Stereoscopic motion magnification in minimally-invasive robotic prostatectomy
Document Type
Conference Proceeding
Publication Date
1-1-2016
Journal
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume
9515
First Page
35
Last Page
45
URL with Digital Object Identifier
10.1007/978-3-319-29965-5_4
Abstract
© Springer International Publishing Switzerland 2016. The removal of the prostate is a common treatment option for localized prostate cancer. Robotic prostatectomy uses endoscopic cameras to provide a stereoscopic view of the surgical scene to the surgeon. Often, this surgical scene is difficult to interpret because of variants in anatomy and some critical structures such as the neurovascular bundles alongside the prostate, are affected by variations in size and shape of the prostate. The objective of this article is to develop a real-time stereoscopic video processing framework to improve the perceptibly of the surgical scene, using Eulerian Motion Magnification to exaggerate the subtle pulsatile behavior of the neurovascular bundles. This framework has been validated on both digital phantoms and retrospective analysis of robotic prostatectomy video.