The conference, taking place this week in San Diego, has drawn healthcare leaders, researchers, and entrepreneurs to discuss how healthcare should map out an AI strategy.
AI is having its moment. And healthcare leaders are fully invested, excited about the potential for the technology but wary of the dangers.
The technology that's on everyone's lips and in everyone's pilot programs could be used to address healthcare's key pain points, be it a shrinking workforce, surging stress and burnout rates, or care coordination and management inefficiencies. Advocates point out that AI can handle burdensome and tedious tasks that take providers away from providing care, while also gathering and analyzing data far more quickly and efficiently than the human mind.
"It's what we hear all day, every day now," said Karen Seagraves, PhD, MPH, NEA-BC, a senior healthcare consultant and former vice president of Atrium Health's Neuroscience Institute.
But while some are calling it an unguided missile, capable of causing great harm, others see it as a transformative technology poised to reinvigorate healthcare, if only healthcare would listen.
"We wouldn't have used the iPhone," points out Chip Steiner, a product manager for healthcare at Kore.ai, a digital health company focused on language-based AI technology. "We didn't know we needed it until now we do."
The good and the bad are on display at the AIMed (Artificial Intelligence in Medicine) Global Summit, taking place this week in San Diego. The brainchild of Anthony Chang, MD, MPH, MS, MBA, a pediatric cardiologist at Children's Hospital of California (CHOC) and Freddy White, a UK-based events organizer and author of Intelligence Based Medicine, the five-year-old conference boasts a registered attendance of some 1,500 healthcare executives, clinicians, researchers, and vendors.
With a high-level and international speaker list and an intimate exhibit hall ringed by track-level stages similar to the HLTH and ViVE conferences, AIMed is poised to capture the conversation. That includes heeding the concerns of those who argue for tapping the brakes on the hype.
Just remember what happened with the EHR.
"There is enthusiasm about this disruptive technology," said Jesse Ehrenfeld, MD, MPH, a senior associate dean, tenured professor of anesthesiology and director of the “Advancing a Healthier Wisconsin Endowment” at the Medical College of Wisconsin, and president-elect of the American Medical Association, while also bringing up the "horror stories" of EHR adoption caused by a provider population that clearly wasn't ready or willing to embrace the new technology. "The existing regulatory framework is clearly not equipped to handle [AI governance]."
Ehrenfeld said the healthcare community needs to make sure that AI adoption doesn't follow the same path as EHR adoption, and that healthcare executives and clinicians play an active role in shepherding the technology forward.
"They've got to include clinician voices at the front end, not as an afterthought," he said.
During a panel composed primarily of healthcare executives, the general consensus was that AI—defined as augmented intelligence rather than artificial intelligence—would help healthcare make some early gains in reducing administrative tasks and improving workflows. That's an important selling point for an industry dealing with stress, burnout, and shortages up and down the roster, from clinicians and nurses down to tech support.
'We're always asked to do more with less," said Lynn Jeffers, MD, MBA, FACS, chief medical officer at Dignity Health.
"Efficiency is at the crux of how we solve this," added Stephanie Lahr, MD, CHCIO, the former CHIME board member and CIO and CMIO at Monument Health who's now president of digital health company Artisight.
The panel even featured one of the first and few healthcare executives whose role is specifically focused on AI: Ashley Beecy, MD, FACC, an assistant professor at Weill Cornell Medical College and medical director of AI operations at New York Presbyterian Hospital. Beecy noted her role was created to bring clinical leadership to the table when discussing AI strategy, so that clinicians can be part of the process in developing, testing, and scaling AI projects.
And that's where AI should start. While Chris DiRienzo, MD, MPP, senior vice president and chief physician executive for the American Hospital Association and an adjunct professor at the Duke University School of Medicine, pointed out that AI not only can help clinicians do their work better but also do work that clinicians can't do, the inclination is to reach immediately for the stars and use the technology to, say, find a cure for cancer. Instead, he and others said, start with the low-hanging fruit and build up the small successes.
"We have to cultivate the culture," said Eric Eskioglu, MD, MBA, chief medical and scientific officer at Novant Health.
That's going to take some time. When asked to predict the future for AI acceptance in healthcare, some foresaw 10 failures for every success and a gradual annoyance of the ChatGPT craze. But mixed with that was an understanding that healthcare leaders would move slowly to embrace more AI applications in healthcare, primarily because consumers and clinicians will be learning how to use the technology and will be pushing for more opportunities to use it.
Chang sees the landscape remaining unsettled for another one or two years, then a gradual understanding of what can and can't be done in three to five years.
"I do think there is more hope than ever before," he said.
“We have to cultivate the culture.”
— Eric Eskioglu, MD, chief medical and scientific officer, Novant Health
Eric Wicklund is the associate content manager and senior editor for Innovation, Technology, Telehealth, Supply Chain and Pharma for HealthLeaders.
The AIMed Global Summit, launched roughly six years ago, drew an estimated 1,500 attendees to this year's event in San Diego to talk about AI.
Attendees included a diverse mix of healthcare executives, researchers, vendors, and entrepreneurs.
Through the first two days, the consensus was that healthcare should be slow and methodical with AI, developing small projects that prove themselves before tackling bigger pain points, and that clinicians should always play a part in the process.