A knowledge-guided dual-path framework for automated liver MRI series classification

The authors aim to develop a robust automated classification system for liver MRI series to enhance clinical workflows, addressing limitations of current methods. They introduce a knowledge-guided dual-path (KDP) framework that combines CNN-extracted imaging features with metadata from DICOM headers, achieving high classification accuracy across 18 series categories. The system demonstrated excellent performance in both internal and external validation, indicating its potential for clinical integration.