Dear Editor,
We would like to respond to a comment on the published article entitled “Breaking barriers: can ChatGPT compete with a shoulder and elbow specialist in diagnosis and management?1” This study aimed to assess the accuracy of an artificial intelligence model, ChatGPT, in diagnosing and managing shoulder and elbow symptoms in a private clinical setting. A total of 29 patients were included in the study, and their information was submitted through the ChatGPT portal for analysis. The results showed that ChatGPT correctly diagnosed the condition in 93% of patients and provided the appropriate management in 83% of patients. However, out of the 24 patients whose management was correct, ChatGPT failed to specify the appropriate management in 6 patients and only provided one management option in 5 patients where both options were applicable. This shows that 55% of ChatGPT's management recommendations were deemed poor.
One of the study's weaknesses is its small sample size—only 29 patients were included—which restricts how broadly the results can be applied. Furthermore, the use of a research fellow to complete the standard form raises the risk of data collection errors or inconsistencies. Furthermore, ChatGPT's performance cannot be compared to that of human specialists in order to determine whether it is superior or equivalent. In terms of future directions, it would be beneficial to conduct a larger-scale study with a more diverse patient population to increase the robustness of the findings. Additionally, incorporating a comparative analysis between ChatGPT and human specialists could provide valuable insights into the artificial intelligence model's strengths and limitations. Furthermore, it would be important to address the weaknesses identified in this study, such as improving the accuracy of management recommendations, to enhance the overall performance of ChatGPT.
Disclaimers:
Funding: No Funding was disclosed by the authors.
Conflicts of interest: The authors, their immediate families, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
Footnotes
Institutional review board approval was not required for this response to the letter to the editor.
Contributor Information
Hinpetch Daungsupawong, Email: hinpetchdaung@gmail.com.
Viroj Wiwanitkit, Email: viroj.t1776@cumail.in.
References
- 1.Daher M., Koa J., Boufadel P., Singh J., Fares M.Y., Abboud J.A. Breaking barriers: can ChatGPT compete with a shoulder and elbow specialist in diagnosis and management? JSES Int. 2023;7:2534–2541. doi: 10.1016/j.jseint.2023.07.018. [DOI] [PMC free article] [PubMed] [Google Scholar]