Abstract
Brain tumor is regarded as most severe and aggressive medical condition which shortens life of patients and accurate diagnoses and detection of the condition is vital in curing the disease. The availability of computer assisted diagnoses procedures have improved tumor detection techniques, however, manual classification of brain tumor are often hindered by the vast quantity of MRI data, moreover, the accurate diagnoses are limited to few images. This necessitated the introduction of automatic classification system that can lower mortality rate from brain tumor. This review explores the efficacy of Convolutional Neural Networks (CNN) in classifying brain tumor. The CNN is trained on a larger dataset that consists of pre processed images which corresponds to malignant tumor type. After the completion of training, the CNN is capable of generating a probability map which indicates the presence of malignant tumor. Future scope lies within the advancement that will happen in the creation of novel architectures and inclusion of numerous data obtained from various sources along with the introduction of accepted AI techniques. By finding solutions to these, neural networks can ensure an accurate, and personalized detection of brain tumor BTC, while contributing towards an improved clinical outcomes and improving the understandings about brain pathology.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2024 African Journal of Biomedical Research