Efficacy of exponentiation method with a convolutional neural network for classifying lung nodules on CT images by malignancy level

Attention-guided deep neural network with a multichannel architecture for lung nodule classification

“…Despite the advantages of deep learning techniques in simplifying the procedure of lung nodule analysis, the accuracy of lung nodule classification still needs improvement. Takuma et al [ 14 ] evaluated the effectiveness of the exponentiation method for improving the performance of a CNN in the task of classifying lung nodules on CT images according to malignancy level. Experimental results improved accuracy of the CNN and the ability to adjust sensitivity and specificity by selecting the exponent value.…”