Indian Journal of Science and Technology
Year: 2023, Volume: 16, Issue: 37, Pages: 3129-3138
Original Article
Vipin Saxena1*, Sugandha Singh1, Karm Veer Singh1
1Department of Computer Science, Babasaheb Bhimrao Ambedkar University, 226025, Lucknow, India
*Corresponding Author
Email: [email protected]
Received Date:25 February 2023, Accepted Date:29 August 2023, Published Date:09 October 2023
Objective: The present research work is focused on brain tumor classification, prediction and to increase the performance to locate the tumor region. Methods: A two-dimensional Convolutional Neural Network (CNN) model is proposed to classify the Magnetic Resonance Images (MRI) into tumor and nontumor categories. The method is applied on a collected dataset consisting of 2056 MRI images. The model is implemented in Python with hyperparameter tuning and activation functions.Findings: In this paper, ReLU and LeakyReLU activation functions are applied with several optimizers. The analysis of the implemented results has been used to gauge performance accuracy. The computed results achieve 99.51% accuracy for predicting the brain tumor using LeakyReLU with Adam optimizer. Novelty: The proposed model provides quick, and accurate approach to classify patients by setting hyperparameter tuning parameters which helps to the doctor to detect patients suffering with tumor and the entire process reduces the computation time.
Keywords: Convolutional Neural Network (CNN); Magnetic Resonance Image (MRI); Digital Imaging and Communications in Medicine (DICOM); Brain Tumor; Deep Learning
© 2023 Saxena et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)
Subscribe now for latest articles and news.