3 March 2017 Visual tracking for multi-modality computer-assisted image guidance
Author Affiliations +
Abstract
With optical cameras, many interventional navigation tasks previously relying on EM, optical, or mechanical guidance can be performed robustly, quickly, and conveniently. We developed a family of novel guidance systems based on wide-spectrum cameras and vision algorithms for real-time tracking of interventional instruments and multi-modality markers. These navigation systems support the localization of anatomical targets, support placement of imaging probe and instruments, and provide fusion imaging. The unique architecture – low-cost, miniature, in-hand stereo vision cameras fitted directly to imaging probes – allows for an intuitive workflow that fits a wide variety of specialties such as anesthesiology, interventional radiology, interventional oncology, emergency medicine, urology, and others, many of which see increasing pressure to utilize medical imaging and especially ultrasound, but have yet to develop the requisite skills for reliable success. We developed a modular system, consisting of hardware (the Optical Head containing the mini cameras) and software (components for visual instrument tracking with or without specialized visual features, fully automated marker segmentation from a variety of 3D imaging modalities, visual observation of meshes of widely separated markers, instant automatic registration, and target tracking and guidance on real-time multi-modality fusion views). From these components, we implemented a family of distinct clinical and pre-clinical systems (for combinations of ultrasound, CT, CBCT, and MRI), most of which have international regulatory clearance for clinical use. We present technical and clinical results on phantoms, ex- and in-vivo animals, and patients.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ehsan Basafa, Ehsan Basafa, Pezhman Foroughi, Pezhman Foroughi, Martin Hossbach, Martin Hossbach, Jasmine Bhanushali, Jasmine Bhanushali, Philipp Stolka, Philipp Stolka, } "Visual tracking for multi-modality computer-assisted image guidance", Proc. SPIE 10135, Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling, 101352S (3 March 2017); doi: 10.1117/12.2254362; https://doi.org/10.1117/12.2254362
PROCEEDINGS
7 PAGES


SHARE
Back to Top