CONVOLUTION NEURAL NETWORK FOR BRAIN TUMOR SEGMENTATION

Computer Engineering

Magnetic Resonance Imaging (MRI) has grown in popularity in recent years, particularly in the field of medical technology, for the diagnosis of brain tumors. Only using MR imaging to detect and partition brain cancers requires a significant amount of time and effort on the side of the involved staff, as well as a high level of competence. This supports the requirement for developing an independent model for brain tumor diagnosis. In our work, we develop a convolutional neural network (CNN) for MR image-based brain tumor diagnosis. The dataset used in this study includes 253 brain MR pictures, of which 155 are said to include malignancies. Our approach has a 96% overall accuracy in identifying the MR pictures with malignancies. In the test dataset, the model performed better than the currently used traditional approaches for the diagnosis of brain tumors (Precision = 0.97, Recall = 0.96, and F1-score = 0.96). In order to speed up the treatment process, clinical specialists may find it useful to use the proposed model to determine whether the patient has a brain tumor.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.