Study shows AI may improve patient discharge summaries
The use of artificial intelligence (AI) could help physicians prepare hospital patient discharge summaries more efficiently and with greater accuracy and completeness. This is the conclusion of a Stony Brook University study published online in npj Digital Medicine.
A recent study published in npj Digital Medicine suggests that artificial intelligence (AI) could potentially streamline and enhance the creation of patient discharge summaries in hospitals. The research, conducted by Stony Brook University, found that AI-generated summaries, developed using a large language model (LLM), were preferred by clinicians and scored higher in quality, readability, accuracy, and completeness compared to those written by human physicians.
These AI-generated summaries, which took only a few minutes to produce, contained the essential information required for patient discharge, such as treatment details, medication needs, unresolved medical issues, and follow-up requirements. Despite both AI and human-generated summaries having errors, clinical reviewers did not find a significant difference in the potential harm caused by those errors.
The study emphasizes the importance of human review and verification before utilizing AI-generated discharge drafts for patient records. The researchers aim to further develop the LLM with a focus on accuracy, safety, time savings, and integration into clinical workflows, including a planned pilot study to test the tool in real-world hospital settings.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
