Answers are grounded in this paper and cite their sources.
Demonstrates high brain stroke classification accuracy in medical imaging using an improved CNN and XGBoost hybrid, indicating potential for automated clinical diagnosis.
Brain stroke detection is one of the important diagnoses in healthcare domain. Accurate classification of brain stroke helps physicians to provide proper medication, and automatic classification reduces human efforts. Machine learning algorithms-based stroke classification frameworks are developed to extract and classify the brain image attributes. The recent development of deep learning techniques presented numerous architectures for different image processing applications. The functioning of the deep learning frameworks depends on their deep feature extraction characteristics and classification abilities. Convolutional Neural Network (CNN) is a familiar deep learning framework utilized in the proposed work for brain stroke classification. The deep extraction characteristics of CNN are improved in the proposed CNN framework, and the extracted attributes are classified using the XGBoost classifier. The Laboratory of Processing Image, Signals, and Computer Science (LAPISCO) dataset for brain stroke is utilized for experimental inspection. Improved functioning in stroke classification is verified through experiments and compared with conventional machine learning algorithms and H-StrokeNet. Experimental outcomes show that combining the improved CNN framework with XGBoost provides 97.67% accuracy, and the H-StrokeNet gives 96.55%.
No takes yet. Share an insight, caveat, or question.
S. Subashini (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: