Determination of Risk Factors Associated with the Failure of 12 Weeks of Direct-Acting Antiviral Therapy in Patients with Hepatitis C: A Prospective Study

Table 3

Comparison of clinical models based on six machine learning classifiers in predicting therapeutic failure among patients with HCV.

Model	No of variables	AUCROC	95% CI (Delong)	Pseudo ^2	BIC	_value (>chi)
Model 1	17	0.979	0.961-0.997	0.89	114
Model 2	16	0.982	0.965-0.998	0.88	110.6	0.18224
Model 3	15	0.982	0.965-0.997	0.86	109.2	0.05621
Model 4	13	0.98	0.962-0.996	0.84	102.2	0.19101
Model 5	12	0.984	0.968-0.998	0.82	101.1	0.04548
Model 6	11	0.986	0.971-0.999	0.8	101.1	0.02378

AUC_ROC: area under the receiver operating characteristic curve; BIC: Bayesian Information Criterion.