Full realization of quantitative cortical brain surface measurement through stereo pair without separate tracking technology

Abstract

Intraoperative soft tissue deformation, referred to as brain shift, compromises the application of current image-guided surgery navigation systems in neurosurgery. A computational model driven by sparse data has been proposed as a cost-effective method to compensate for cortical surface and volumetric displacements. In this thesis, I propose an innovative approach to image guidance within the neurosurgical environment. More specifically, I hypothesize that the surgical operating microscope equipped with two CCD cameras and an intra-operative stereo vision (iSV) system can be used as a complete locoregional (restricted to a localized region of the body) platform to achieve conventional image guidance but also to compensate for intra-operative brain shift, all realized without the use of conventional optical tracking technologies, i.e. a trackerless image guidance approach. Locoregional therapy refers to various minimally invasive therapeutic procedures. The approach centers around locoregional, stereo pair-driven, continuous registration strategies designed to match and possibly outperform the standard-of-care instrumentation of a separately optically-tracked surgical microscope. This dissertation consists of four specific aims: (1) Development of a mock craniotomy and brain shift simulator to evaluate methods for registration and tracking of mock brain shift for evaluation of a stereo-camera platform approach. (2) Translation of the stereo-pair technology to a stereo surgical microscope and validation of brain shift measurement using stereo-pair reconstructions. (3) Development of a novel planner that takes advantage of computer vision approaches to enable the planning of a patient’s craniotomy during image-guided neurosurgery without the need of conventional tracking technologies. (4) Demonstrate a proof-of-concept framework whereby deformation-corrected image guided neurosurgery can be performed without the need for conventional tracking and using stereo-pair microscope technology.