1 of 1
Story summary
- A multicenter study across seven cancer centers enrolled 2,463 patients to develop a deep learning pathomics signature (DLPS) for gastrointestinal cancers, including pancreatic cancer.
- The study followed STROBE guidelines and validated DLPS features and models across external cohorts, reinforcing prognostic value.
- It also analyzed multi-omics data from The Cancer Genome Atlas (TCGA) to improve prognostic stratification using advanced statistical methods.
