Disease risk prediction modeling method based on LI-RADS classification

Abstract

The invention relates to a disease risk prediction modeling method based on LI-RADS classification. The method comprises the following steps: collecting sample data; converting a medical imaging examination result into a quantized end point event state, and setting LR-3 and above LI-RADS classification as an end point event; taking a plurality of physical examination index items mainly based on the detection result of serum cytoplasm thymidine kinase 1, namely TK1, as independent variables; and carrying out modeling through a regression method so as to obtain a hepatocellular carcinoma (HCC) disease risk prediction model. According to the method, the iconography examination result set with the determined HCC risk is set as the risk prediction end point event, so that the acquisition cost and difficulty of the positive result sample are reduced, the stable accumulation of the positive sample becomes possible, and the iteration upgrading feasibility of the risk prediction model obtained by the system is higher in the same time period.

Description

technical field [0001] The present application relates to the field of disease risk prediction modeling, in particular to a disease risk prediction modeling method based on LI-RADS classification. Background technique [0002] Liver cancer is one of the most common malignant tumors. According to statistics in 2018, the global liver cancer patients accounted for 4.7% of the total cancer population, and the mortality rate was as high as 8.2%, especially among male cancer patients (the mortality rate was 10.2%) and lung cancer ranked the top two in terms of mortality rate. Hepatocellular carcinoma (HCC) is the main type of primary liver cancer, accounting for 85%-90% of the total incidence. The causes of HCC are complex, with no typical symptoms in the early stage, rapid pathological progress, and significant differences in prognosis at different stages. The number of primary liver cancer patients in China accounts for about 50% of the world, and the number is increasing year...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the existing HCC risk assessment system is not suitable for healthy groups to monitor the risk of hepatocellular carcinoma, and the cost of sample acquisition is high, difficult, and the efficiency of sample accumulation is low, which makes it difficult to iteratively upgrade the risk assessment model. This application provides a method for predicting and modeling disease risk based on LI-RADS grading

Images