Table 1.

3D-CNN Architecture for extraction of deep learning features.

Layers	Parameter Setting
Conv 3D-1	size = 5 × 5 × 5; stride = 1; zero-padded
Relu-1	Alpha = 0.2
Max Pool 3D-1	size = 4 × 4 × 4; stride = 4; zero-padded
Conv 3D-2	size = 5 × 5 × 5; stride = 1; zero-padded
Relu-2	Alpha = 0.2
Max Pool 3D-2	size = 4 × 4 × 4; stride = 4; zero-padded
Fully connected-1
Flat-1
Relu-3	Alpha = 0.2
Dropout-1	p = 0.5
Fully connected-2
SoftMax