[Retracted] BID-Net: An Automated System for Bone Invasion Detection Occurring at Stage T4 in Oral Squamous Carcinoma Using Deep Learning

Table 3

Hyper tune parameters of BID-Net for BI.

Type of layer	Size	Channel	Filter size	Activation function	No. of parameters
Convo1	222 × 222 × 32	32	3 × 3	Relu	896
Pool1	111 × 111 × 32	—	2 × 2	—	0
Convo2	109, 109, 32	32	3 × 3	Relu	9248
Pool2	54 × 54 × 32	—	2 × 2	—	0
Convo3	52 × 52 × 64	64	3 × 3	Relu	18 496
Pool3	26 × 26 × 64	–	2 × 2	–	0
Flatten	43 264	–	—	—	0
Dense 1	64	–	—	—	2768 960
Dropout	64				0
Dense 2	2	–	–	–	130