OpenAlex · Aktualisierung stündlich · Letzte Aktualisierung: 15.03.2026, 21:44

Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.

Gastroenterology-Specific AI Model GastroGPT Outperforms Attending Physicians’ and ChatGPT in Analyses of Clinical Notes of Real-World Endoscopy Cases

2024·0 Zitationen·Endoscopy
Volltext beim Verlag öffnen

0

Zitationen

7

Autoren

2024

Jahr

Abstract

Aims Endoscopic procedures involve non-endoscopist collaboration in healthcare presentations. Despite AI potential, like GastroGPT, its use is limited, emphasizing the need for greater clinical integration. Our team developed GastroGPT, a Gastroenterology specific LLM for clinical based text generation. Previous studies showed this approach outperforms general LLMs and attending physicians in general gastroenterology tasks and simulated cases. However, its success in an endoscopy field and comparison to physician note texts is not known. As such, the study aims to assess GastroGPT's abilities in comparison with attending physicians and ChatGPT-4 serving as a reference model, across seven different components of patient care sequence using authentic patient data from real-life endoscopy cases.

Ähnliche Arbeiten

Autoren

Institutionen

Themen

Artificial Intelligence in Healthcare and EducationRadiomics and Machine Learning in Medical ImagingMachine Learning in Healthcare
Volltext beim Verlag öffnen