Sign in Subscribe
Artificial intelligence

Artificial Intelligence in Electronic Medical Records: A Scoping Review

Strategic Insights

— 1 min read

Researchers from McMaster University have published a new study on the applications of generative artificial intelligence (GenAI) in electronic medical records (EMRs). The scoping review aimed to explore the current real-world applications of GenAI within EMRs to support an understanding of AI applications in healthcare. The study included 55 studies and identified five themes: data manipulation, patient communication, clinical decision making, clinical prediction, and summarization.

Key Takeaways:

  • The study found that 35 of the 55 included studies originated from the United States, highlighting the need for more research on the applications of GenAI in EMRs globally.
  • ChatGPT was the most commonly referenced large language model (LLM) in the studies, which used proprietary and commercially available models to test their diverse use cases within EMRs.
  • The researchers identified several limitations of GenAI applications in EMRs, including data privacy concerns, trust issues, bias, model hallucinations, and the need for robust evaluation.
  • The study emphasized the potential of GenAI to improve patient outcomes, enhance access to medical data, streamline hospital workflows, and reduce physician workload, but noted that continued research is essential to address existing limitations.
  • The researchers suggested that validation of the findings and addressing the limitations through further research is crucial to support healthcare advancement.

Statistics:

  • 55 studies were included in the scoping review.
  • 35 of the included studies originated from the United States.
  • 24 studies focused on data manipulation.
  • 9 studies focused on patient communication.
  • 8 studies focused on clinical decision making.
  • 8 studies focused on clinical prediction.
  • 4 studies focused on summarization.
  • 2 studies focused on other applications.
  • ChatGPT was the most commonly referenced LLM in the studies.

Sources:

  • "Applications of Generative Artificial Intelligence in Electronic Medical Records: A Scoping Review." Information, 2025,16(4):284. (http://www.mdpi.com/journal/information/)
  • "Researchers from McMaster University Discuss Research in Artificial Intelligence (Applications of Generative Artificial Intelligence in Electronic Medical Records: A Scoping Review)." Health & Medicine Week. May 16, 2025; p 6190.

Read more