As the capabilities of healthcare AI tools grow, vendors continue to insist that as with previous generations of technology, AI will simply do the grunt work and free doctors’ minds up for higher uses.
For what it’s worth, the research strongly suggests that this is true. By all accounts, we are incalculably far from creating technology that can think like a trained human or apply empathy and insight to complex problems.
In the meantime, we may see some unexpected benefits from watching AI tackle healthcare problems. According to Nick Peters, a professor of cardiology and head of cardiac electrophysiology at Imperial College London, observing AI at work may jog physicians’ creativity and push them in directions they never would’ve gone otherwise.
Peters, whose article appears on the World Economic Forum website, believes that because AIs, well, think different(ly), they can sometimes inspire their human partners to try new things. “Machines are beginning to challenge human imagination in a way that may not have been anticipated, and which could…unleash a revolution in creativity,” he asserts.
Among the first changes this revolution may bring in a shift in how we track health. Peters argues that while we currently assess a patient’s status by measuring phenomena like blood pressure, respiration, and pulse, AI will replace these measures with subtler approaches.
Over time, we will use machine learning to identify other signals derived from the use of consumer devices which serve the care process better, Peters argues. “It will enable entirely new fields of cheaper, better and more cost-effective clinical science to emerge that may supersede blunt measurements such as the likes of blood pressure,” he writes.
He predicts that the data which will identify these pathways will spring in part from devices like the Apple Watch 4, which incorporates an ECG. These smart consumer devices, in turn, will eventually be able to alert and recruit a nearby citizen who has registered their competence to deliver CPR, he notes. This could have a major impact on survival rates for time-sensitive problems like cardiac arrest, Peters writes.
As interesting as his observations are, the article is too short. I do wish Peters had extended his argument further and attempted to answer more questions about the impact of AI and analytics on medical practice.
For example, if we are poised to discover health measures which take the place of basic metrics like blood pressure checks, how will we determine whether these new measures deliver the kind of results the old-fashioned ones do? What other medical processes will be transformed, and how? Also, should we focus AI development on finding alternative approaches to traditional care processes or are they just side benefits that might evolve out of other useful analysis?
Still, merely by envisioning AI as a spur to healthcare creativity, Peters has done us a service. Perhaps physicians will benefit from inevitable differences in which humans and AI software process information rather than working at cross-purposes.