Artificial intelligence (AI) driven solutions have the potential to transform how physicians navigate and interpret patient charts. Reviewing prior notes within the electronic health record (EHR) is a fundamental aspect of clinical practice. This has become more difficult due to “note bloat,” with recent evidence showing that note length has increased by 60% over the past decade. This phenomenon results in clinicians spending over 5 min to review a single patient chart based on a recent analysis of 100 million patient encounters with 155,000 physicians across 400 health systems.
Recent advancements in natural language processing (NLP) and deep learning have resulted in large language models (LLMs) capable of processing and summarizing the heterogenous, unstructured, narrative free text that comprises medical notes. Synthesizing patient clinical information from medical notes using LLMs has the potential to improve efficiency, decrease cognitive burden, lower the likelihood of missing key information, and reduce physician burnout.