Table 3.

Colorectal polyp detection model architecture.

Layers	Filters	Size/Stride	Output
Input			128 × 128
Convolutional	16	(3 × 3 + 3 × 1 + 1 × 3)/1	128 × 128
Batch Normalization			128 × 128
ReLu			128 × 128
Max pooling		2 × 2/2	64 × 64
Convolutional	32	(3 × 3 + 3 × 1 + 1 × 3)/1	64 × 64
Batch Normalization			64 × 64
ReLu			64 × 64
Max pooling		2×2/2	32 × 32
Convolutional	64	(3 × 3 + 3 × 1 + 1 × 3)/1	32 × 32
Batch Normalization			32 × 32
ReLu			32 × 32
Max pooling		2 × 2/2	16 × 16
Convolutional	128	(3 × 3 + 3 × 1 + 1 × 3)/1	16 × 16
Batch Normalization			16 × 16
ReLu			16 × 16
Convolutional	128	(3 × 3 + 3 × 1 + 1 × 3)/1	16 × 16
Batch Normalization			16 × 16
ReLu			16 × 16
Convolutional	128	(3 × 3 + 3 × 1 + 1 × 3)/1	16 × 16
Batch Normalization			16 × 16
ReLu			16 × 16
Convolutional	24	1 × 1/1	16 × 16
Transform			16 × 16
Output