Max Robotics — Global Service Robot Directory + US Certification

About

Thomas Kurmann is a computer vision researcher specializing in surgical instrument analysis, with a particular focus on applying deep learning techniques to computer-assisted interventions and minimally invasive surgery. His work addresses critical challenges in intraoperative tool detection, segmentation, pose estimation, and tracking — capabilities essential for advancing robotic and computer-assisted surgical systems. Kurmann's most influential contribution, "Articulated Multi-Instrument 2-D Pose Estimation Using Fully Convolutional Networks" (2018, 136 citations), introduced a pioneering deep neural network approach to detecting articulated surgical instrument poses in video, tackling one of the field's most persistent technical hurdles. Complementing this, his involvement in the widely recognized 2017 Robotic Instrument Segmentation Challenge (57 citations) helped establish community benchmarks that have shaped evaluation standards across the field. His comparative evaluation study (50 citations) further provided researchers with systematic insights into the relative strengths of competing segmentation and tracking approaches. More recently, his "Mask then Classify" framework (2021) advanced multi-instance instrument segmentation, improving upon semantic segmentation limitations to deliver more precise tool-type identification. Collectively, Kurmann's body of work has accumulated over 260 citations, reflecting meaningful and growing influence on the intersection of surgical robotics and medical computer vision.