Summary of Enhancing Large Language Models For Clinical Decision Support by Incorporating Clinical Practice Guidelines, By David Oniani et al.
Enhancing Large Language Models for Clinical Decision Support by Incorporating Clinical Practice Guidelines
by David Oniani, Xizhi Wu, Shyam Visweswaran, Sumit Kapoor, Shravan Kooragayalu, Katelyn Polanska, Yanshan Wang
First submitted to arxiv on: 20 Jan 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed research develops and evaluates three methods for incorporating Clinical Practice Guidelines (CPGs) into Large Language Models (LLMs), enhancing their ability to provide accurate recommendations for Clinical Decision Support (CDS). The methods, Binary Decision Tree (BDT), Program-Aided Graph Construction (PAGC), and Chain-of-Thought-Few-Shot Prompting (CoT-FSP), are tested on four LLMs (GPT-4, GPT-3.5 Turbo, LLaMA, and PaLM 2) using synthetic patient descriptions and both automatic and human evaluation. The results show that all four LLMs perform better when enhanced with CPGs compared to the baseline Zero-Shot Prompting (ZSP), with BDT outperforming CoT-FSP and PAGC in automatic evaluation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research takes large language models and makes them smarter by adding clinical practice guidelines. The goal is to help doctors make better decisions when treating patients. Four different ways of combining the two are tested, and all of them do a better job than just using the guidelines alone. This could be useful for many medical situations, not just COVID-19. |
Keywords
* Artificial intelligence * Decision tree * Few shot * Gpt * Llama * Palm * Prompting * Zero shot
