Wearable Surgical Optical Tracking System Based on Multi-Modular Sensor Fusion

Optical tracking system (OTS) is an important part of surgical navigation and has been commonly employed in clinical practice due to high accuracy and precision. However, current OTS is bulky, which reduces the flexibility for clinical procedures. To address this drawback, we propose a wearable surgical instrument tracking system based on a multi-modular sensor fusion approach. The proposed tracking system consists of a multispectrum red-green-blue-depth (RGB-D) sensor and the accompanying tracking target. It significantly increases the agility of the tracking device while obtaining high tracking accuracy. We propose a plane-based refinement algorithm (PRA) to estimate marker centers, which ensures the accuracy and precision for the short baseline of the wearable stereo camera. A 2-D marker-based instrument identification algorithm (MIIA) has also been proposed to identify different instruments. Both simulations and real experiments have been carried out to validate the feasibility and performance of the proposed system. These accurate and robust experimental results have verified the feasibility of the proposed wearable surgical instrument tracking system.