Tomas G. Neilan: Can large language models outperform ICD codes in detecting severe immune-related adverse events?

Tomas G. Neilan, Director of the Cardio-Oncology Program at Massachusetts General Hospital, shared on X:

“Very proud of amazing Ginny Sun from MGH IM Residents. Co-led with Kerry Reynolds from Mass General Cancer Center. Large language models with an excellent sensitivity and specificity for several key ICI associated adverse events and in 9 secs. Journal of Clinical Oncology.

Enhancing Precision in Detecting Severe Immune-Related Adverse Events: Comparative Analysis of Large Language Models and International Classification of Disease Codes in Patient Records

Authors: Virginia H. Sun, Julius C. Heemelaar, Ibrahim Hadzic, Vineet K. Raghu, Chia-Yun Wu,  Leyre Zubiri, Azin Ghamari, Nicole R. LeBoeuf, Osama Abu-Shawer,  Kenneth L. Kehl, Shilpa Grover, Prabhsimranjot Singh, Giselle A. Suero-Abreu, Jessica Wu, Ayo S. Falade, Kelley Grealish,  Molly F. Thomas, Nora Hathaway,  Benjamin D. Medoff, Hannah K. Gilman, Alexandra-Chloe Villani, Jor Sam Ho, Meghan J. Mooradian, Meghan E. Sise, Daniel A. Zlotoff, Steven M. Blum, Michael Dougan, Ryan J. Sullivan, Tomas G. Neilan, and Kerry L. Reynolds. ”

Source: Tomas G. Neilan/X