@inproceedings{liu2023exploring,
author = {Liu, Qianchu and Hyland, Stephanie and Bannur, Shruthi and Bouzid, Kenza and Coelho de Castro, Daniel and Wetscherek, Maria Teodora and Tinn, Robert and Sharma, Harshita and Pérez-García, Fernando and Schwaighofer, Anton and Rajpurkar, Pranav and Khanna, Sameer Tajdin and Poon, Hoifung and Usuyama, Naoto and Thieme, Anja and Nori, Aditya and Lungren, Matthew P and Oktay, Ozan and Alvarez-Valle, Javier},
title = {Exploring the Boundaries of GPT-4 in Radiology},
booktitle = {EMNLP 2023},
year = {2023},
month = {October},
abstract = {The recent success of general-domain large language models (LLMs) has significantly changed the natural language processing paradigm towards a unified foundation model across domains and applications. In this paper, we focus on assessing the performance of GPT-4, the most capable LLM so far, on the text-based applications for radiology reports, comparing against state-of-the-art (SOTA) radiology-specific models. Exploring various prompting strategies, we evaluated GPT-4 on a diverse range of common radiology tasks and we found GPT-4 either outperforms or is on par with current SOTA radiology models. With zero-shot prompting, GPT-4 already obtains substantial gains (≈ 10% absolute improvement) over radiology models in temporal sentence similarity classification (accuracy) and natural language inference (F1). For tasks that require learning dataset-specific style or schema (e.g. findings summarisation), GPT-4 improves with example-based prompting and matches supervised SOTA. Our extensive error analysis with a board-certified radiologist shows GPT-4 has a sufficient level of radiology knowledge with only occasional errors in complex context that require nuanced domain knowledge. For findings summarisation, GPT-4 outputs are found to be overall comparable with existing manually-written impressions.},
url = {http://approjects.co.za/?big=en-us/research/publication/exploring-the-boundaries-of-gpt-4-in-radiology/},
}