Regular Issue - 5
Stroke is one of the leading causes of disability and mortality worldwide, with ischemic and haemorrhagic strokes being the two primary types. Early and accurate detection of these stroke types from medical imaging, such as CT scans, is crucial for timely intervention. This study proposes a deep learning-based approach for automated classification of ischemic stroke, haemorrhagic stroke, and normal brain scans using a modified ResNet50V2 architecture enhanced with a Channel Attention Mechanism (CAM). The dataset, comprising CT scan images, was pre-processed and balanced using Synthetic Minority Over-sampling Technique (SMOTE) to address class imbalance. Data augmentation techniques were employed to improve model generalization. The proposed model utilizes ResNet50V2 as the feature extractor while integrating CAM to refine feature representation by emphasizing important channels. The final classification layer outputs three categories using a SoftMax activation function. The model was trained using categorical cross-entropy loss and optimized with the Adam optimizer. Experimental results demonstrated an overall accuracy of 99%, with class- wise F1-scores exceeding 97%, indicating robust performance in stroke classification. Additionally, Grad-CAM visualization was employed to enhance interpretability by highlighting critical regions in the input images influencing model decisions. The proposed approach provides an efficient and explainable deep learning solution for automated stroke detection, potentially aiding radiologists in early diagnosis and reducing clinical workload.
100
75
43
100
75
43
100
75
43
India
""""
Copyright © 2025, This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC-BY-NY-SA). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Corresponding Author: LALUPRASAD, laluprasadjaladi@gmail.com
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.
Conflict of interest: The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher’s note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
No. of Readers
No. of Reaction
No. of Comments
No. of Downloads
Or share your Opinion
Lung cancer is a leading cause of cancer-related problems worldwide, and its early diagnosis is helpful for improving...
Clay soil stabilization is a critical aspect of geotechnical engineering, as untreated clay often exhibits low strength, high...
Unified framework for noise reduction in both audio and image domains using advanced deep learning techniques, specifically a...
Cancer is a one type of diseases characterized by the uncontrolled growth and spread of abnormal cells, which...
Even if technology has been rapidly adopted in healthcare, a large number of small clinics still use outdated...
Impact Factor
Scopus & Web of Science under Evalution
Comments(0)