay kit to respectively measure the contents of alpha fetoprotein (AFP), carcino-embryonic antigen (CEA) and carbohydrate antigen 125 (CA125) in serum; measuring the level of sialic acid (SA) in the serum by applying a visible spectrophotometry; measuring the level of Ca in the serum by using an azo arsenic III end-point method; carrying out descriptive statistical analysis on all data by using statistics, and taking diagnostic sensitivity, specificity, accuracy, a positive predictive value and a negative predictive value as evaluation indexes; adopting a back-propagation neural network algorithm to obtain a trained model; using the trained model to predict a corresponding test set. The method provided by the invention is high in accuracy rate and can be used for well distinguishing the liver cancer, benign and normal, thus having good popularization and application prospects.

Description

technical field [0001] The invention belongs to the technical field of liver cancer detection, and in particular relates to a method for auxiliary diagnosis of liver cancer based on artificial neural network combined detection of tumor markers. Background technique [0002] Primary liver cancer is one of the most common malignant tumors in my country, with high morbidity and mortality, and most of them are primary hepatocellular carcinoma. The main reason is that there are many hepatitis B patients in my country, and most primary liver cancers develop on the basis of chronic hepatitis such as hepatitis B or C. At present, people's consensus on primary liver cancer is early detection, early diagnosis and early treatment. [0003] At present, there are mainly the following detection methods: [0004] 1) The serological indicator that is widely used in clinical practice is alpha-fetoprotein. It is generally believed that when its content is greater than 400 μg / L, it may indic...

AI Technical Summary

Problems solved by technology

[0004] 1) The serological indicator that is widely used in clinical practice is alpha-fetoprotein. It is generally believed that when its content is greater than 400 μg / L, it may indicate the existence of liver cancer, and the possibility of liver cancer should be highly suspected, but its diagnostic sensitivity and specificity are limited.

[0005] 2), detection techniques such as ultrasound, computerized X-ray tomography, magnetic resonance imaging, digital subtraction angiography, hepatic arteriography, and liver pathological puncture are also widely used clinically, but these techniques are not effective for the early diagnosis of primary liver cancer. The ability is limited, and most of them have ray radiation and cause certain trauma to the patient's body

This brings certain difficulties to the diagnosis of tumors, but it brings the possibility for the joint detection of several markers to improve the diagnosis rate of tumors

Images