Prediction of the risk of developing hepatocellular carcinoma in health screening examinees: a Korean cohort study

Table 4 Performance of ensemble machine learning model for probability of HCC development within 9 years in subgroups of patients with preexisting medical conditions: alcoholic or non-alcoholic fatty liver disease, and diabetes mellitus

Model	Subset of selected risk factors	CV or OOB error in the training cohort	Validation in the test cohort
Model	Subset of selected risk factors	Value (±SD)	AUC (95% CI)
DM subgroup	Age, sex, ALT, GGT, CLD, HIV, Schizophrenia	0.873 (±0.006)	0.851 (0.794–0.863)
NAFLD subgroup	Age, Sex, Income, ALT, GGT, cholesterol, CLD, CVH, HIV	0.882 (±0.006)	0.853 (0.801–0.822)
AFLD subgroup	Age, Sex, FHx of CLD, ALT, cholesterol, CLD, CVH, HIV	0.874 (±0.006)	0.849 (0.837–0.861)

HCC hepatocellular carcinoma, CV cross validation, OOB out-of-bag, SD standard deviation, AUC area under receiver operating characteristics curve, CI confidence interval, DM diabetes mellitus, NAFLD non-alcoholic fatty liver disease, AFL alcoholic fatty liver disease, ALT alanine aminotransferase, GGT gamma-glutamyl transferase, CLD chronic liver disease, HIV human immunodeficiency virus, CVH chronic viral hepatitis, FHx family history

ISSN: 1471-2407