Computational Intelligence and Metaheuristic Techniques for Brain Tumor Detection through IoMT-Enabled MRI Devices

Table 2

The related work of the deep learning approaches for brain tumor.

Authors	Dataset	Access	Techniques	Limitation
Jasm, D et al. [7]	MRI dataset	Open	Image mining techniques, neural network	The techniques did not work for video database.
Polat, O et al. [9]	T1-weighted MRI images	Open	VGG16, VGG19, ResNet50, DenseNet21, Adadelta optimizer	The training process of the VGG net was very slow and also had complex architecture.
Deep, A. et al. [11]	BRATS dataset	Open	Neural network, adaptive firefly, kernel principal component analysis	The techniques dealt with computational complexity, time complexity, and feature selection complexity.
Mohsen, H et al. [10]	MRI dataset	Open	DWT, deep neural network, CNN	Training of neural network had been time-consuming as it needed a large-size dataset for training purpose.
Rani, P et al. [13]	Figshare dataset	Open	Deep neural network, R-CNN, ChanVese algorithm	The algorithm segmented the unwanted regions in the brain MRI.
Pernas, F et al. [20]	MRI dataset	Open	Deep convolution neural network, sliding window technique	The proposed work needed prior information about the images and took high computational time.