Artificial intelligence, or AI, has gained attention because of predictions that it could help eliminate a huge number of jobs across virtually every industry. A concern many have is whether there will be other jobs to replace ones that are lost.
But another consideration is whether AI could enable types of activities impossible to undertake by humans at large scale. Healthcare could benefit greatly from the capabilities, enabling aspects of care never before possible while freeing many practitioners to focus on patients in a way they seldom can.
One example was an application of IBM’s Watson suite of tools (not a single artificial entity as the performance in Jeopardy years ago led many to assume) in cancer diagnosis. A test at Memorial Sloan-Kettering Cancer Center found in tests that the software had a 90 percent accuracy rate in diagnosing lung cancer, compared to human rates of 50 percent, which is no better than flipping a coin.
What made the difference was the ability to absorb and apply massive amounts of information: 600,000 medical findings, 1.5 million patient records, and 2 million pages of medical journals. A doctor would need an estimated 160 hours a week to keep up with all the new medical knowledge published — an impossible task. Oncologists clearly aren’t made obsolete, but they could use such systems as decision support tools.
Deloitte has partnered with the Institute of Mental Health to develop a predictive system that could help better manage outpatient behavioral monitoring. By using data, the organizations hope to find patterns that can be applied looking forward to predict how certain patients will react and to reduce the number of missed appointments.
There are robotic systems guided by AI software that can administer anesthesia for colonoscopies. Although they require supervision, the approach allows one doctor to do more, reducing otherwise needed human headcount at a facility. So-called machine learning and pattern matching have already been applied in radiology to help professionals better detect problems in imaging. Even operations could benefit. Interactive programs called bots could help patients set up appointments, address billing, and otherwise free up administrative time from handling phone calls so personnel can perform more valuable services.
With all the positive potential, however, AI is not some sentient machine ready to take over tasks. There can be significant shortcomings, with predictions being wrong or the possibility that machine learning can amplify mistakes, making things worse than they were. Learning to implement AI requires insight and expertise on the part of healthcare managers and executives.
Are you interested in finding a rewarding and lucrative healthcare career that fits your individual strengths and interests? Find out how education can help you adapt to the changing healthcare landscape. American Sentinel University is an innovative, accredited provider of healthcare management degrees, including an MBA Healthcare and Master of Science Business Intelligence and Analytics.