Skip to main content

Table 9 A comparison of the classification performance obtained by dSPIC-AM with and without pooling layers on the testing samples in dataset D_2

From: dSPIC: a deep SPECT image classification network for automated multi-disease, multi-lesion diagnosis

Classifer

Accuracy

Precision

Sensitivity

Specificity

F-1 score

AUC

dSPIC-AM with pooling

0.7747

0.7883

0.7863

0.8820

0.7860

0.9272

dSPIC-AM without pooling

0.7558

0.7561

0.7740

0.8642

0.7649

0.9060