Hybrid CNN–GCN framework for brain tumor MRI classification: A graph‐based approach to smart healthcare diagnostics

Key Points

• To develop a hybrid CNN-GCN framework for automated brain tumor classification using MRI scans.
• Utilized a CNN for hierarchical feature extraction from MRI images.
• Employed a GCN for relational reasoning among extracted features.
• Integrated both approaches into a cohesive diagnostic pipeline.
• Evaluated the framework's performance in a smart healthcare context.
• Achieved enhanced diagnostic precision for brain tumors.
• Significantly reduced the workload on radiologists.
• Facilitated more timely therapeutic interventions.
• Demonstrated strong potential for integration in clinical settings.

Abstract