Evaluation of GPT-4's Chest X-Ray Impression Generation: A Reader Study on Performance and Perception.

Sebastian Ziegelmayer Alexander W Marka Nicolas Lenhart Nadja Nehls Philipp-Alexander Neumann Felix N Harder Andreas Philipp Sauter Marcus R Makowski Markus Graf Joshua Gawlitza

Published in: Journal of medical Internet research (2023)

Exploring the generative capabilities of the multimodal GPT-4, our study uncovered significant differences between radiological assessments and automatic evaluation metrics for chest x-ray impression generation and revealed radiological bias.

Keyphrases

high resolution
machine learning
deep learning
single cell
pain management
mass spectrometry
dual energy