Personalized Risk and Outcome Assessment Benefits through Machine Learning
In This Video
- In this video, Joseph Schwab, MD, discusses how the Mass General Spine Center is utilizing machine learning algorithms to help patients understand their risks and outcomes for surgical procedures with a personalized approach
- The risk and outcomes assessments are tailored specifically for the patients based on all of their parameters with information derived from a biobank and large data repository backed up by machine learning
- In the future, patients may use an app to help clinicians measure their outcomes through real-time analytics once they leave the clinic
Advances in machine learning are now enabling orthopedic spine surgeons at Massachusetts General Hospital to help patients understand their risks and outcomes for surgical procedures.
Subscribe to the latest updates from Orthopaedics Advances in Motion
Joseph Schwab, MD, chief of the Orthopaedic Spine Center, and colleagues have developed a large biobank and data repository based on all the patients for which they have provided care. This information, when combined with machine learning algorithms, enables clinicians to provide their patients with risk and outcomes assessments based on their specific parameters.
In the future, machine learning advances will help provide clinicians with even more information, as patients will be able to use an app to track their activities outside the clinic. These real-time analytics will help clinicians personalize risk and outcomes assessments even further, thus equipping both patient and provider with more and better information to make health care decisions.
We're at an interesting time in medicine but also in society, where we're using big data as well as machine learning algorithms to try to solve problems in the clinic. We've developed several applications that we're using currently. So when a patient comes into our clinic we can place certain parameters that the patient has into our machine learning algorithms that are online, and they can give us real time data on risks.
As an example, if someone comes in for surgery and say they're going to have procedure X and they don't know the risks, we may talk to them about the risks, but actually we can give them very specific risk assessments. So, we might say, "You have a risk of having an infection of X percent," but that's actually tailored specifically for the patients based on all of their parameters.
So, we can give them a great personalized risk assessment and also outcomes assessment, which is very different than in the past where it was sort of a cookie-cutter approach. Mass General has utilized or has developed a biobank but also a very large data repository, which includes all the patients that we've cared for here at this institution. So, that combination of our very large data repository with computing to back it up and then clinical questions that are currently unsolved, leads us to trying to use these machine learning algorithms to solve problems.
One of the important aspects of the future will be analytics that are in real time. So, for instance, patients are not just going to have an interaction when they come into our clinic, they may use an app where we're measuring their outcomes once they leave the clinic. So, we'll be collecting data, in real time, at all times. So, we may be asking them questions at different times during their regular activities to really get an idea of how they're doing as opposed to simply when they're in our clinic and having patient-reported outcomes.
Learn more about the Orthopaedic Spine Center