A retrospective multicenter study conducted at three tertiary medical centers in China used pretreatment hematoxylin and eosin-stained whole-slide images and a weakly supervised deep learning model to stratify progression-free survival risk in patients with locally advanced nasopharyngeal carcinoma. The model included 253 patients and 265 whole-slide images and achieved an AUC of 0.721 and a concordance index of 0.710 for predicting 3-year progression-free survival in the locked external test cohort, with an external sensitivity of 0.941.
Source:BIOENGINEER.ORG · Read Original
Leave a Reply