Magnetic resonance imaging-based radiomics for early detection of primary Sjögren's disease using ZOOMit-DWI: a non-invasive diagnostic approach

Key Points

• To explore the use of ZOOMit-DWI radiomics for non-invasive early detection of primary Sjögren's disease.
• Utilized RFEP model based on parotid ZOOMit DWI radiomics
• Assessed diagnostic accuracy of the model
• Analyzed interpretability for clinical use
• Demonstrated high diagnostic accuracy for early parotid gland injury
• Showed potential for clinical translation in pSjD detection
• Introduced AI-driven radiomics as a novel diagnostic tool