Evaluation of GPT-4's Chest X-Ray Impression Generation: A Reader Study on Performance and Perception

被引：9

作者：

Ziegelmayer, Sebastian ^{[1
,2
]}

Marka, Alexander W. ^{[1
]}

Lenhart, Nicolas ^{[1
]}

Nehls, Nadja ^{[1
]}

Reischl, Stefan ^{[1
]}

Harder, Felix ^{[1
]}

Sauter, Andreas ^{[1
]}

Makowski, Marcus ^{[1
]}

Graf, Markus ^{[1
]}

Gawlitza, Joshua ^{[1
]}

机构：

[1] Tech Univ Munich, Sch Med, Dept Diagnost & Intervent Radiol, Ismaninger Str 22, D-81675 Munich, Germany

[2] Tech Univ Munich, Klinikum Rechts Isar, Ismaninger Str 22, D-81675 Munich, Germany

来源：

JOURNAL OF MEDICAL INTERNET RESEARCH | 2023年 / 25卷

关键词：

generative model; GPT; medical imaging; artificial intelligence; imaging; radiology; radiological; radiography; diagnostic; chest; x-ray; x-rays; generative; multimodal; impression; impressions; image; images; AI;

D O I：

10.2196/50865

中图分类号：

R19 [保健组织与事业（卫生事业管理）];

学科分类号：

摘要：

Exploring the generative capabilities of the multimodal GPT-4, our study uncovered significant differences between radiological assessments and automatic evaluation metrics for chest x-ray impression generation and revealed radiological bias.

引用

页数：5

共 17 条

[1] Leveraging GPT-4 for Post Hoc Transformation of Free-text Radiology Reports into Structured Reporting: A Multilingual Feasibility Study
Adams, Lisa C.
Truhn, Daniel
Busch, Felix
Kader, Avan
Niehues, Stefan M.
Makowski, Marcus R.
Bressem, Keno K.
[J]. RADIOLOGY, 2023, 307 (04)
[2] GPT-4 in Radiology: Improvements in Advanced Reasoning
Bhayana, Rajesh
Bleakney, Robert R.
Krishna, Satheesh
[J]. RADIOLOGY, 2023, 307 (05)
[3] Endo M, 2021, MACH LEARN HLTH
[4] How to Create a Great Radiology Report
Hartung, Michael P.
Bickle, Ian C.
Gaillard, Frank
Kanne, Jeffrey P.
[J]. RADIOGRAPHICS, 2020, 40 (06) : 1658 - 1670
[5] Jain Shantanu, PREPRINT
[6] Human heuristics for AI-generated language are flawed
Jakesch, Maurice
Hancock, Jeffrey T.
Naaman, Mor
[J]. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2023, 120 (11)
[7] Kojima Takeshi, 2023, arXiv
[8] Lewis P, arXiv
[9] BLEU: a method for automatic evaluation of machine translation
Papineni, K
Roukos, S
Ward, T
Zhu, WJ
[J]. 40TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, PROCEEDINGS OF THE CONFERENCE, 2002, : 311 - 318
[10] Radford A, 2021, PR MACH LEARN RES, V139

← 1 2 →