Value of Endoscopic Ultrasonography for Distinguishing Malignant from Benign Non-pancreatic Periampullary Lesions: An Explainable Machine Learning Study.

BACKGROUND: Early discrimination of non-pancreatic periampullary lesions (NPLs) is challenging owing to their complex anatomy and the absence of representative clinical symptoms. PURPOSE: To establish an interpretable machine learning (ML) model that…

Preview only. Read the full abstract at the source