Comparative diagnostic accuracy of GPT-4o and LLaMA 3-70b: Proprietary vs. open-source large language models in radiology☆

被引：3

作者：

Li, David ^{[1
]}

Gupta, Kartik ^{[1
]}

Bhaduri, Mousumi ^{[1
]}

Sathiadoss, Paul ^{[1
]}

Bhatnagar, Sahir ^{[2
]}

Chong, Jaron ^{[1
]}

机构：

[1] London Hlth Sci Ctr, Dept Med Imaging, 800 Commissioners Rd E, London, ON, Canada

[2] McGill Univ, Dept Epidemiol Biostat & Occupat Hlth, 2001 McGill Coll, Montreal, PQ, Canada

来源：

CLINICAL IMAGING | 2025年 / 118卷

关键词：

Large language model; Generative artificial intelligence; Generative pre-trained transformer; Open source; Radiology;

D O I：

10.1016/j.clinimag.2024.110382

中图分类号：

R8 [特种医学]; R445 [影像诊断学];

学科分类号：

1002 ; 100207 ; 1009 ;

摘要：

[No abstract available]

引用

页数：4

共 10 条

[1] Can Generative AI Models Extract Deeper Sentiments as Compared to Traditional Deep Learning Algorithms? [J].

Anas, Mohammad ;

Saiyeda, Anam ;

Sohail, Shahab Saquib ;

Cambria, Erik ;

Hussain, Amir .

IEEE INTELLIGENT SYSTEMS, 2024, 39 (02) :5-10

[2] Performance of an Open-Source Large Language Model in Extracting Information from Free-Text Radiology Reports [J].

Le Guellec, Bastien ;

Lefevre, Alexandre ;

Geay, Charlotte ;

Shorten, Lucas ;

Bruge, Cyril ;

Hacein-Bey, Lotfi ;

Amouyel, Philippe ;

Pruvo, Jean-Pierre ;

Kuchcinski, Gregory ;

Hamroun, Aghiles .

RADIOLOGY-ARTIFICIAL INTELLIGENCE, 2024, 6 (04)

[3] Laterality: A Potential Pitfall in Applying Multimodal Large Language Models to Radiology [J].

Li, David ;

Chong, Jaron .

RADIOLOGY, 2024, 313 (02)

[4] Comparing GPT-3.5 and GPT-4 Accuracy and Drift in Radiology Diagnosis Please Cases [J].

Li, David ;

Gupta, Kartik ;

Bhaduri, Mousumi ;

Sathiadoss, Paul ;

Bhatnagar, Sahir ;

Chong, Jaron .

RADIOLOGY, 2024, 310 (01)

[5]

Li David, 2023, Radiology, V308, pe232082, DOI 10.1148/radiol.232082

[6] Comment on "Evaluation of responses to cardiac imaging questions by the artificial intelligence large language model ChatGPT" [J].

Lone, Mohd Rafi ;

Sohail, Shahab Saquib .

CLINICAL IMAGING, 2024, 114

[7] ChatGPT and Other Large Language Models Are Double-edged Swords [J].

Shen, Yiqiu ;

Heacock, Laura ;

Elias, Jonathan ;

Hentel, Keith D. ;

Reig, Beatriu ;

Shih, George ;

Moy, Linda .

RADIOLOGY, 2023, 307 (02)

[8] Diagnostic Performance of ChatGPT from Patient History and Imaging Findings on the Diagnosis Please Quizzes [J].

Ueda, Daiju ;

Mitsuyama, Yasuhito ;

Takita, Hirotaka ;

Horiuchi, Daisuke ;

Walston, Shannon L. ;

Tatekawa, Hiroyuki ;

Miki, Yukio .

RADIOLOGY, 2023, 308 (01)

[9] Limited generalizability of deep learning algorithm for pediatric pneumonia classification on external data [J].

Xin, Kevin Z. ;

Li, David ;

Yi, Paul H. .

EMERGENCY RADIOLOGY, 2022, 29 (01) :107-113

[10]

Zhou YL, 2024, RADIOLOGY, V311, DOI 10.1148/radiol.233270

← 1 →