Artificial Intelligence

How healthcare organizations are implementing AI/ML tools.

Even the best algorithm could still need months to verify its performance, a healthcare systems expert says. So IT leaders should ensure they budget for the training process whenever they evaluate any artificial intelligence technology.

Previewing HIMSS24 panels discussing informatics nurses and the realities of genAI for physicians – and how to make the tech viable within healthcare – with Tom Lawry, managing director at Second Century Tech.  

Large language models can provide a big boost here for providers, says the CEO of Medicomp. But algorithms should be trained on hierarchical condition categories for risk adjustment – and challenges with data quality and bias need addressing.

There are a variety of applications of ChatGPT and LLMs in healthcare, and various uses that have the potential to help or harm patients. Dr. Srinivasan Suresh, VP, CIO and CMIO at UPMC Children's Hospital of Pittsburgh, digs into these tools.

The National Institutes of Health may seek as much as $200 million to acquire a new electronic health record that does not depend on institutional knowledge to maintain – and can handle continuous machine learning validation. 

Technology, including AI, should serve data as the foundation for the optimal patient experience, says Noyo National Practice Leader Gary Davis.

Ram Rimal and Mike Plesh from UNC Health preview their HIMSS24 panel session, where they'll review how multiple stakeholders across UNC created a framework for adopting both vendor and in-house artificial intelligence applications.

Bommae Kim, lead data scientist at Hackensack Meridian Health, will present her findings at HIMSS24 in Orlando.

And the tool deployed at Arizona-based SimonMed, which complements the work of subspecialty-trained radiologists, has a 96.9% to 100% sensitivity range per bone, research shows.