Research Analyzer
← Back ICRA 2026

Real-Time SLAM-Guided Closed-Loop Photodynamic Therapy with Pixel-Accurate Light-Dose Control

Hyesung Lee, Sungwook Yang

PDF

AI summary

Key figure (auto-extracted from paper)
A single handheld probe enables real-time, closed-loop photodynamic therapy with pixel-accurate dose control and simultaneous diagnosis using SLAM-guided tracking.
Photodynamic therapy SLAM closed-loop control digital micromirror device endomicroscopy dose control

Problem

Current handheld photodynamic therapy systems are operator-dependent, lose accuracy during motion, and lack simultaneous diagnosis or pixel-level dose feedback.

Approach

The system fuses optical tracking with endomicroscopic images via an extended Kalman filter to steer a digital micromirror device in real time, enforcing pixel-level light-dose accumulation and control.

Key results

  • Sub-millimeter 6-DoF probe tracking (0.3 mm RMSE) in texture-limited scenes
  • Camera-to-DMD spatial registration error under 5.2 µm
  • Pixel-accurate closed-loop dose control with uniformity within ±0.186 mJ/cm²
  • Real-time simultaneous photodynamic diagnosis and therapy at 15–30 Hz

Why it matters

Provides surgeons with a motion-robust, single-probe platform for precise, tissue-sparing photodynamic treatment in complex anatomical regions.

Abstract

Precise light-dose delivery is essential for pho- todynamic therapy (PDT), yet current handheld systems re- main operator-dependent and lose accuracy under motion. We present a SLAM-guided, closed-loop control framework that enables co-temporal and co-spatial photodynamic diagnosis (PDD) and PDT with a single handheld endomicroscopic probe, while enforcing pixel-level dose control. The probe integrates a fiber bundle that shares a common optical path for both PDD and PDT and is paired with a digital micromirror device (DMD) for μm-scale pattern projection. An extended Kalman filter fuses optical-tracking measurements with texture-limited endomicroscopic images at 30 Hz, providing robust six-degree- of-freedom pose estimates that expand the probe’s effective field of view and drive real-time pattern updates. A dose-map SLAM algorithm accumulates light dose over the reconstructed lesion surface during handheld scanning, while pixel-level dose control is enforced by referencing previously accumulated light at each location. Quantitative evaluation shows a spatial registration error between diagnostic and therapeutic systems within 5.2 μm. Experiments on fluorescence phantoms achieved sub- millimeter localization accuracy (0.3 mm RMSE), significantly outperforming vision-only and tracker-only baselines. Finally, tests on targets with quadrant-specific dose limits confirmed SLAM-based dose control, achieving dose uniformity within ±0.186 mJ/cm2 across millimeter-scale regions.

Index terms

Medical Robots and Systems SLAM Hardware-Software Integration in Robotics

Related papers