Brain Tumor Classification Using Deep Learning Through MRI Images

A brain tumor is a dangerous neural illness produced by the strict growing of prison cell in the brain or head. The amount of persons suffering from brain tumor remains increasingly cumulative. Initial detection of wicked cancers is vital to provide cure to sickness, and early identification reduces the risk of death. If a brain cancer is not predicted in initial phase, it can assuredly cause to death. Hence, primary identification of brain tumors requires the usage of a mechanical means. The segmentation, analysis, and separation of unclean tumor parts from MRI images are the main source of anxiety. Nevertheless, the situation is a boring and slow procedure that radiologists or scientific professionals need to assume, and their act is only reliant on their knowledge. To report the segmented MRI images including tumor, the usage of computer-assisted methods come to be necessary. In this thesis, a Convolutional Neural Network (CNN) approach is used to identify brain cancers in MRI images. The presented model focuses on improving accuracy because there has been a significant amount of research in this sector. This investigation is carried out using Python and Google Colab. Two datasets are used for this study, namely Kaggle Brain MRI dataset and Figshare Brain MRI dataset. Models of deep CNN, consisting of VGG16, AlexNet, and ResNet, are utilized to extract deep features. The classification accuracies of the aforementioned deep learning models are used to measure the efficiencies of the implemented systems. For the Kaggle dataset, AlexNet achieves a 98% accuracy, VGG16 has 97% accuracy, and ResNet has 66% accuracy. Among these networks, AlexNet has provided the highest level of accuracy. In the Figshare dataset, AlexNet and VGG16 both achieve 99% accuracy, and ResNet has 96% accuracy. In terms of accuracy, AlexNet and VGG16 outperform ResNet. These performances aid in the early detection of cancers before they cause physical harm such as paralysis and other complications.