Medical Image Retrieval Using Empirical Mode Decomposition with Deep Convolutional Neural Network

Table 1

The details of deep CNN used for medical image retrieval.

Layer name	Output size	Layer
Conv1		, 64, stride 2
Max pooling		stride 2
Residual attention unit
		Attention mask
Residual attention unit
		Attention mask
Residual attention unit
		Attention mask
Residual unit
Conv2		, 32, stride 1, padding: same
Max pooling		stride 1
Average pooling		stride 1
FC, softmax, Center loss	116
Trunk depth	57