Exploring chat generated pre-trained transformer-3 ability to interpret MRI knee images and generate reports
Journal of Arthroscopic Surgery and Sports Medicine · 2024
DOI: 10.25259/jassm_16_2024 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Arthroscopic Surgery and Sports Medicine, 2024. Verfasst von Sonal Saran und 6 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 10 Literaturangaben.
Abstract (Originalsprache)
Objectives
The study’s objective was to determine if Chat Generated Pre-Trained Transformer-3 (ChatGPT)-4V can interpret magnetic resonance imaging (MRI) knees and generate preliminary reports based on images and clinical history provided by the radiologist. Materials and Methods: This cross-sectional observational study involved selecting 10 MRI knees with representative imaging findings from the institution’s radiology reporting database. Key MRI images were then input into the ChatGPT-4V model, which was queried with four questions: (i) What does the image show?; (ii) What is the sequence?; (iii) What is the key finding?; and, (iv) Finally, the model generated a report based on the provided clinical history and key finding. Responses from ChatGPT-4 were documented and independently evaluated by two musculoskeletal radiologists through Likert scoring.
Results
The mean scores for various questions in the assessment were as follows: 2 for “What does the image show?,” 2.10 for “What is the sequence?,” 1.15 for “What is the key finding?,” and the highest mean score of 4.10 for the command “Write a report of MRI of the…” Radiologists consistently gave mean scores ranging from 2.0 to 2.5 per case, with no significant differences observed between different cases ( P > 0.05). The interclass correlation coefficient between the two raters was 0.92 (95% Confidence interval: 0.85–0.96).
Conclusion
ChatGPT-4V excelled in generating reports based on user-fed clinical information and key findings, with a mean score of 4.10 (good to excellent proficiency). However, its performance in interpreting medical images was subpar, scoring ≤2.10. ChatGPT-4V, as of now, cannot interpret medical images accurately and generate reports.
Zitieren
APA
Saran, S., Shirodkar, K., Ariyaratne, S., Iyengar, K., Jenko, N., Durgaprasad, B. K., & Botchu, R. (2024). Exploring chat generated pre-trained transformer-3 ability to interpret MRI knee images and generate reports. Journal of Arthroscopic Surgery and Sports Medicine. https://doi.org/10.25259/jassm_16_2024
Vancouver
Saran S, Shirodkar K, Ariyaratne S, Iyengar K, Jenko N, Durgaprasad BK, et al. Exploring chat generated pre-trained transformer-3 ability to interpret MRI knee images and generate reports. Journal of Arthroscopic Surgery and Sports Medicine. 2024. doi:10.25259/jassm_16_2024
BibTeX
@article{Saran2024,
author = {Saran, Sonal and Shirodkar, Kapil and Ariyaratne, Sisith and Iyengar, Karthikeyan and Jenko, Nathan and Durgaprasad, B. K. and Botchu, Rajesh},
title = {Exploring chat generated pre-trained transformer-3 ability to interpret MRI knee images and generate reports},
journal = {Journal of Arthroscopic Surgery and Sports Medicine},
year = {2024},
doi = {10.25259/jassm_16_2024}
}
EUVIMED