ChatGPT falls short in assisting doctors with medical diagnoses

Science & Medicine
Written By Fatima Soltan

Image credit: nci via Unsplash

An RCT study published in the journal JAMA Network Open, revealed that using ChatGPT, a large language model (LLM), did not significantly enhance diagnostic reasoning performance compared to using only conventional resources. This study underscores the limitations of current AI tools in effectively supporting medical diagnostics. Despite the potential of AI to transform healthcare, this research suggests that AI technologies, including advanced models like ChatGPT, still require substantial development to be reliably integrated into clinical decision-making processes.

Source

Goh E, Gallo R, Hom J, et al. Large Language Model Influence on Diagnostic Reasoning: A Randomized Clinical Trial. JAMA Netw Open. 2024;7(10):e2440969. doi:10.1001/jamanetworkopen.2024.40969

Additional Reading

https://www.washingtonpost.com/health/2024/11/19/chatgpt-health-doctor-diagnosis/

https://www.nytimes.com/2024/11/17/health/chatgpt-ai-doctors-diagnosis.html

https://www.sciencefocus.com/news/gps-use-chatgpt-study-finds

Fatima Soltan
Previous

The one beverage to avoid on planes
Next

Lunchables removed from U.S. school lunch programs