Using AI to Generate Research Ideas for Adolescent Idiopathic Scoliosis
Author Information
Author(s): Muacevic Alexander, Adler John R, Leonardo Christian J, Melcer Kevin, Liu Steven H, Komatsu David E, Barsi James M
Primary Institution: Stony Brook University
Hypothesis
Can artificial intelligence effectively generate novel research questions in the field of adolescent idiopathic scoliosis?
Conclusion
AI models like ChatGPT, Llama, and Copilot can generate novel ideas in orthopaedics research.
Supporting Evidence
- ChatGPT generated 12 prompts, with 16.67% being completely novel.
- Gemini produced 14 prompts, with none being original.
- Copilot generated 10 prompts, with 10.00% being completely novel.
- Llama produced 12 prompts, with 8.33% being completely novel.
Takeaway
This study shows that computers can help doctors come up with new ideas for studying scoliosis, a condition that affects the spine in teenagers.
Methodology
A comparative evaluation study using AI models to generate research questions and assess their originality through PubMed searches.
Potential Biases
There is a risk of AI hallucination, where the AI may present misinformation as fact.
Limitations
The study did not involve human subjects and may not fully capture the AI's ability to generate contextually appropriate research ideas.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website