Yiftach Barash

The people who follow the emerging evidence on large language models (LLMs) in clinical medicine will notice something unusual. A systematic review of 30 studies found diagnostic accuracy ranging from 25% to 98%, a nearly fourfold spread across similar models on similar tasks1. Minor changes in prompt wording and format can shift model performance substantially, even when the underlying clinical task is unchanged2.

Open AccessArticle by Asaf Levartovsky 1,2,*, Tal Ovdat 1,3, Yiftach Barash 1,4,5, Zohar Ben-Shatach 1,2, Yael Skinezes 1,2, Stuart Jesin 1,2, Robert Klempfner 1,3, Ehud Grossman 1,6, Uri Kopylov 1,2, Shomron Ben-Horin 1,2 and Bella Ungar 1,2 1 Sackler Medical School, Tel Aviv University, Tel Aviv 69978, Israel 2 Department of Gastroenterology, Sheba Medical Center, Tel Hashomer, Ramat Gan 52621, Israel 3 Leviev Heart Center, Sheba Medical Center, Tel Hashomer, Ramat Gan 52621, Israel 4...

Alerting on mortality among patients discharged from the emergency department: a machine learning model Yiftach Barash1,2,3, Shelly Soffer2,3, Ehud Grossman2,4, Noam Tau1,2, Vera Sorin1,2, Eyal BenDavid2, Avinoah Irony2,5, Eli Konen1,2, Eyal Zimlichman2,6, http://orcid.org/0000-0002-4567-3108Eyal Klang1,2,3 1 Department of Diagnostic Imaging, The Chaim Sheba Medical Center, Ramat Gan, Israel 2 Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel 3 DeepVision lab, The Chaim Sheba...