SurgeryLLM: a retrieval-augmented generation large language model framework for surgical decision support and workflow enhancement
2024

SurgeryLLM: A New AI Tool for Surgical Decision Support

Sample size: 12 publication Evidence: moderate

Author Information

Author(s): Ong Chin Siang, Obey Nicholas T., Zheng Yanan, Cohan Arman, Schneider Eric B.

Primary Institution: Yale School of Medicine

Hypothesis

Can a retrieval-augmented generation large language model improve surgical decision-making and workflow?

Conclusion

SurgeryLLM successfully demonstrated the feasibility of incorporating external domain-specific information to support surgical decision-making.

Supporting Evidence

  • SurgeryLLM incorporated domain-specific knowledge from current evidence-based surgical guidelines.
  • SurgeryLLM improved the identification of abnormal lab values compared to VanillaLLM.
  • SurgeryLLM provided clinically accurate recommendations based on surgical guidelines.
  • SurgeryLLM was able to draft preliminary operative notes, while VanillaLLM could not.

Takeaway

SurgeryLLM is a smart computer program that helps doctors make better decisions during surgery by using the latest medical guidelines.

Methodology

The study compared outputs from a retrieval-augmented LLM (SurgeryLLM) and a standard LLM (VanillaLLM) using simulated patient data.

Potential Biases

Future efforts will focus on bias detection and correction techniques.

Limitations

The study used limited examples and may face challenges related to patient data availability and accuracy.

Digital Object Identifier (DOI)

10.1038/s41746-024-01391-3

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication