Brain MRI detection by Sematic Segmentation models- Transfer Learning approach

TL;DR

This paper explores the application of transfer learning with CNN models like VGG16, ResNet50, and ResU-net for brain tumor detection in MRI scans, highlighting ResNet50's superior performance despite common imaging challenges.

Contribution

It introduces the use of transfer learning with specific CNN architectures for brain MRI segmentation and evaluates their effectiveness in tumor detection.

Findings

ResNet50 achieved high accuracy and F1 score.

Transfer learning improves segmentation performance.

ResNet50 outperforms VGG16 and ResU-net in this task.

Abstract

The paper discusses the use of MRI for segmentation techniques, specifically focusing on brain tumor detection. It discusses the use of convolutional neural networks (CNN) for automatic segmentation but also discusses challenges such as non-isotropic resolution, Rician noise, and bias field effects. The paper proposes models like VGG16, ResNet50, and ResU-net to predict MRI images based on original and predicted masks. ResNet50 is found to be a promising model with high accuracy and F1 score.