AAOS Annual Meeting Town Hall: Executive shares industry perspective on AI’s role in orthopaedic surgery

Editor’s note: This article is the second in a two-part series covering the Artificial Intelligence (AI) Town Hall at the AAOS 2025 Annual Meeting. Part one, “Town hall details how AI is transforming orthopaedics,” was published in the June issue of AAOS Now.

AI has become ubiquitous, with applications in imaging, self-driving cars, reasoning agents, and, of course, orthopaedic surgery. In orthopaedics, AI is improving surgical planning, patient monitoring, device navigation, and many other areas.

The AAOS 2025 Annual Meeting hosted a Town Hall titled “The use of artificial intelligence in orthopaedic surgery,” featuring varied perspectives on the evolving landscape of these tools. Following an overview of the basics of AI from several surgeons leading the field, the session pivoted to focus on industry perspectives on applications of AI in orthopaedics.

Siddarth Satish, vice president of digital innovation for surgical technologies at Stryker, outlined ways AI can improve the surgical environment. “There’s a ton of excitement and a lot of development in this space,” Satish said. “It’s gotten to the point where there are some very advanced concepts in play, like so-called agentic AI, [which makes decisions with minimal human support].”

Satish believes healthcare is poised to adopt many of these technologies because the sector is already becoming increasingly digital. In the OR, many tools are equipped with user interfaces. Even implants are getting smart. In addition, there are many rich orthopaedic data sources to inform AI algorithms: electronic health records, imaging, even OR cameras.

“Many forms of decision-making, whether it’s surgical approaches or, more broadly, treatment planning, can become more personalized and patient-specific by using AI algorithms,” Satish said. “In the context of arthroplasty, consider a world in which you have truly accessible automated planning.”

AI in the OR
Modern surgical environments have become highly digital, with smart implants, connected devices, and algorithmically assisted tools — advances that are powering a variety of insights. Preoperatively, AI supports risk stratification, treatment planning, and 3D modeling from imaging data. Postoperatively, AI can improve patient care by tracking wearables and improving image analysis.

AI is making surgical tools even smarter. Navigation and guidance systems perform segmentation tasks pre- and intraoperatively. Stereotactic navigation is becoming increasingly enabled by AI algorithms. In bone resection, for example, a navigated drill is tracked by a stereotactic camera, essentially providing GPS for the instrument. Because the algorithm knows where the drill is in relation to the patient, this intelligent device can give surgeons real-time notifications around critical anatomy.

Satish also pointed out that surgical suites have their own cameras, which can offer a bird’s-eye view of almost everything that is happening in the room. Hospitals can now squeeze more value from these videos with an emerging technology called vision language models.

These AI tools can transcribe entire cases, providing incredibly detailed, step-by-step accounts of what is happening during a surgery. These databases will also be searchable and could become invaluable tools as surgeons go to specific points in a video to observe how a particularly challenging step was performed.

The possibilities are broad. Another new technology measures blood loss by scanning sponges and estimating the hemoglobin content in each one. “The surgeons I’ve spoken with don’t lose very much blood, usually about 100 to 200 ccs,” Satish said. “But in some cases, it may be significantly more, or we might be overestimating blood loss. Technologies like this can eliminate the guesswork and bring more precision, clarity, and accuracy to blood-loss estimations.”

Challenges with implementation
AI still has a long way to go before it is fully accepted, and there are a number of hurdles. Satish noted that the industry must be particularly concerned about ethics and privacy, which means conducting robust bias and risk evaluations and continuously monitoring performance in the clinic.

“It’s incumbent upon us as device manufacturers to work closely with our customers to make sure that feedback loop is there, whether it’s receiving clinical feedback or truly automating the process of gathering data-driven feedback,” Satish said.

Although AI has made important inroads in orthopaedic surgery, Satish believes the technology has not fully entered the industry’s consciousness, particularly when it comes to problem-solving. “AI is such a new type of technology that we’re not often looking at it to provide answers,” he said. “We’re looking for something else as an answer.”

Rather than default to mechanical or electromechanical approaches to address problems, Satish encouraged session attendees to keep an open mind about AI, which he believes can provide the best possible solutions in a number of situations. As these technologies become more accessible, he believes utilization will increase.

“The future of orthopaedics isn’t just AI-driven,” Satish said. “You all have a huge part to play in how this unfolds. There’s a saying out there that it’s not AI that will replace radiologists — it’s other radiologists who are using AI. Orthopaedic surgeons are perfectly positioned to adopt these next-generation technologies and help lead the way.”

Josh Baxt is a freelance writer for AAOS Now.