Spine robots are great when it comes to navigation and instrumentation, but surgeons have yet to fully tap into their potential, Katherine Wagner, MD, said at Becker’s 22nd Orthopedic and Pain Management-Driven ASC + The Future of Spine Conference.
Dr. Wagner, of Ventura (Calif.) Neurosurgery, shared her insights on where she wants to see more development from the technology during a June 19 panel.
Question: What big improvements do you want for spine robots in the future?
Dr. Katherine Wagner: There’s two things that we’re under utilizing with the spine robots. The first thing is visualization. One problem with the robot is we’re just using it for instrumentation and neuro-navigation. We have this machine that’s literally $900,000 to $1.5 million, and you’re just using it to hold an arm and a pathway to put screws in. If you’re more advanced, if you have the latest software updates, you may be using it to pre-plan your rods … You’ve optimized your instrumentation and interop navigation, but we have a tool for visualization where we could add the endoscope, but also we can add intraoperative cameras. We’re not utilizing these tools at all for visualization when we have a scope that is halfway into the spine. There’s a multitude of different products in the market and there’s also exoscopes. Depending on how you set that up, they can be very expensive or if you use it a lot you can make it worthwhile. You can marry the visualization technology to the robot, and we’re really not doing a good job of that right now. We have not explored that.
The other thing is that you can go ahead and use it for data.There’s already machine learning built into this operating software that we’re using, and so you may be using it a little bit to optimize your own construct, but why not collect data and amass data and look at what screws failed, what hardware constructs failed, and use information from individual surgeons pull that and use machine learning and other algorithms to figure out how to optimize our hardware pathways? I have my independent data. My colleague has his independent data. So we’re learning from our independent data and from papers and conferences. But why not pull that information again? We have this million dollar computer. It’s not just an arm, it’s also a computer, and there’s no reason that we can’t make those software improvements for data sharing, data collection, and improving those machine learning algorithms over time.
