Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcoma

Summary: Clinical reliability assessment of large language models is necessary due to their increasing use in healthcare. This study assessed the performance of ChatGPT-3.5 and ChatGPT-4 in answering questions deducted from the German evidence-based S3 guideline for adult soft tissue sarcoma (STS)....

Full description

Saved in:
Bibliographic Details
Main Authors: Cheng-Peng Li, Jens Jakob, Franka Menge, Christoph Reißfelder, Peter Hohenberger, Cui Yang
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:iScience
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2589004224027202
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850121698490187776
author Cheng-Peng Li
Jens Jakob
Franka Menge
Christoph Reißfelder
Peter Hohenberger
Cui Yang
author_facet Cheng-Peng Li
Jens Jakob
Franka Menge
Christoph Reißfelder
Peter Hohenberger
Cui Yang
author_sort Cheng-Peng Li
collection DOAJ
description Summary: Clinical reliability assessment of large language models is necessary due to their increasing use in healthcare. This study assessed the performance of ChatGPT-3.5 and ChatGPT-4 in answering questions deducted from the German evidence-based S3 guideline for adult soft tissue sarcoma (STS). Reponses to 80 complex clinical questions covering diagnosis, treatment, and surveillance aspects were independently scored by two sarcoma experts for accuracy and adequacy. ChatGPT-4 outperformed ChatGPT-3.5 overall, with higher median scores in both accuracy (5.5 vs. 5.0) and adequacy (5.0 vs. 4.0). While both versions performed similarly on questions about retroperitoneal/visceral sarcoma and gastrointestinal stromal tumor (GIST)-specific treatment as well as questions about surveillance, ChatGPT-4 performed better on questions about general STS treatment and extremity/trunk sarcomas. Despite their potential as a supportive tool, both models occasionally offered misleading and potentially life-threatening information. This underscores the significance of cautious adoption and human monitoring in clinical settings.
format Article
id doaj-art-051e2cb123514c1697b9b3341b4c5fb2
institution OA Journals
issn 2589-0042
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series iScience
spelling doaj-art-051e2cb123514c1697b9b3341b4c5fb22025-08-20T02:35:01ZengElsevieriScience2589-00422024-12-01271211149310.1016/j.isci.2024.111493Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcomaCheng-Peng Li0Jens Jakob1Franka Menge2Christoph Reißfelder3Peter Hohenberger4Cui Yang5Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Sarcoma Center, Peking University Cancer Hospital & Institute, Beijing, China; Department of Surgery, University Medical Center Mannheim, Medical Faculty Mannheim, University of Heidelberg, Mannheim, GermanyDepartment of Surgery, University Medical Center Mannheim, Medical Faculty Mannheim, University of Heidelberg, Mannheim, GermanyDepartment of Surgery, University Medical Center Mannheim, Medical Faculty Mannheim, University of Heidelberg, Mannheim, GermanyDepartment of Surgery, University Medical Center Mannheim, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany; DKFZ-Hector Cancer Institute, Medical Faculty Mannheim, Heidelberg University, Mannheim, GermanyDivision of Surgical Oncology and Thoracic Surgery, Medical Faculty Mannheim, University of Heidelberg, Mannheim, GermanyDepartment of Surgery, University Medical Center Mannheim, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany; AI Health Innovation Cluster, German Cancer Research Center (DKFZ), Heidelberg, Germany; Corresponding authorSummary: Clinical reliability assessment of large language models is necessary due to their increasing use in healthcare. This study assessed the performance of ChatGPT-3.5 and ChatGPT-4 in answering questions deducted from the German evidence-based S3 guideline for adult soft tissue sarcoma (STS). Reponses to 80 complex clinical questions covering diagnosis, treatment, and surveillance aspects were independently scored by two sarcoma experts for accuracy and adequacy. ChatGPT-4 outperformed ChatGPT-3.5 overall, with higher median scores in both accuracy (5.5 vs. 5.0) and adequacy (5.0 vs. 4.0). While both versions performed similarly on questions about retroperitoneal/visceral sarcoma and gastrointestinal stromal tumor (GIST)-specific treatment as well as questions about surveillance, ChatGPT-4 performed better on questions about general STS treatment and extremity/trunk sarcomas. Despite their potential as a supportive tool, both models occasionally offered misleading and potentially life-threatening information. This underscores the significance of cautious adoption and human monitoring in clinical settings.http://www.sciencedirect.com/science/article/pii/S2589004224027202OncologyArtificial intelligence
spellingShingle Cheng-Peng Li
Jens Jakob
Franka Menge
Christoph Reißfelder
Peter Hohenberger
Cui Yang
Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcoma
iScience
Oncology
Artificial intelligence
title Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcoma
title_full Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcoma
title_fullStr Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcoma
title_full_unstemmed Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcoma
title_short Comparing ChatGPT-3.5 and ChatGPT-4’s alignments with the German evidence-based S3 guideline for adult soft tissue sarcoma
title_sort comparing chatgpt 3 5 and chatgpt 4 s alignments with the german evidence based s3 guideline for adult soft tissue sarcoma
topic Oncology
Artificial intelligence
url http://www.sciencedirect.com/science/article/pii/S2589004224027202
work_keys_str_mv AT chengpengli comparingchatgpt35andchatgpt4salignmentswiththegermanevidencebaseds3guidelineforadultsofttissuesarcoma
AT jensjakob comparingchatgpt35andchatgpt4salignmentswiththegermanevidencebaseds3guidelineforadultsofttissuesarcoma
AT frankamenge comparingchatgpt35andchatgpt4salignmentswiththegermanevidencebaseds3guidelineforadultsofttissuesarcoma
AT christophreißfelder comparingchatgpt35andchatgpt4salignmentswiththegermanevidencebaseds3guidelineforadultsofttissuesarcoma
AT peterhohenberger comparingchatgpt35andchatgpt4salignmentswiththegermanevidencebaseds3guidelineforadultsofttissuesarcoma
AT cuiyang comparingchatgpt35andchatgpt4salignmentswiththegermanevidencebaseds3guidelineforadultsofttissuesarcoma