By Edmondo J. Robinson, M.D., MBA, FACP, Chief Transformation Officer; Senior Vice President, Consumerism; Senior Vice President and Executive Director, Christiana Care Wilmington; at Christiana Care Health System
Artificial intelligence (AI) is sometimes used to describe an ultimate goal of teaching computers how to think for themselves. This is known as “artificial general intelligence” and is still quite a ways away. But one form of AI, known as “machine learning,” is beginning to come into its own in medical fields, and it may someday transform everything from x-rays to hospital billing.
As researchers explained in the New England Journal of Medicine, machine learning (ML) “algorithms sift through vast numbers of variables, looking for combinations that reliably predict outcomes.” The ability of ML to leverage large volumes or incredibly complex data to develop predictions is a significant advance.
For example, ML may improve accuracy of medical imaging and pathology. It is well known that having a study evaluated by two reviewers is more accurate than having it evaluated by one alone. With ML the computer can serve as “quality control” and be trained to focus on commonly missed features to improve quality and outcomes. Several early stage companies have developed focused on this potential use case.
ML may also be used to identify hidden predictors of disease and appropriate therapeutic interventions. This could be especially important for addressing illnesses that are important to diagnose early, such as in newborn screening, as well as optimizing medication administration.
Rady Children’s Institute for Genomic Medicine in San Diego recently partnered with several cutting edge genomics and ML companies to break a GUINNESS WORLD RECORDS title for fastest genetic diagnosis.
However, as exciting as ML may seem, ML won’t be replacing physicians and other care providers any time soon. Machine learning may confirm a diagnosis and even predict outcomes, but it cannot explain why something happened. For that, we still need the insight and wisdom of an experienced clinician.
Also, experts rightly warn that pushes to adopt ML should not outpace studies of its efficacy. Machines require an astronomical amount of data before they can learn. For example, in a 2018 study, investigators looked to predict 30-day readmissions and inpatient mortality. To do so, they analyzed records of 216,000 patients—resulting in 47 billion data points. The ability of ML to manage such a large data set is commendable however; the end result is a slightly improved predictive model that still fails to determine the necessary intervention needed to avoid undesirable outcomes.
Still, artificial intelligence is no longer the stuff of science-fiction. The science-reality is that ML may be soon be a dream-come-true for patients and health care providers alike.
ABOUT THE AUTHOR
Edmondo J. Robinson, M.D., MBA, FACP, serves as the Chief Transformation Officer and Senior Vice President, Consumerism for Christiana Care Health System.
In this role, Dr. Robinson has responsibility for transformation of health care delivery to advance population health initiatives and move from volume-based to value-based care with a special focus on developing and managing Christiana Care’s consumerism strategy. Using an entrepreneurial spirit he works closely with Service Line and Essential Services leadership as well as other internal and external partners to meet strategic goals of optimal health, exceptional experience and organizational vitality with an emphasis on using innovative tools.
As a Christiana Care Value Institute Scholar, he conducts research on analytic methodology to inform management decisions, with particular interest in transitions of care, interdisciplinary collaboration and the intersection of technology and health care.
Dr. Robinson joined Christiana Care in 2008 as the founding medical director of Christiana Care Hospitalist Partners. In that position he recruited, mentored and managed numerous hospital medicine faculty physicians. In 2011 he transitioned to Physician-in-Chief of Wilmington Hospital, a role that developed into Senior Vice President and Executive Director, Christiana Care-Wilmington. He continues to serve in that capacity.
Dr. Robinson is an associate professor of medicine at Thomas Jefferson University’s Sidney Kimmel Medical College and an adjunct senior fellow in the Leonard Davis Institute of Health Economics at the University of Pennsylvania. He is also a fellow of the American College of Physicians and a senior fellow of the Society of Hospital Medicine. He holds a medical degree from the David Geffen School of Medicine at the University of California, Los Angeles; an MBA with an emphasis in health care management from the Wharton School at the University of Pennsylvania; and a master’s degree in health policy research also from the University of Pennsylvania.