The performance of ChatGPT-4.0o in medical imaging evaluation: a cross-sectional study
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
This study investigated the performance of ChatGPT-4.0o in evaluating the quality of positioning in radiographic images. Thirty radiographs depicting a variety of knee, elbow, ankle, hand, pelvis, and shoulder projections were produced using anthropomorphic phantoms and uploaded to ChatGPT-4.0o. The model was prompted to provide a solution to identify any positioning errors with justification and offer improvements. A panel of radiographers assessed the solutions for radiographic quality based on established positioning criteria, with a grading scale of 1–5. In only 20% of projections, ChatGPT-4.0o correctly recognized all errors with justifications and offered correct suggestions for improvement. The most commonly occurring score was 3 (9 cases, 30%), wherein the model recognized at least 1 specific error and provided a correct improvement. The mean score was 2.9. Overall, low accuracy was demonstrated, with most projections receiving only partially correct solutions. The findings reinforce the importance of robust radiography education and clinical experience.
Abstract
This study investigated the performance of ChatGPT-4.0o in evaluating the quality of positioning in radiographic images. Thirty radiographs depicting a variety of knee, elbow, ankle, hand, pelvis, and shoulder projections were produced using anthropomorphic phantoms and uploaded to ChatGPT-4.0o. The model was prompted to provide a solution to identify any positioning errors with justification and offer improvements. A panel of radiographers assessed the solutions for radiographic quality based on established positioning criteria, with a grading scale of 1–5. In only 20% of projections, ChatGPT-4.0o correctly recognized all errors with justifications and offered correct suggestions for improvement. The most commonly occurring score was 3 (9 cases, 30%), wherein the model recognized at least 1 specific error and provided a correct improvement. The mean score was 2.9. Overall, low accuracy was demonstrated, with most projections receiving only partially correct solutions. The findings reinforce the importance of robust radiography education and clinical experience.
참고문헌 (15)
- Khalifa M, "AI in diagnostic imaging : revolutionising accuracy and efficiency", Comput Methods Programs Biomed Update, 5, pp.100146~ (2024)[DOI]
- Keshavarz P, "ChatGPT in radiology : a systematic review of performance, pitfalls, and future perspectives", Diagn Interv Imaging, 105, pp.251~265 (2024)[DOI]
- "Standards of proficiency:radiographers" (2023)[URL]
- "Professional capabilities for medical radiation practice" (2020)[URL]
- Al-Malki MA, "A study on radiographic repeat rate data of several hospitals in Jeddah", Radiat Prot Dosimetry, 103, pp.323~330 (2003)[DOI]
- Hasaneen M, "Assessment of image rejection in digital radiography", J Med Life, 16, pp.731~735 (2023)[DOI]
- Almalki AA, "Educational module intervention for radiographers to reduce repetition rate of routine digital chest radiography in Makkah region of Saudi Arabia tertiary hospitals : protocol of a quasi-experimental study", JMIR Res Protoc, 6, pp.e185~ (2017)[DOI]
- Amedu C, "Harnessing the benefits of ChatGPT for radiography education: a discussion paper", Radiography(Lond), 30, pp.209~216 (2024)[DOI]
- Currie G, "ChatGPT in medical imaging higher education", Radiography(Lond), 29, pp.792~799 (2023)[DOI]
- Lampignano JP, "Bontrager’s textbook of radiographic positioning and related anatomy" (2018), Elsevier
- Pieterse T, "A model to facilitate critical thinking of radiography students", J Med Radiat Sci, 70, pp.369~379 (2023)[DOI]
- Yigit S, "Views of health sciences undergraduates on ChatGPT, an artificial intelligence-powered language model: a qualitative study", Innov Educ Teach Int (2024)[DOI]
- Zhang JS, "Exploring the usage of ChatGPT among medical students in the United States", J Med Educ Curric Dev, 11 (2024)[DOI]
- Shieh A, "Assessing ChatGPT 4.0's test performance and clinical diagnostic accuracy on USMLE STEP 2 CK and clinical case reports", Sci Rep, 14, pp.9330~ (2024)[DOI]
- Lourenco AP, "Rise of ChatGPT : it may be time to reassess how we teach and test radiology residents", Radiology, 307, pp.e231053~ (2023)[DOI]