An Efficient Method for Diagnosing Brain Tumors Based on MRI Images Using Deep Convolutional Neural Networks

Table 4

Table of comparison results between algorithms before and after data normalization.

CNN architectures	Training accuracy		Validation accuracy		Time (s)
	Before data normalization	After data normalization	Before data normalization	After data normalization	Before data normalization	After data normalization
DenseNet201	99.87%	99.88%	91.63%	94.14%	3165	455
ResNet152V2	99.95%	99.95%	92.86%	93.31%	3639	927
MobileNetV3	99.66%	99.70%	88.70%	91.21%	1869	1398
VGG19	99.91%	99.91%	89.54%	92.88%	3165	2691