Local and Deep Features Based Convolutional Neural Network Frameworks for Brain MRI Anomaly Detection

Table 5

Comparison of the proposed approach with state-of-the-art algorithms.

Methods	KACD	Brain tumor
RBFNN [19]	0.92	0.94
CNN [21]	0.80	0.87
MobileNet-ELM-CBA [12]	0.88	0.94
MobileNet-SNN-CBA [12]	0.82	0.92
BrainMRNet [13]	0.94	0.96
ResNet-50 (augmentation) [11]	0.93	0.95
The naive bayes with ELM [20]	0.81	0.84
DSRCN (our approach)	0.95	0.96
Multibranch (our approach)	0.93	0.88
DBP-DAE (our approach)	0.85	0.90