Improving Large Language Models’ Summarization Accuracy by Adding Highlights to Discharge Notes: Comparative Evaluation
JMIR Medical Informatics · 2025
DOI: 10.2196/66476 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift JMIR Medical Informatics, 2025. Verfasst von Mahshad Koohi Habibi Dehkordi und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, New Jersey Institute of Technology, Newark, NJ, United States. Hauptthemen laut MEDLINE: Large Language Models, Patient Discharge Summaries, Plain Language Summaries, Electronic Health Records. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
The American Medical Association recommends that electronic health record (EHR) notes, often dense and written in nuanced language, be made readable for patients and laypeople, a practice we refer to as the simplification of discharge notes. Our approach to achieving the simplification of discharge notes involves a process of incremental simplification steps to achieve the ideal note. In this paper, we present the first step of this process. Large language models (LLMs) have demonstrated considerable success in text summarization. Such LLM summaries represent the content of EHR notes in an easier-to-read language. However, LLM summaries can also introduce inaccuracies.
Objective
This study aims to test the hypothesis that summaries generated by LLMs from highlighted discharge notes will achieve increased accuracy compared to those generated from the original notes. For this purpose, we aim to prove a hypothesis that summaries generated by LLMs of discharge notes in which detailed information is highlighted are likely to be more accurate than summaries of the original notes.
Methods
To test our hypothesis, we randomly sampled 15 discharge notes from the MIMIC III database and highlighted their detailed information using an interface terminology we previously developed with machine learning. This interface terminology was curated to encompass detailed information from the discharge notes. The highlighted discharge notes distinguished detailed information, specifically the concepts present in the aforementioned interface terminology, by applying a blue background. To calibrate the LLMs’ summaries for our simplification goal, we chose GPT-4o and used prompt engineering to ensure high-quality prompts and address issues of output inconsistency and prompt sensitivity. We provided both highlighted and unhighlighted versions of each EHR note along with their corresponding prompts to GPT-4o. Each generated summary was manually evaluated to assess its quality using the following evaluation metrics: completeness, correctness, and structural integrity.
Results
We used the study sample of 15 discharge notes. On average, summaries from highlighted notes (H-summaries) achieved 96% completeness, 8% higher than the summaries from unhighlighted notes (U-summaries). H-summaries had higher completeness in 13 notes, and U-summaries had higher or equal completeness in 2 notes, resulting in P=.01, which implied statistical significance. Moreover, H-summaries demonstrated better correctness than U-summaries, with fewer instances of erroneous information (2 vs 3 errors, respectively). The number of improper headers was smaller for H-summaries for 11 notes and U-summaries for 4 notes (P=.03; implying statistical significance). Moreover, we identified 8 instances of misplaced information in the U-summaries and only 2 in the H-summaries. We showed that our findings supported the hypothesis that summarizing highlighted discharge notes improves the accuracy of the summaries.
Conclusions
Feeding LLMs with highlighted discharge notes, combined with prompt engineering, results in higher-quality summaries in terms of correctness, completeness, and structural integrity compared to unhighlighted discharge notes.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Computer Science, New Jersey Institute of Technology, Newark, NJ, United States.
- Department of Informatics, New Jersey Institute of Technology, Newark, NJ, United States.
- School of Information, Florida State University, Tallahassee, FL, United States.
- Department of Medical Informatics, Yale University, New Haven, CT, United States.
- Department of Computer Science, Montclair State University, Montclair, NJ, United States.
- School of Medicine, University of Nevada, Reno, NV, United States.
ORCID-Profile: Mahshad Koohi Habibi Dehkordi, Yehoshua Perl, Fadi P Deek, Zhe He, Vipina K Keloth, Hao Liu, Gai Elhanan, Andrew J Einstein
Zitieren
APA
Koohi Habibi Dehkordi, M., Perl, Y., Deek, F. P., He, Z., Keloth, V. K., Liu, H., Elhanan, G., & Einstein, A. J. (2025). Improving Large Language Models’ Summarization Accuracy by Adding Highlights to Discharge Notes: Comparative Evaluation. JMIR Medical Informatics. https://doi.org/10.2196/66476
Vancouver
Koohi Habibi Dehkordi M, Perl Y, Deek FP, He Z, Keloth VK, Liu H, et al. Improving Large Language Models’ Summarization Accuracy by Adding Highlights to Discharge Notes: Comparative Evaluation. JMIR Medical Informatics. 2025. doi:10.2196/66476
BibTeX
@article{KoohiHabibiDehkordi2025,
author = {Koohi Habibi Dehkordi, Mahshad and Perl, Yehoshua and Deek, Fadi P and He, Zhe and Keloth, Vipina K and Liu, Hao and Elhanan, Gai and Einstein, Andrew J},
title = {Improving Large Language Models’ Summarization Accuracy by Adding Highlights to Discharge Notes: Comparative Evaluation},
journal = {JMIR Medical Informatics},
year = {2025},
doi = {10.2196/66476}
}
EUVIMED