Artificial intelligence assists health care

From development to clinical use, the rise of artificial intelligence has inspired new tools in health care. AI programs can listen and document clinician visits, recognize patterns of disease before physicians do and give people more autonomy in their health. Although there are many valid concerns about accuracy and patient privacy, those in the health care space say proper use of the technology can save time, money and energy for clinicians and patients alike. "There's just not enough people here to do the mass quantity of work that we have," said Dr. Kent Wright, a local physician who owns telemedicine-based practice Maximum Performance Medicine. "So if we can make the people that we have more efficient and make their job more enjoyable, just to be honest, we will get a higher level of care that's available to the people that are here." Local business builds AI tools Wright uses several AI tools, including a program founded and developed by Kingsport-based entrepreneur Wesley Combs. Phiniti AI can listen to clinic visits and transcribe them into notes for physicians and patients. It can combine data from those conversations with other uploaded patient information, too. The program, Combs said, can save physicians hours of time they would usually spend charting after work. They can also see more patients in a day — and have more face-to-face interactions with them instead of typing into a computer during conversations. "The last thing you want the doctor doing is even more work," he said, adding that the tool takes care of physician chores. "Those are just extra steps that really don't require their medical brain." Wright added that the tool, along with others, has allowed him to run a sizable practice solo and keep his prices down. His patients love the program and look forward to their summarized notes after visits. "It has already transformed how I work, and it has huge potential," Wright said. "This is the kind of communication with patients that keeps them in a collaborative state with their clinician." The tool encrypts all information and is HIPAA compliant. And it doesn't store audio files, Combs added, to protect patients from identity theft. While most of his clients are currently in other states, the entrepreneur encourages clinicians to consider looking local for new technology. "We're building some really great stuff right here in our own backyard," he said. Ballad Health implements AI About 250 Ballad Health physicians also use a listening tool to record patient and physician interactions, a practice that Chief Transformation Officer Tony Keck agreed saves hours of time. Currently, Keck said, Ballad is on the front end of its journey with AI. The company is discussing how to evaluate AI tools to make sure they're working as intended and continuing to produce good decisions over time. "The monitoring is just as important as the evaluating, because you can't just turn these AI tools on and then walk away," he said. "One of the things that we want to make sure that we don't get caught up in is the shiny object syndrome." The health system is piloting and evaluating a number of tools. One of those, Ballad Chief Physician Executive Dr. Clay Runnels said, is a program that looks for undiagnosed aortic disease in CAT scans ordered for other purposes. It can detect calcifications in the aorta that may lead to future disease so physicians can intervene early, Runnels said. Other tools include a "telesitter" program that monitors hospital patients at risk of falling. It recognizes movement and notifies a clinician before a fall happens. "Our hope is that in the vast majority of cases where AI is being used, the patient is the beneficiary of better outcomes and better patient experience without knowing that AI was involved at all," Runnels said. "All the family knows and all the patient knows is they didn't fall." Ballad's approach favors keeping humans in the loop, Keck said. AI does not make diagnoses, but assists and helps clinicians in their practice. "That's probably the near future for health systems," Runnels added, "is finding those use cases where AI assists the clinician in being better at detecting early disease in a variety of different disease states."