Publications by authors named "Edoardo Agosta"
Article Synopsis
- The study evaluates ChatGPT's performance in answering questions and solving clinical scenarios in interceptive orthodontics, involving 21 questions and 7 cases created by specialized orthodontists.
- The results indicate that ChatGPT achieved a median accuracy score of 4.9/6 and a completeness score of 2.4/3 for open-ended questions, with 40.5% of answers deemed entirely accurate and 50.5% entirely complete.
- While the AI demonstrated a high level of accuracy and completeness, it still falls short of 100%, emphasizing that it cannot fully replace human expertise in clinical decision-making.
