Portable Dual Sensor Large Area Visualization System for Robotic Laser Surgery

Abstract

Laser scalpels are precise, dexterous, and efficient tools for soft tissue surgeries. However, surgical lasers are hard to control manually, require experience, and are often incompatible with conventional intraoperative imaging sensors. Integrating compatible sensing technology with laser scalpels for precise soft tissue surgery opens vital avenues for widespread adoption and previously unrealized automation. This paper proposes a dual-sensor strategy to generate high-resolution surgical scene visualization based on surgeon feedback for robotic laser surgery. The proposed method uses a coarse depth sensor to localize the tissue of interest in the surgical scene, and a fine optical coherence tomography (OCT) sensor to create a detailed (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$&lt;30 \mu ~\mathrm{m}$</tex> lateral resolution) tissue representation. The method achieves RMSE error in the range of 0.0878mm to 0.102mm in large-area tissue reconstruction and 0.050mm to 0.427mm in pattern-based laser ablation using user feedback over various fiducial samples. The findings demonstrate the proposed system's capability in large-area tissue imaging for precise laser-based surgery.