Method for establishing early lung cancer risk assessment model in combination with imaging analysis result

Abstract

The invention provides a method for establishing an early lung cancer risk assessment model in combination with an imaging analysis result and belongs to the field of computer aided medicine. On the basis of a traditional risk assessment model, an analysis result based on iconography is combined. A lung nodule benign and malignant judgment result based on CT images is used as one of dangerous factors. The lung nodule benign and malignant judgment result and other dangerous factors such as the smoking history and the family medical history are used as factors of the same level to establish an early-stage lung cancer risk assessment model. Therefore, the accuracy and the reliability of the prediction model are effectively improved.

Description

technical field [0001] The invention relates to the field of computer-aided medicine, and relates to a method for establishing an early lung cancer risk assessment model combined with imaging analysis results. Background technique [0002] In recent years, some studies on tumor risk assessment have been actively carried out in China, but most of them still use foreign models to assess the risk of cancer in Chinese people, such as using the Gail model to assess the risk of breast cancer in Chinese women. There are differences between the Chinese population and the European and American population in terms of physiological characteristics, living environment, and living habits. Therefore, the risk model developed based on the European and American population is not fully applicable to the Chinese population. At present, there is still a lack of specific malignant tumor risk assessment models suitable for the Chinese population. It is of great significance to build a tumor ris...

AI Technical Summary

Problems solved by technology

[0003] At present, domestic and foreign studies on chronic disease risk assessment models mainly have the following characteristics: (1) most of the studies are prospective cohort studies, and most of the observation time is 5 or 10 years; (2) the universality of the model is poor, and there are local (3) Most of the variables included in the model are unmodifiable factors (such as age, gender, family history, susceptibility, etc.), and lack of intervention indicators such as behavior habits, diet, and physical exercise; There are many studies on risk assessment of common chronic diseases such as vascular disease and diabetes, but relatively few studies on malignant tumors (except breast cancer); (5) There is a lack of cohort data in China, and most common chronic diseases lack specific risk assessment and prediction models

Images