Dana-Farber’s Breast Oncology Center – Accuracy and Reproducibility of ChatGPT Responses to Breast Cancer Tumor Board Patients
Dana-Farber’s Breast Oncology Center posted on X:
“New study evaluates the accuracy and reproducibility of ChatGPT responses to Breast Cancer Tumor Board Patients.”
Nancy Lin, Breast cancer specialist at the Dana-Farber Cancer Institute and Associate Professor of Medicine at Harvard Medical School, shared this post on X, adding:
“What happens when you query chatGPT with 362 breast cancer cases?
- 46% responses differed from breast cancer expert opinion
- when the same case was entered 3 times, the responses differed ~2/3 of the time!
59 F T1a (2 mm) N0 TNBC s/p mastect/SLNB
- chat GPT recommended/consider chemo and in 2 of 3 duplicate cases consider PMRT (!)
- none of breast ca experts rec chemo or RT
43 F ER+HER2- MBC with bone mets
- cdk4/6i recommended only 1/3 entries
- bone agents rec only 1/3 entries.”
Title: Accuracy and Reproducibility of ChatGPT Responses to Breast Cancer Tumor Board Patients
Authors: Ning Liao, Cheukfai Li, William J. Gradishar, V. Suzanne Klimberg, Joshua A. Roshal, Taize Yuan, Sanjiv S. Agarwala, Vincente K. Valero, Sandra M. Swain, Julie A. Margenthaler, Isabel T. Rubio, Sara A. Hurvitz, Charles E. Geyer Jr, Nancy U. Lin, Hope S. Rugo, Guochun Zhang, Nanqiu Liu, Charles M. Balch
Read The Full Article at JCO Clinical Cancer Informatics.
More posts featuring Dana-Farber’s Breast Oncology Center on OncoDaily.
-
Challenging the Status Quo in Colorectal Cancer 2024
December 6-8, 2024
-
ESMO 2024 Congress
September 13-17, 2024
-
ASCO Annual Meeting
May 30 - June 4, 2024
-
Yvonne Award 2024
May 31, 2024
-
OncoThon 2024, Online
Feb. 15, 2024
-
Global Summit on War & Cancer 2023, Online
Dec. 14-16, 2023