Medtronic PLC is preparing to introduce Touch Surgery Aide, an AI compute platform built to run inside the operating room and connect every stage of a surgical case. The system forms the core of the Touch Surgery Ecosystem and is scheduled for an early showcase at the Society of Robotic Surgery 2026 Annual Meeting.
Touch Surgery Aide uses NVIDIA accelerated computing infrastructure, specifically Holoscan, CUDA, and TensorRT, to run multiple computer-vision and multimodal models simultaneously on live surgical video. The platform processes both visual feeds and procedural context in real time, generating actionable overlays without requiring surgeons to leave the sterile field.
The first cleared application is Instrument Exit Point (IEP), which received FDA clearance for use with the Hugo RAS system. IEP employs computer vision to detect when selected instruments move beyond the visible field and issues an immediate visual notification. This marks Medtronic’s initial step toward embedding real-time AI directly into robotic procedures.
Hugo itself has accumulated substantial clinical volume. The system received FDA clearance for urologic procedures in December 2025 and, by June 2026, Medtronic had filed 510(k) submissions to add general surgery and gynecologic indications. The robot has already been used in tens of thousands of cases across more than 35 countries.
Medtronic executives frame the platform as a shift from hand-assist technologies to team-level decision support. Chief technology officer Jim Peichel noted that real-time AI is moving surgery from tools that guide the hand to systems that help teams work faster and with greater situational awareness.
From a commercial standpoint, Touch Surgery Aide gives Medtronic a software layer that can be updated after hardware installation. Hospitals that purchase Hugo gain access to an expanding library of AI applications without additional capital equipment purchases, a model that contrasts with competitors whose AI features remain more tightly coupled to specific system generations.
The architecture also supports tele-mentoring and post-case review, allowing a single expert to oversee multiple rooms or review annotated video for quality improvement. This capability is particularly relevant for Medtronic’s stated goal of expanding access beyond high-volume academic centers.
Longer term, the company intends to extend the same compute platform from robotic-assisted cases into conventional laparoscopic procedures, broadening the addressable installed base. The combination of FDA-cleared hardware clearances, NVIDIA infrastructure, and a modular AI application model gives Medtronic a concrete path to increase procedure volume and recurring software value per installed Hugo system.