Welcome to Hipócrates 3.0! In this episode, we explore an innovative study published this year in the Journal of Medical Internet Research (IF 7.4) that puts artificial intelligence, specifically ChatGPT’s GPT-4, head-to-head with resident physicians in diagnosing medical emergencies. Developed by OpenAI, GPT-4 has proven to be a promising tool, surpassing both its predecessor, GPT-3.5, and resident physicians in diagnostic accuracy for internal medicine cases.

We discuss the study’s surprising results, its implications for the future of medicine, and how AI can be an essential ally in clinical practice. And for those of you striving to pass Step 1, we’ve included a message of encouragement and a shout-out to the amazing USMLE Privateers group for their unwavering support. Press play and join us as we dive into the impact of AI in healthcare!

John Michael Hoppe, et al. ChatGPT with GPT-4 Outperforms Emergency Department Physicians in Diagnostic Accuracy: Retrospective Analysis. Journal of Medical Internet Research, vol. 26, 8 July 2024, pp. e56110–e56110, https://doi.org/10.2196/56110.

1) A systematic literature review of machine learning based risk prediction models for diabetic retinopathy progression:

https://pubmed.ncbi.nlm.nih.gov/37673580/

2) Differential diagnosis of secondary hypertension based on deep learning

https://pubmed.ncbi.nlm.nih.gov/37295898/

3) Gynecological cancer prognosis using machine learning techniques: A systematic review of the last three decades (1990-2022)

Journal: Artificial Intelligence in Medicine

Follow me on instagram: @drthiagoflorencio.ia