A Deep Learning Pathomic Signature Predicts Survival and Recurrence in Pancreatic Ductal Adenocarcinoma Independent of CA19-9 - PubMed
5 hours ago
- #Deep Learning
- #Pancreatic Cancer
- #Prognostic Biomarker
- Deep learning-based pathomic signature (PCPSDL) predicts survival and recurrence in pancreatic ductal adenocarcinoma.
- PCPSDL uses H&E-stained whole-slide images and outperforms CA19-9, especially in CA19-9-negative patients.
- PCPSDL stratifies CA19-9-negative patients into high- and low-risk groups with distinct outcomes.
- High-risk score associated with immunosuppressive microenvironment and basal-like epithelial cell state.
- PCPSDL is a biologically interpretable and clinically actionable prognostic biomarker.