Table 2 Mean AUC for three ML algorithms, support vector machines classifier (SVM), neural network (NN), and random forest (RF), trained on the sets of features from three stains, H&E (n=12), ERG (n=13), and PIN-4 (n=16), see text

From: Evaluation of automatic discrimination between benign and malignant prostate tissue in the era of high precision digital pathology

	SVM		RF		NN
	Default	Tuned	Default	Tuned	Default	Tuned
H&E	\(0.80 \pm 0.07\)	\(0.90 \pm 0.06\)	\(0.82 \pm 0.08\)	\(0.82 \pm 0.08\)	\(0.84 \pm 0.07\)	\(0.89 \pm 0.07\)
ERG	\(0.83 \pm 0.06\)	\(0.84 \pm 0.07\)	\(0.85 \pm 0.06\)	\(0.85 \pm 0.05\)	\(0.85 \pm 0.06\)	\(0.86 \pm 0.05\)
PIN-4	\(0.92 \pm 0.05\)	\(0.93 \pm 0.06\)	\(0.91 \pm 0.06\)	\(0.92 \pm 0.05\)	\(0.93 \pm 0.04\)	\(\mathbf {0.94} \pm 0.04\)

Back to article page