A novel network-level fused deep learning architecture with shallow neural network classifier for gastrointestinal cancer classification from wireless capsule endoscopy images

Table 7 Comparison of the proposed network-level fusion architecture with SOTA pre-trained models based on the accuracy of the selected datasets of this work

Network	Kvasir V1 (%)	Kvasir VII (%)	Parameters (M)	Parameter memory (PM)
Proposed	99.60	95.10	14.7	51 MB
Inception V3	96.40	92.86	23.9	91 MB
DenseNet201	95.10	93.04	20.0	77 MB
Resnet18	94.50	91.28	11.7	45 MB
Resnet50	94.86	90.20	25.5	98 MB
Resnet101	92.36	90.14	44.6	171 MB
Xception	94.90	93.68	22.9	88 MB
NasnetLarge	95.60	93.60	88.9	340 MB

ISSN: 1472-6947