AI-Assisted Endoscopic Ultrasound to Distinguish Benign from Malignant Liver Masses

A trained computer program identified and classified focal liver lesions accurately.

In patients with known or suspected hepatic malignancy, the evaluation of liver lesion is often critically important: The presence of a malignant metastatic liver lesion often renders a cancer unresectable, and some primary liver lesions are benign while others harbor cancer. However, it can sometimes be difficult to distinguish benign from malignant liver lesions. In a recent study, researchers evaluated the ability of artificial intelligence (AI) to identify and classify liver lesion seen on endoscopic ultrasound (EUS).

The authors used 210,685 EUS-derived images of focal liver lesions (FLLs) from 256 patients to train, validate, and test a convolutional neural network (CNN) model. The model was able to locate FLLs in 92% of EUS video samp…

Reviewing Author

Douglas G. Adler, MD, FACG, AGAF, FASGE

Disclosures

Consultant/Advisory BoardBSC; Merit; Olympus

Speaker’s BureauAbbVie

Editorial BoardsGastrointestinal Endoscopy

Douglas G. Adler, MD, FACG, AGAF, FASGE

Disclosures

Consultant/Advisory BoardBSC; Merit; Olympus

Speaker’s BureauAbbVie

Editorial BoardsGastrointestinal Endoscopy

AI-Assisted Endoscopic Ultrasound to Distinguish Benign from Malignant Liver Masses

Topics

Reviewing Author