Cross-platform integration of robotic intraoperative ultrasound using system-agnostic drop-in probes: technical feasibility and workflow standardization

Technical Summary: Cross-Platform Integration of Robotic Intraoperative Ultrasound

Problem Statement
Intraoperative ultrasound (IOUS) is a critical tool for real-time image guidance in minimally invasive surgery, yet its integration into robotic workflows remains fragmented. Current solutions often rely on proprietary, platform-specific probes that limit scalability and hinder the transfer of skills between different robotic systems. The rigid shafts of traditional laparoscopic ultrasound probes constrain acquisition angles and make intracorporeal manipulation difficult. While "drop-in" probes (featuring a flexible cable and a clip for robotic graspers) offer a solution, the lack of a standardized, system-agnostic workflow across diverse robotic architectures poses a barrier to the widespread adoption of Robotic Intraoperative Ultrasound (RIOUS). There is a need to determine if a single procedural framework can be successfully applied across architecturally distinct multi-port robotic platforms using system-agnostic hardware.

Methodology
This technical feasibility study was conducted at the IRCAD Research and Training Center in Strasbourg, France, using a non-clinical, experimental setting with an abdominal ultrasound phantom. The study evaluated the reproducibility of a standardized six-step RIOUS workflow across four distinct multi-port robotic surgical systems:

1. da Vinci Xi (Intuitive Surgical)
2. Hugo™ RAS (Medtronic)
3. Toumai (MicroPort® MedBot™)
4. hinotori™ (Medicaroid Corporation)

Equipment and Protocol:

• Ultrasound System: ARIETTA 750VE (FUJIFILM Healthcare Corporation).
• Probes: Two system-agnostic drop-in linear probes were tested: the L43K (12–2 MHz, 26 mm footprint) and the L51K (15–3 MHz, 13 mm footprint). Both feature flexible cables and clips designed for standard robotic graspers.
• Phantom: A validated US-3 IOUSFAN abdominal phantom housed in a box trainer to simulate realistic trocar configurations and anatomical structures.
• Workflow: A structured six-step protocol was applied identically to all platforms:
  1. Ultrasound machine positioning relative to the robotic cart and console.
  2. Pre-activation safety checks (probe attachment, cable routing, arm interference).
  3. Trocar configuration (utilizing an additional 12-mm accessory trocar).
  4. Probe insertion under direct endoscopic vision.
  5. Console configuration and image display integration.
  6. Instrument selection and probe manipulation using standard robotic graspers.

Key Results
The study successfully demonstrated that a standardized RIOUS workflow is feasible and reproducible across all four evaluated platforms.

• Hardware Integration: Both drop-in probes were inserted and manipulated intracorporeally on all systems without probe detachment, cable entanglement, or robotic arm interference. The smaller L51K probe offered advantages in confined spaces, while the L43K provided superior image quality for larger structures.
• Image Display: Real-time B-mode imaging was successfully displayed at the surgeon's console on every system, though the integration modalities differed:
  • da Vinci Xi: Utilized native TilePro™ for simultaneous split-view display.
  • Hugo™ RAS: Required an external workflow using the Touch Surgery™ DS1 ecosystem to transmit the signal to an auxiliary monitor adjacent to the console.
  • Toumai: Used native Picture-in-Picture (PiP) functionality to overlay the ultrasound on the endoscopic view.
  • hinotori™: Employed a split-view or triple-view functionality via the system tower connection.
• Instrumentation: Stable grasping and manipulation were achieved using standard robotic instruments on all platforms. While specific preferred graspers varied by system (e.g., Maryland bipolar forceps for da Vinci Xi, Cadiere forceps for Hugo RAS and Toumai, Versatile Grasping Forceps for hinotori), no significant difference in clip grip stability was observed between the two probe models.
• Consistency: Core procedural steps (machine positioning, trocar strategy, probe insertion) remained consistent across systems, with differences confined to room layout, visualization modality, and specific instrument preferences.

Key Contributions

1. First Multi-Platform Demonstration: This is the first technical demonstration of a RIOUS workflow performed across four architecturally distinct robotic systems under identical experimental conditions.
2. Standardized Protocol: The study established a reproducible, platform-agnostic six-step protocol that addresses the technical specificities of each system while maintaining a common procedural framework.
3. System-Agnostic Viability: The research validates that system-agnostic drop-in probes can facilitate cross-platform skill transfer, reducing the economic and technical burden of developing proprietary hardware for each robotic manufacturer.
4. Educational Resources: The study generated four structured, video-based reference materials documenting the complete workflow for each platform, intended to support surgeon training and the integration of RIOUS into robotic surgical curricula.

Significance and Claims
The paper claims that a reproducible, platform-agnostic RIOUS workflow is technically feasible, suggesting that proficiency in robotic ultrasound manipulation may be transferable across different robotic systems without the need for entirely separate learning curricula. This finding addresses a critical gap in the field, potentially preventing the creation of "platform-specific silos" that limit the adoption of intraoperative ultrasound.

The authors highlight that this work lays the procedural groundwork for future applications in telesurgery and tele-mentoring, where a remote operator could manipulate a drop-in probe under image guidance while a primary surgeon operates. Furthermore, the study posits that RIOUS can partially compensate for the lack of haptic feedback in robotic surgery by providing an additional layer of visual information for tissue characterization and lesion localization.

The authors remain modest regarding their claims, acknowledging that the study was limited to a phantom model and did not include single-port systems or formal telesurgical simulations. They emphasize that while the procedural framework is transferable, the cross-platform transferability of sonographic interpretation skills requires further formal investigation. The study concludes that standardized video documentation and system-agnostic hardware represent a practical step toward collaborative, image-guided robotic surgery.