Abstract
The growing importance of efficient and accurate medical image classification has led to increased research interest in the application of deep learning techniques. This research paper investigates the utilization of Convolutional Neural Networks (CNNs) for the classification of brain strokes from computed tomography (CT) images. Leveraging a diverse dataset, we systematically examine the performance of the CNN model in distinguishing between 'Normal' and 'Stroke' image classes. The study employs a comprehensive set of evaluation metrics, including accuracy, precision, recall, and F-score, to assess the model's diagnostic efficacy. The training process is rigorously monitored, ensuring convergence of learning rates and model stability. Our results demonstrate promising levels of accuracy and an increasing trend in precision, recall, and F-score, underscoring the potential of deep learning in enhancing stroke diagnosis. Furthermore, the study addresses data scarcity challenges through data augmentation techniques, emphasizing their pivotal role in model training. While acknowledging the need for larger and more diverse datasets and potential model enhancements, this research contributes to the broader goal of harnessing artificial intelligence for improved healthcare, particularly in the timely and accurate detection of critical medical conditions like brain strokes through non-invasive imaging modalities.