Table 3.

Evaluation indicators.

name	Formula	Meaning
PRE	PRE=TP+TN/TP+TN+FP+FN	Number of sample proportions with correct decisions
REC	REC=TP/TP+FN	The number of correctly predicted proportions in the normal sample
F-score	(α2+1) × PRE × REC/α2 × PRE+REC	Harmonic mean value of PRE and REC (α = 1 in this paper)
TNR	$\mathrm{T}\mathrm{N}\mathrm{R}=\sqrt{\mathrm{T}\mathrm{P}/\mathrm{T}\mathrm{P}+\mathrm{F}\mathrm{N}\times \mathrm{T}\mathrm{N}/\mathrm{T}\mathrm{N}+\mathrm{F}\mathrm{P}}$	Average predictive accuracy of normal and abnormal samples