In 2018, almost three-quarters of Americans believed AI would eliminate more jobs than it stood to create. Almost a quarter expected healthcare to be among the earliest and hardest hit of all employment sectors. However, in 2023, McKinsey & Co. projected overall demand for healthcare workers to grow by 30% by 2030.

That’s one of a number of points on healthcare AI that can be viewed from two conflicting perspectives: glass half-full and glass half-empty.

The points are fleshed out in commentary posted by staff analysts at Health IT Analytics. Here are five more.

1. As recently as 2021, radiologists were being told their days were numbered: AI was still coming for their jobs.

However, it’s now apparent that there aren’t enough radiologists—or pathologists, surgeons or PCPs. In fact, name the specialty and it’s probably facing a physician shortage. Meanwhile, AI can help retain clinicians as it helps “alleviate the stresses of burnout that drive healthcare workers to resign,” the Health IT Analytics authors write.

2. Concerns persist that clinicians may become de-skilled by relying on AI and related technologies for various clinical tasks.

However, this scenario is unlikely to materialize on a broad scale since automation bias isn’t new to healthcare—and since time-tested strategies exist to ward it off.

3. Healthcare consumers are increasingly comfortable with the notion of AI as a tool for improving care delivery.

However, a research letter published in JAMA Network Open in 2022 showed two-thirds of 1,000 adults surveyed consider it “very important” for providers to inform patients when AI is used in their care for any reason.

4. Recent research shows patients strongly prefer to have consequential care tasks—prescribing medications, diagnosing skin conditions, those kinds of things—performed by human experts.

However, as the Healthcare IT Analytics authors point out, “whether patients and providers are comfortable with the technology or not, AI is advancing in healthcare. Many health systems are already deploying the tools across a plethora of use cases.”

5. If health data is to be safeguarded for use in AI, privacy laws and regulations must be updated. The potential for supposedly de-identified data to be re-identified is an attention-grabbing concern, and it’s not the only thing to worry about.

However, AI “falls into a regulatory gray area, making it difficult to ensure that every user is bound to protect patient privacy and will face consequences for not doing so.”

On the latter point, the authors comment that security and privacy “will always be paramount, but this ongoing shift in perspective—as stakeholders get more familiar with the challenges and opportunities of data sharing is vital for allowing AI to flourish in a health IT ecosystem where data is siloed and access to quality information—is one of the industry’s biggest obstacles.”

The article also looks at light and shade in the areas of ethics, responsibility and oversight with regard to healthcare AI.

Read the whole thing.

Buzzworthy developments of the past few days.

• Today’s healthcare consumers often expect a little too much of their doctors. AI can help set those relationships right. That’s one surprising point of view to come out of a think-in conducted at the University of California, Davis. Personal wellbeing is primarily our own responsibility, says David Lubarsky, MD, the institution’s CEO and vice chancellor of human health sciences. “To the extent that AI—especially generative AI technologies—can help direct people to live healthier lives, they’re going to need less care,” Lubarsky says. Also weighing in is Dennis Chornenky, chief AI advisor to UC-Davis Health and a former technology advisor to the White House. The discussion runs about 40 minutes. Check out some or all of it here.
   
• Down the road a piece, the San Diego Union-Tribune picks the brain of one Karandeep Singh, MD. The newspaper is keenly interested in the physician’s mind because he’s the incoming CAIO—chief AI officer—at UC-San Diego. “I come from a place of optimism but also skepticism about any one tool,” Singh tells health reporter Paul Sisson. “I think a lot of what I’ll be doing in my role is trying to help us cut through the clutter, rigorously looking at what’s out there so we can be leaders in helping others identify what works and what doesn’t.” Don’t expect the University of Michigan to celebrate the hiring, much less the interview. It’s Ann Arbor that Singh is leaving for San Diego. Q&A here.
   
• The global business consultancy Bain & Co. expects healthcare AI of the generative kind to advance in ‘select and focused’ use cases in 2024. Looking further ahead, the technology’s effect on healthcare may be of sufficient depth and breadth to totally transform workflows, applications and “ways of working.” Bain’s advice to investors: Be thoughtful about generative AI’s impact on your present and future selections, and you should be able to “harness this technological change to generate returns and accelerate the transformation of the healthcare sector.”
   
• In the cold war between providers and payers, generative AI is a superweapon. Watch for an escalating arms race as both sides vie for GenAI superiority. Politico takes a look at the gathering war, suggesting that the $4 trillion Americans annually accrue in medical expenses may make some fingers trigger-happy with the GenAI button.
   
• St. Michael’s Hospital in Toronto is at the forefront of healthcare AI adoption in Canada. It’s been helped along by a $10 million donation from a businessman in Hong Kong on top of a $5 million annual injection from the hospital’s own fundraising operations. But the institution has done its part to spend well, “quietly” testing 50 or so AI products over the past couple of years, the Globe and Mail newspaper reports. A standout performer is an AI-powered clinical alert system called ChartWatch. On one floor, this has delivered a 26% reduction in the risk of death among patients receiving curative care. Read the article.
   
• On the other side of the world, AI’s potential to transform the healthcare landscape is ‘undeniable.’ So report two researchers in the daily newspaper Modern Ghana. AI proponents in that Western African nation face challenges familiar to every healthcare system in the world—ethical considerations, data privacy concerns, ensuring equitable access and so on, the authors note. Still, the future of AI-aided healthcare in Ghana “looks promising, with the potential to significantly improve the quality of life and health outcomes for the population.” Read the rest.
   
• Personalized treatment planning, wellness management and virtual medical assistance. That’s one correct and acceptable answer to: Name three personal healthcare trends to watch for in 2024 as predicted by self-professed data geek and AI enthusiast Pooja Choudhary. The tech reporter summarizes those items and seven more at AIThority.
   
• Meet the top 10 people in healthcare AI. And maybe gnash teeth over who didn’t make the list. These kinds of things are subjective, after all. But isn’t their inherent incompleteness what makes them fun to argue about? The present one is from Business Insider. See it here.
   
• From AIin.Healthcare’s news partners:
   
  • Cardiovascular Business: HeartFlow’s AI-powered CAD offering reaches a significant milestone
  • Health Imaging: Mayo Clinic rings in the new year with two AI-powered cancer care initiatives