The implementation of sophisticated artificial intelligence and machine learning algorithms has the potential to transform how spine surgeons approach preoperative clinic visits and guide decisions based on predictive models.
Three surgeons discuss the evolution of spine care through AI and machine learning.
Note: Responses are lightly edited for style and clarity.
David Kaye, MD. Rothman Orthopaedic Institute (Philadelphia): Today, I believe the most exciting technology in spine surgery is in artificial intelligence and machine learning and their applications in predictive analytics. For example, in spinal deformity, algorithms have been developed, capturing over 100 variables, which can quickly and accurately — in real time — inform the surgeon of the risk and benefit of a particular operation for a specific patient. Similar algorithms have been created to suggest 'ideal' alignment parameters for a specific patient based on their unique profile, and tools such as patient-specific rods have been developed to help the surgeon achieve these end goals.
As outcomes from these surgeries are collected and added to the datasets, machine learning allows the algorithm to become even more accurate. In a drive to improve patient outcomes, AI allows for assimilation of big data and interpretation in a meaningful and trainable way. These tools will become an increasingly important part of the preoperative, intraoperative and postoperative management of our patients, and may allow safer, more effective and more cost-efficient surgery moving forward.
William Taylor, MD. University of California San Diego: Improved outcomes will be based on our use of artificial intelligence to help in patient selection. This will not only be based on entering appropriate patients in preoperative education and rehab, but lead to a decrease in the use of narcotics. That will allow us to select and individualize operative procedures that are appropriate for specific patients. Intraoperatively, the continued use of robotics, navigation and augmented reality will allow us to avoid some routine issues that have caused reoperations, infections or neurologic injury.
Ronald Lehman Jr., MD. Columbia Orthopedic Surgery (New York City): One of the fascinating things about the current advancements in spine surgery is that we are moving toward spine 'solutions' and not just using and placing 'widgets.' The advancement in technology is rapidly changing the landscape for us and we are truly embarking on surgical synergy. We now have very advanced ability to:
1. Plan surgical strategy with advanced planning platforms.
2. Safely, reliably and expeditiously place implants.
3. Use machine learning and artificial intelligence to 'predict' the correction and define the goals for our surgery.
4. Appreciate if we have achieved the desired goals of the surgery intraoperatively.
5. We will advance incorporating EMR and long-term follow up with patient-specific monitoring after they leave the hospital.
It is an exciting time for spine surgery, and we will truly have a more comprehensive approach to the patients in terms of planning, execution and patient-reported outcomes. Using data and predictive analytics will allow us to tell each patient what their 'expected' results will be before they consider a spine surgery, and also perhaps who best to perform the surgery as all of our metrics, as surgeons, will be available as well.