This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American
Computing capacity is advancing more every hour today than it did in the first 90 years of the information age, and some have estimated that by the year 2045 computing capacity is going to exceed the cognitive ability of all human brains on Earth combined.
With all this data at our fingertips, why do vast discrepancies exist as to how and to what extent physicians treat patients? The biggest issue is that our decision-making ability is flawed. We lack the technology, or predictive tool, that can reconcile all of the data in real time during a patient visit. That is why artificial intelligence (AI) is so important: its ability in this area exceeds that of any MD.
From treatment rates to decision-making regarding surgical procedures, the wide-ranging differences in the care patients get—many of them depending on geography for no discernible reason—can be shocking to observers both inside and outside the medical community.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Perhaps no specialty shows more heterogeneity with respect to the delivery of nonsurgical and surgical care than treatment of spinal disorders. While orthopedic surgery, neurosurgery and medical spine treatments have advanced dramatically in the 21st century, the variance in decision-making for treatment methods can be quite dramatic.
For instance, studies reveal that in the United States there is a reported 69 percent and 75 percent disagreement, respectively, among orthopedic and neurological surgeons in the management of recurrent lumbar disc herniations and low back pain—two very common conditions. And it’s easy to understand why: the spine is a tricky and complicated region of the body. Several different conditions and diseases can affect the spine, resulting in a wide range of procedures used to treat them. Factor in the experience, expertise and pedigree of treating physicians; the wide variations in the utilization of pre- and post-operative resources; and a slew of patient-specific factors that may arise, and you end up with a swirling tempest of confusion—with the patient in the eye of the storm.
While the spine community has a wealth of knowledge in peer-reviewed medical literature, it remains extremely difficult, if not impossible, for the practicing physician or surgeon to reconcile in real time all of the data that will ultimately determine the most cost-effective choice for a particular treatment. This can include all aspects of a specific patient, the physician’s own expertise, and all relevant financial data pertaining to the patient and the proposed care.
AI TO THE RESCUE
Fortunately, a solution is within our grasp. At Cleveland Clinic, and elsewhere, we are taking steps to leverage AI.
By analyzing millions of discrete data points housed in electronic medical records and financial databases, AI can complete in milliseconds, and with more precision, what would take the human mind hours or days. AI has the capacity to identify with high probability the best decisions on surgery or non-operative treatments that will lead to optimal patient outcomes within the most appropriate cost and reimbursement model. Plus, AI can, if necessary, suggest an alternative provider in the same health care system who would likely perform better on a particular patient.
The implications of this in clinical practice are staggering. An AI platform would be the first legitimate clinical decision-making tool in spine medicine, delivering on the value equation while serving as a resource to improve physician performance and enabling financial viability during this era of financial uncertainty.
An AI-driven platform is currently under development at Cleveland Clinic. By collecting extensive amounts of historical data on spine patients and storing them in a database, we are identifying many important and previously unrecognized variables that are collectively contributing to optimal patient outcomes. We expect at least 50 percent better accuracy with AI than with physician-only approaches.
Importantly, AI in spine care will also provide higher quality, homogenization of care and resource utilization, and minimization of cost. It will also afford more logical and successful strategies to manage larger patient populations with predictable expenditure and results.
Computing capacity continues to grow. And when coupled with human expertise, the result is higher value for patients and all other stakeholders.
