Table 5.

Performance of proposed lobe-based radiomics method compared with other machine learning methods for COPD severity staging on Dataset1

Task	Classifier	Accuracy	Precision	Recall	F1-score	AUC	Training time (s)
Four categories	KNN	0.50	0.53	0.50	0.49	0.47	--
	Decision Tree	0.52	0.57	0.52	0.50	0.48	0.0349
	AdaBoost	0.51	0.51	0.51	0.50	0.44	0.2813
	Gradient Boosting	0.57	0.57	0.57	0.56	0.45	4.6552
	XGBoost	0.64	0.66	0.64	0.60	0.45	0.5206
	Random Forest	0.60	0.68	0.60	0.59	0.44	0.3022
	CatBoost	0.64	0.68	0.64	0.63	0.48	150.8
	Proposed method	0.63	0.71	0.63	0.62	0.49	0.0621
Two categories	KNN	0.80	0.80	0.80	0.79	0.87	--
	Decision Tree	0.73	0.73	0.73	0.73	0.72	0.0312
	AdaBoost	0.79	0.79	0.79	0.79	0.85	0.3504
	Gradient Boosting	0.84	0.84	0.84	0.84	0.89	1.5213
	XGBoost	0.82	0.82	0.81	0.82	0.89	0.1366
	Random Forest	0.83	0.83	0.82	0.83	0.89	0.3191
	CatBoost	0.84	0.84	0.84	0.84	0.88	64.7
	Proposed method	0.87	0.87	0.87	0.87	0.93	0.0440