System for deducing cancer risk probability by using multi-modal risk factors

Abstract

The invention discloses a system for deducing a cancer risk probability by using multi-modal risk factors, and the system comprises a data collection module which is used for obtaining a risk feature data set of various cancer high-risk groups and a data set of main risk factors affecting the onset of various cancers, and forming a multi-modal data set; a feature extraction module which is used for preprocessing the multi-modal data set and performing risk feature extraction on the preprocessed multi-modal data set; a feature fusion module which is used for determining a multi-modal fusion mode of the normalized mutual information to obtain fusion feature parameters; a model building module which is used for building a risk assessment model by using a Logistic regression analysis method; and a risk assessment module which is used for deducing the risk probabilities of various cancers through the risk assessment model. According to the method, the risk assessment model of the cancer incidence risk factors of the cancer high-risk group is constructed, so the automatic cancer risk assessment and quantitative expression of the early cancer risk based on cross-modal fusion are realized.

Description

technical field [0001] The invention belongs to the field of disease prediction, and in particular relates to a system for inferring cancer risk probability by using multimodal risk factors. Background technique [0002] Unlike infectious diseases, which have a definite cause, the pathogenesis of cancer is diverse and often has complex and comprehensive effects. Most cancers are very difficult to prevent and the risk of cancer is difficult to quantify. In particular, the degree of early warning of different tumors is different. Therefore, it is very important to formulate corresponding screening indicators, establish a scientific evaluation system and risk perception indicators. Contents of the invention [0003] In view of this, the present invention proposes a system for inferring cancer risk probability by using multimodal risk factors, which is used to solve the problem that early cancer risk cannot be effectively quantified. [0004] In the first aspect of the pres...

AI Technical Summary

Problems solved by technology

[0003] In view of this, the present invention proposes a system for inferring cancer risk probability by using multimodal risk factors to solve the problem that early cancer risk cannot be effectively quantified

Images