VGG16-based Brain Stroke Prediction in CT Scan

Authors: Rudrani Mahajan, Akshay Jain, Shubham Tiwari, Chetan Yeshod, Gayatri Deore

Country: India

Full-text Research PDF File:   View   |   Download

Abstract: Brain strokes pose a significant global health challenge, ranking as the second leading cause of mortality worldwide. In India, reports indicate a frequency of three stroke incidents per minute. Ischemic strokes, resulting from arterial blockages, constitute about 80% of cases, with hemorrhagic strokes, caused by blood vessel ruptures, making up the remaining 20%. This study explores the integration of Generative Adversarial Networks (GANs) and the VGG16 model for brain stroke detection. GANs, known for their ability to generate diverse medical images, are used to augment the training datasets for machine learning models. The VGG16 architecture, recognized for its deep convolutional layers, is employed for robust feature extraction, crucial for stroke identification. By combining GAN-synthesized data with VGG16-based classification, the proposed methodology aims to enhance the accuracy of stroke detection from brain scans. This innovative approach holds promise for improving early stroke detection, enabling timely intervention, and ultimately enhancing patient outcomes.

Keywords: Brain Stroke, Machine Learning, Generative Adversarial Networks, Visual Geometry Group 16

Paper Id: 230629

Published On: 2024-05-12

Published In: Volume 12, Issue 3, May-June 2024

Cite This: VGG16-based Brain Stroke Prediction in CT Scan - Rudrani Mahajan, Akshay Jain, Shubham Tiwari, Chetan Yeshod, Gayatri Deore - IJIRMPS Volume 12, Issue 3, May-June 2024.

Share this