Recent Papers

Brain stroke is a life-threatening medical condition that requires rapid and accurate diagnosis to ensure timely treatment and reduce mortality rates. Traditional diagnostic approaches often rely heavily on manual interpretation of computed tomography (CT) images, which may lead to delayed or inconsistent results. To address these challenges, this research proposes a Brain Stroke Detection System based on CT images using Deep Learning. The system integrates two deep learning architectures Convolutional Neural Networks (CNNs) for robust image feature extraction and Long Short-Term Memory (LSTM) networks for capturing sequential dependencies. A dataset of 2,501 CT images, comprising 1,551 normal scans and 950 stroke-affected scans, was used to train and evaluate the models. Experimental results demonstrate that the CNN model achieved a training accuracy of 99.00% and a validation accuracy of 98.00%, outperforming the LSTM model, which achieved a training accuracy of 99.00% and a validation accuracy of 95.00%. The system was developed using Python with Flask as the backend framework and an intuitive web-based interface built using HTML, CSS, and JavaScript, making it scalable and user-friendly. The findings highlight the effectiveness of CNN-based deep learning models in stroke detection and their potential to provide clinicians with a reliable diagnostic support tool for early and precise identification of brain strokes.

Copyright © 2025 Harsha Kumar V. This is an open access article distributed under the Creative Commons Attribution License.