A digital stereo microscope platform for microsurgery training

Abstract

We describe a software defined surgical microscope platform for developing AI applications in surgical training. The microscope has facility to merge and render multiple streams of live and/or stored video, and has the ability to enhance, annotate, and measure using a 3D position and orientation tracking forceps. A configuration mechanism controls the zoom, focus and disparity of the stereo view and stores surgeon and procedure specific configuration. The system tracks the surgical motion and analyses its quality in realtime. Several measures of quality of motion are described and can be used as a platform to develop AI applications in surgical training.