Neuroimage-Based Stroke Identification: A Machine Learning Approach

Abstract

This study presents a novel machine learning-based diagnostic model for the timely identification of strokes using neuro imaging data. Stroke diagnosis is critical, as rapid and accurate detection significantly influences patient outcomes. The proposed model employs a comprehensive methodology utilizing various machine learning algorithms, including Logistic Regression, Sup port Vector Machine (SVM), Random Forest, Decision Tree, and Convolutional Neural Network (CNN), to analyze neuroimages and predict stroke  occurrences. The model was trained and validated on an extensive dataset of brain images, achieving remarkable per formance in distinguishing between stroke and non-stroke cases. Notably, the CNNalgorithm demonstrated superior accuracy, achiev ing an accuracy of 95.6%, sensitivity of 94.2%, and specificity of 96.5%. The Random Forest and SVM models also performed well, with accuracies of 93.1% and 92.5%, respectively. This research highlights the potential of machine learning techniques to enhance stroke diagnosis, providing healthcare professionals with valuable tools for informed decision-making and ultimately improving pa tient outcomes while reducing the economic burden associated with strokes.