Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 8, 2025JMIR Medical InformaticsOpen Access

Application of Large Language Models in Complex Clinical Cases: Cross-Sectional Evaluation Study

View Full Paper
Ask AI
Bookmark
Share

Authors

YHYuanheng HuangGYGuozhen YangYSYeyu Shen

Discussion

Loading...

Member takes

Overview

Cross-sectional evaluation highlights LLMs' efficiency and accuracy in complex medical cases, suggesting their potential in clinical decision support.

Key Points

  • GPTo1 and Deepseek-R1 demonstrated strong accuracy and efficiency, outperforming traditional experts in clinical decision-making.
  • Experts took an average of 33.60 minutes to provide recommendations, while LLMs completed the task significantly faster, often in under a minute.
  • Deepseek-R1 achieved the highest accuracy score with a mean Likert score of 4.19, indicating effective performance in complex cases.
  • All evaluated LLMs had lower decision costs compared to the Multidisciplinary Team, with Deepseek-R1 offering a zero direct cost advantage.

Cite This Study

Huang et al. (2025) studied this question.

synapsesocial.com/papers/68e6bc5f38ca8e474d549e29https://doi.org/10.2196/73941
View Full Paper
Ask AI
Bookmark
Share