How machine learning will fundamentally change the lives of healthcare providers

Artificial intelligence and machine learning technologies could fundamentally change healthcare forever, both for providers and their patients. A new analysis published in the New England Journal of Medicine examined that potential shift in great detail.

“What if every medical decision, whether made by an intensivist or a community health worker, was instantly reviewed by a team of relevant experts who provided guidance if the decision seemed amiss?” asked Google’s Alvin Rajkomar, MD, and colleagues. “Patients with newly diagnosed, uncomplicated hypertension would receive the medications that are known to be most effective rather than the one that is most familiar to the prescriber. Inadvertent overdoses and errors in prescribing would be largely eliminated. Patients with mysterious and rare ailments could be directed to renowned experts in fields related to the suspected diagnosis.”

This, according to the authors, describes the true potential of machine learning in medicine. Rajkomar et al. explored this topic in great detail, and these are five areas where they believe machine learning can augment the work of healthcare providers on a day-to-day basis:

Prognosis

By “learning” patterns, the authors noted, machine learning algorithms will be able to anticipate how patients may recover from illnesses or respond to treatments. They key to such advances is data, which presently lives in a wide variety of places, including electronic health records, PACS and more.

“A natural solution would be to systematically place data in the hands of patients themselves,” the authors wrote. “We have long advocated for this solution, which is now enabled by the rapid adoption of patient-controlled application programming interfaces.”

Diagnosis

Machine learning will be able to help physicians identify conditions quicker, and with more accuracy, than ever before.

“With data collected during routine care, machine learning could be used to identify likely diagnoses during a clinical visit and raise awareness of conditions that are likely to manifest later,” the authors wrote.

Rajkomar and colleagues noted that this approach would still be limited by the quality of the prediction models and the clinicians themselves. However, they added, models could also be built to “suggest questions or tests” to physicians based on certain data being collected in real time.

Treatment

With so many physicians treating so many patients at any given time, machine learning models can help “sort through these natural variations” in care to identify which techniques work better than others. And that’s not the only example of how treatment can be potentially augmented by these growing technologies.

“Machine learning can also be used to automatically select patients who might be eligible for randomized, controlled trials on the basis of clinical documentation or to identify high-risk patients or subpopulations who are likely to benefit from early or new therapies under study,” the authors wrote. “Such efforts can empower health systems to subject every clinical scenario for which there is equipoise to more rigorous study with decreased cost and administrative overhead.”

Clinician Workflow

Machine learning will be able to help clinicians use electronic health records faster and more efficiently, the authors explained. Models can also be implemented to improve workflow in other ways, including analyzing video in real time and “more mundane tasks” such as keeping track of key supplies.

Expanding the Availability of Clinical Expertise

One of the most important ways machine learning could change healthcare is its ability to improve patient access to clinicians.

“For example, patients with new rashes may be able to obtain a diagnosis by sending a picture that they take on their smartphones, thereby averting unnecessary urgent-care visits,” the authors wrote. “A patient considering a visit to the emergency department might be able to converse with an automated triage system and, when appropriate, be directed to another form of care.”

Michael Walter
Michael Walter, Managing Editor

Michael has more than 16 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.

Around the web

After reviewing years of data from its clinic, one institution discovered that issues with implant data integrity frequently put patients at risk. 

Prior to the final proposal’s release, the American College of Radiology reached out to CMS to offer its recommendations on payment rates for five out of the six the new codes.

“Before these CPT codes there was no real acknowledgment of the additional burden borne by the providers who accepted these patients."

Trimed Popup
Trimed Popup