Table 3.

Performance metrics of the CLARE-XR classification methodology.

Class	Accuracy		Precision		Recall		F1-score
	R1	R2	R1	R2	R1	R2	R1	R2
Ca	0.700	0.475	0.025	0.026	0.328	0.622	0.047	0.051
Pa	0.634	0.631	0.477	0.481	0.418	0.644	0.446	0.551
Pn	0.943	0.891	0.355	0.224	0.347	0.596	0.351	0.326
Pl	0.872	0.794	0.510	0.369	0.680	0.808	0.538	0.507
Ot	0.974	0.961	0.424	0.314	0.473	0.609	0.447	0.414
MEAN	0.825	0.750	0.358	0.283	0.449	0.656	0.375	0.370

Class	Accuracy		Precision		Recall		F1-score
	R3	R4	R3	R4	R3	R4	R3	R4
Ca	0.558	0.499	0.028	0.027	0.541	0.608	0.052	0.052
Pa	0.630	0.629	0.470	0.469	0.426	0.434	0.447	0.451
Pn	0.956	0.943	0.500	0.382	0.320	0.454	0.390	0.415
Pl	0.878	0.870	0.529	0.504	0.663	0.688	0.588	0.582
Ot	0.975	0.971	0.446	0.395	0.499	0.529	0.471	0.452
MEAN	0.799	0.782	0.394	0.355	0.490	0.543	0.390	0.390

Results obtained for all instances predicted as positive for a given class, regardless of whether they are single-pathology or multi-label. The precision, recall, and F1-score values were calculated based on the predictions obtained by each set of rules (R).