Should we be using LLMs for discharge summarisation?
Should we be using LLMs for discharge summarisation?
This episode, we discuss some of the challenges in using large language models (LLMs) for the task of summarising inpatient encounters.
00:30 - Technical wrap - new models, MiT’s State of AI in Business 2025
12:40 - Medical summarisation
19:30 - Context Rot: How Increasing Input Tokens Impacts LLM Performance
30:50 - Verifiable Summarization of Electronic Health Records Using Large Language Models to Support Chart Review
40:00 - Evaluating large language models for drafting emergency department encounter summaries
42:25 - Evaluating Hospital Course Summarization by an Electronic Health Record-Based Large Language Model
Some resources and papers we discuss:
State of AI in 2025: https://www.artificialintelligence-news.com/wp-content/uploads/2025/08/ai_report_2025.pdf
Context Rot: How Increasing Input Tokens Impacts LLM Performance: https://research.trychroma.com/context-rot
Verma R et al, Verifiable Summarization of Electronic Health Records Using Large Language Models to Support Chart Review, medRxiv 2025.06.02.25328807; doi: https://doi.org/10.1101/2025.06.02.25328807
Williams CYK, Bains J, Tang T, Patel K, Lucas AN, Chen F, et al. 2025 Evaluating large language models for drafting emergency department encounter summaries. PLOS Digit Health 4(6): e0000899. https://doi.org/10.1371/journal.pdig.0000899
Small WR et al Evaluating Hospital Course Summarization by an Electronic Health Record-Based Large Language Model. JAMA Netw Open. 2025 Aug 1;8(8):e2526339. doi: 10.1001/jamanetworkopen.2025.26339. PMID: 40802185; PMCID: PMC12351420.