Epic is diving deep into artificial intelligence, incorporating the technology into existing platforms and new products to support health system operations and clinical outcomes.
The EHR giant has a well-documented relationship with Microsoft focused on artificial intelligence, with Epic integrating Microsoft's generative AI technology into its EHR platform last year.
The company has grown its internal team to develop products with generative AI and predictive models. Epic has more than 60 significant AI development projects underway, including:
- Using AI to draft messages and notes as well as quickly summarizing patient stories
- Predictive modeling to identify areas for early intervention for improved outcomes
- AI assistants that can use conversational language to "slice and dice data" and integrate it with evidence-based care insights
- An intelligent coding assistant that can automate the revenue cycle and streamline documentation to address prior authorization, denials and appeals
- Tools to identify clinical risk adjustment opportunities and proactively text patients
- Patient self-scheduling capabilities
- Medical and billing jargon translator to language that makes sense to patients and families
- Patient flow predictions and capacity forecasts to optimize staffing and utilization
- Insights for streamlined discharges and OR utilization
At its European Group Meeting in Bristol, U.K., Epic leaders and customers showcased new AI products and collaboration. In May, two hospitals in the Netherlands became among the first to deploy the AI-enabled Patient Summaries technology, which distills years' worth of clinical notes into a short summary for clinician review.
Manchester University NHS Foundation Trust also shared results from implementing self-scheduling capabilities. Nearly all, 98%, of patients who used the self-scheduling through Epic's MyMFT said they would recommend it to others.
Epic has also partnered with several health systems to pilot generative AI projects, including the University of California San Diego Health. CHris Longhurst, MD, chief medical officer of UC San Diego Health, said in the next three to five years, generative AI could have a huge impact on healthcare, in part due to the work they're doing.
"[Large language models] offer a lot of promise when it comes to leveraging unstructured data like notes, images and scans. We'll be able to draw value from data we haven't historically been using for decision support purposes," Dr. Longhurst said in an interview last year. "If we use an LLM to do the first draft of coding, we could realize tremendous efficiency."