Advances in Scene Reconstruction and Tracking for Endoscopic Surgery

Watch his talk here.

Surgical navigation is a promising avenue to increasing the safety and efficacy of minimally invasive interventions. Vision-based techniques, which enable tracking of the tool to tissue relationship directly from interventional images, are thought to dramatically increase the availability and use of surgical navigation as they avoid additional hardware and cost. In this talk, I will first characterize the challenges of vision-based surgical navigation and then illustrate how recent techniques for stereo reconstruction, scene flow estimation, and multi-modal learning contribute to turning image-based surgical navigation into reality.

Mathias Unberath is an Assistant Professor in the Department of Computer Science at Johns Hopkins University, core faculty in the Laboratory for Computational Sensing and Robotics, and Fellow of the Malone Center for Engineering in Healthcare.
With his group - the Advanced Robotics and Computationally AugmenteD Environments (ARCADE) Lab - he advances healthcare by creating collaborative intelligent systems that support clinical workflows. Through synergistic research on imaging, computer vision, machine learning, and interaction design, they build human-centered solutions that are embodied in emerging technology such as mixed reality and robotics.