Predicting liver cancers using skewed epidemiological data

作者：

Highlights：

• An artificial intelligence model that can predict the occurrence of liver cancer within five years is exploited.

• The model considers 84 risk factors, far more than those considered by relevant studies.

• Two undersampling algorithms for processing extremely skewed data are proposed.

• An averaged area under the curve (AUC) of 0.76 has been achieved.

摘要

•An artificial intelligence model that can predict the occurrence of liver cancer within five years is exploited.•The model considers 84 risk factors, far more than those considered by relevant studies.•Two undersampling algorithms for processing extremely skewed data are proposed.•An averaged area under the curve (AUC) of 0.76 has been achieved.

论文关键词：Liver cancer,Cancer prediction,Risk assessment,Machine learning,Clustering

论文评审过程：Received 14 December 2020, Revised 26 November 2021, Accepted 21 December 2021, Available online 6 January 2022, Version of Record 8 January 2022.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102234