Improved image-guidance has led to the implementation of minimally invasive alternatives to many complex procedures that are routinely performed across many disciplines. While percutaneous surgery benefits the patient, the task load for the surgeon increases drastically. As a consequence, there is substantial risk of failure that, depending on the intervention, is associated with poor functional outcome, need for revision surgery, or more severe complications. Assisting the surgeon in interventional decision making via advanced imaging, image processing, and visualization is, therefore, a key step in improving outcomes. In this talk, I will present our recent work on human-centric end-to-end systems that assist surgeons in improving on-task performance. In the context of image-guided interventions, this consists of acquiring images that contain information specific to the task, analyzing the images to extract cues otherwise inaccessible to the surgeon, and augmenting the perception of the surgeon to inform a decision that, finally, enables optimal actions.
Mathias Unberath is currently a postdoctoral fellow in the Laboratory for Computational Sensing and Robotics at Johns Hopkins University. He holds a BSc in Physics, a MSc in Optical Technologies, and a PhD in Computer Science from the Friedrich-Alexander-Universität Erlangen-Nürnberg from which he graduated summa cum laude in 2017. Previously, he was appointed as ERASMUS scholar at the University of Eastern Finland in 2011 and DAAD fellow at Stanford University throughout 2014. Mathias has won numerous awards including the “SPIE Outstanding Thesis Award” for his Master’s Thesis, and the “German Society of Medical Image Processing Award” and the “Staedtler Award” for his Dissertation. His research focuses on Multi-modal Imaging Systems, Machine Learning for interventional medicine, and Augmented Reality in the context of medical education and percutaneous procedures