Human-AI collaboration in large language model-assisted brain MRI differential diagnosis: a usability study
Question While large language models (LLM) have the potential to support radiological differential diagnosis, the role of human-LLM collaboration in this context remains underexplored. Findings LLM-assisted brain MRI differential diagnosis yielded superior accuracy over conventional internet search. Inaccurate case descriptions, LLM hallucinations, and insufficient contextualization were identified as potential challenges. Clinical relevance Our results highlight the potential of an LLM-assisted workflow to increase diagnostic accuracy but underline the necessity to study collaborative efforts between humans and LLMs over LLMs in isolation.
Authors
- Christopher Klenk (ORCID: https://orcid.org/0000-0002-0180-8437)
- Benedikt Wiestler (ORCID: https://orcid.org/0000-0002-2963-7772)
- Severin Schramm (ORCID: https://orcid.org/0000-0002-3299-5861)
- Lena Schmitzer (ORCID: https://orcid.org/0000-0003-1644-3335)
- Enrike Rosenkranz
- Dennis M. Hedderich (ORCID: https://orcid.org/0000-0001-8994-5593)
- Su Hwan Kim (ORCID: https://orcid.org/0000-0002-5383-2041)
- Cornelius Berberich
- Nicolas Lenhart
- Claus Zimmer (ORCID: https://orcid.org/0000-0002-4805-9103)
- Kerem Serguen
- Jonas Wihl
Institutions
- TUM Klinikum (DE)
- Technical University of Munich (DE)
Publication Details
- Journal
- European Radiology
- Published
- 2025-03-07
- DOI
- https://doi.org/10.1007/s00330-025-11484-6
- Citations
- 21
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 13.09