Comparison of Performance of Large Language Models on Lung-RADS Related Questions

Eren Çamur; Turay Cesur; Yasin Celal Güneş

doi:10.1200/GO.24.00200

JCO Global Oncology (Aug 2024)

Comparison of Performance of Large Language Models on Lung-RADS Related Questions

Eren Çamur,
Turay Cesur,
Yasin Celal Güneş

Affiliations

Eren Çamur: Eren Çamur, MD, Department of Radiology, Ministry of Health Ankara 29 Mayis State Hospital, Ankara, Turkey; Turay Cesur, MD, Department of Radiology, Ankara Mamak State Hospital, Ankara, Turkey; Yasin Celal Güneş, MD, Department of Radiology, TC Saglik Bakanligi Kirikkale Yuksek Ihtisas Hastanesi, Kırıkkale, Turkey
Turay Cesur: Eren Çamur, MD, Department of Radiology, Ministry of Health Ankara 29 Mayis State Hospital, Ankara, Turkey; Turay Cesur, MD, Department of Radiology, Ankara Mamak State Hospital, Ankara, Turkey; Yasin Celal Güneş, MD, Department of Radiology, TC Saglik Bakanligi Kirikkale Yuksek Ihtisas Hastanesi, Kırıkkale, Turkey
Yasin Celal Güneş: Eren Çamur, MD, Department of Radiology, Ministry of Health Ankara 29 Mayis State Hospital, Ankara, Turkey; Turay Cesur, MD, Department of Radiology, Ankara Mamak State Hospital, Ankara, Turkey; Yasin Celal Güneş, MD, Department of Radiology, TC Saglik Bakanligi Kirikkale Yuksek Ihtisas Hastanesi, Kırıkkale, Turkey

DOI: https://doi.org/10.1200/GO.24.00200
Journal volume & issue: no. 10

Abstract

Read online

This study evaluates LLM integration in interpreting Lung-RADS for lung cancer screening, highlighting their innovative role in enhancing radiological practice. Our findings reveal that Claude 3 Opus and Perplexity achieved a 96% accuracy rate, outperforming other models.

Published in JCO Global Oncology

ISSN: 2687-8941 (Online)
Publisher: American Society of Clinical Oncology
Country of publisher: United States
LCC subjects: Medicine: Internal medicine: Neoplasms. Tumors. Oncology. Including cancer and carcinogens
Website: https://ascopubs.org/journal/go

About the journal