Vivek Subbiah, Director of Early Phase Drug Development Program at NYU Langone Health, shared on LinkedIn:
“Your Outlook email may not be working and your flight may be delayed so you might have a few minutes to read our Perspective paper, just published hot off the press.
Open access, free for all to read – at least LinkedIn & internet are working․
Tissue exhaustion and slow turnaround still keep patients from the right targeted therapy.
Deep learning models can now predict actionable genomic alterations directly from routine H&E, and the prospective EAGLE (EGFR, NSCLC) and ANNAR (FGFR, urothelial) studies show this holds across diverse global cohorts.
Near-instant, non-destructive results let us triage patients for molecular testing or clinical trials at the time of diagnosis.
Every lab in the world makes an H&E slide, and this puts molecular information within reach of all of them.”
Title: AI-based prediction of actionable genomic alterations from H&E-stained slides
Authors: Woochan Hwang, Fernando Lopez-Rios, Kashif Firozvi, Vivek Subbiah, Siraj M. Ali
Other articles about artifical intelligence on OncoDaily.