Transforming static C-arm hardware into an active, patient-centric anatomical navigator. Eliminating 78.4% of procedural radiation and automating continuous sub-millimeter target lock during complex ERCP and interventional radiology.
Complex Endoscopic Retrograde Cholangiopancreatography (ERCP)—such as the Orlando Protocol for difficult bile duct stones (>15 mm) or stricture cannulation—imposes severe physical, radiation, and cognitive burdens on surgical teams.
Complex therapeutic ERCP procedures average 82.4 Gy·cm² in Dose Area Product (DAP). Cumulative scatter causes early cataracts, thyroid malignancies, and leukemia risks for interventionalists wearing heavy 15 kg lead aprons that cause severe chronic orthopedic spine disease.
Under sedation, natural diaphragmatic excursions (14 BPM) and involuntary patient coughs displace the anatomical region of interest (ROI) by up to 79 mm. The surgeon must drop instruments, break sterile focus, and manually re-jog the C-arm or table, compounding fatigue.
Radiopaque iodine dye rapidly dissipates under continuous bile secretion. Once contrast washes out, conventional 2D tracking algorithms lose the target completely, leaving the surgeon operating blind and risking lethal ductal perforation or avulsion.
A patented multi-level perception pipeline combined with Levenberg-Marquardt damped Image-Based Visual Servoing (IBVS) to deliver true real-time motion compensation.
A deep multi-network perception stack ensuring continuous coordinate extraction even under extreme noise, device overlap, and anatomical motion.
Locks onto global intensifier / Flat-Panel Detector (FPD) circular boundary to establish physical coordinate frame origin [270.0, 270.0].
Tracks rigid vertebral pedicle shadows and diaphragm curvature. Forms the spatial anchor for zero-contrast extrapolation.
Binds the distal duodenoscope articulation point at the ampulla of Vater, providing continuous relative mechanical grounding.
Extracts sub-pixel mathematical spline vectors tracing 0.035" hydrophilic guidewires through tortuous biliary duct geometry.
Computes 2D centroid and angular orientation theta of Dormia lithotripsy baskets, biliary stones, and stricture bifurcations.
Levenberg-Marquardt Damped Pseudoinverse: Calculates optimal instantaneous 3D Cartesian velocity commands (\(v_c\)) for the motorized C-arm and lead collimators. An adaptive gain law \(\lambda(e) = \lambda_{\text{base}} \cdot [0.80 + 0.20 \tanh(e/8.0)]\) guarantees rapid response to gross spasms (\(85\text{ mm/s}\)) while ensuring sub-millimeter damping near center without overshoot.
When contrast dissipates and Level 5 confidence drops, the Anatomical Extrapolation Engine instantly computes the calibrated vector offset from the Level 2 Vertebral Spine anchor, projecting persistent virtual biliary coordinates with zero drift.
Motorized high-density lead shutters actively follow the tracked target across the detector, restricting high-dose X-ray beams exclusively to the active ROI while providing 78.4% ALARA radiation dose shielding across peripheral tissue.
Strict human-in-the-loop foot pedal safety gating. Releasing the pedal instantly severs CAN bus actuation commands and clamps mechanical brakes in under 12 milliseconds, meeting strict IEC 62304 Class C medical safety standards.
Audited and validated across three distinct real-world clinical surgical datasets comprising over 2,650 frame tracks.
Patient undergoing single-operator cholangioscopy-guided mechanical lithotripsy for an impacted 16.5 mm common bile duct stone. Under respiratory excursions, the Dormia extraction basket oscillates across a 42 mm sweep.
Clinical Clip 1: Lithotripsy basket and guidewire locked under closed-loop visual servoing.
Direct clinical comparison against Omega FluoroShield static ROI collimation during selective cannulation of the major duodenal papilla.
Clinical Clip 2: Omni Rovis active dynamic collimator vs. background static shutter.
Surgical suite workflow constraints during a complex 71.52 mm proximal Common Bile Duct stricture dilation. High-attenuation guidewire navigation through intrahepatic ductal arborization.
Clinical Clip 3: 71.5mm CBD stricture traversal with multi-level perception cascade.
Quantified radiation dose reductions (ALARA Tier 1) and procedural time savings modeled across standard interventional operating room workflows.
| Performance Metric | Conventional Fluoroscopy | Omega FluoroShield | Omni Rovis Autonomous Autopilot |
|---|---|---|---|
| Mean DAP Exposure / Case | 82.4 Gy·cm² | 42.1 Gy·cm² | 17.8 Gy·cm² (-78.4%) |
| Occupational Eye Scatter (Staff) | 528 µGy / case | 295 µGy / case | 116 µGy / case (-78.0%) |
| Patient Motion Compensation | None (Manual Jog) | Manual Box Drag | Autonomous Closed-Loop (<33ms) |
| Contrast Washout Protection | Zero (Blind Operation) | None | L2 Skeletal Anchor Extrapolation |
| Manual Reset Interruption Rate | 12–18 resets / case | 8–12 resets / case | 0 resets (100% Tax Eliminated) |
| Hardware Actuation Safety | Manual Foot Switch | Touchscreen UI | Hardware Dead-Man Foot Pedal (<12ms Brake) |
Modeled across 450 ERCP cases/year via 8.5 minute reduction in manual C-arm re-centering and fluoroscopy dead-time.
Reduces need for maximum lead apron thickness during prolonged procedures, directly alleviating surgeon spinal disc compression.
Engineered in compliance with IEC 62304 Class C medical software architecture and ALARA radiation minimization guidelines.
Compatible with Siemens CIARTIC Move, Philips Azurion, GE Innova, and legacy robotic C-arm systems via standardized CANopen / J1939 telemetry.
