Automatic Differentiation for GPU-Accelerated 2D/3D Registration
نویسندگان
چکیده
A common task in medical image analysis is the alignment of data from different sources, e.g., X-ray images and computed tomography (CT) data. Such a task is generally known as registration. We demonstrate the applicability of automatic differentiation (AD) techniques to a class of 2D/3D registration problems which are highly computationally intensive and can therefore greatly benefit from a parallel implementation on recent graphics processing units (GPUs). However, being designed for graphics applications, GPUs have some restrictions which conflict with requirements for reverse mode AD, in particular for taping and TBR analysis. We discuss design and implementation issues in the presence of such restrictions on the target platform and present a method which can register a CT volume data set (512×512×288 voxels) with three X-ray images (512×512 pixels each) in 20 seconds on a GeForce 8800GTX graphics card.
منابع مشابه
2D / 3D Registration of X-Ray and CT Data Using Accelerated Volume Rendering and Gradient Calculation
As the role of computer-aided biopsy and surgery becomes more and more important in medical treatment, the need for a common frame of reference for all devices that participate in the operation is ever increasing. One indispensable task in that context is the determination of the patient’s position and orientation in a pre-defined world coordinate system to locate and position medical instrumen...
متن کاملGPU Accelerated Image Registration in Two and Three Dimensions
Medical image registration tasks of large volume datasets, especially in the non-rigid case, often put a heavy burden on computing resources. GPUs are a promising new approach to address computational intensive image processing tasks. We investigate recently introduced GPU hardware features that accelerate 2D and 3D rigid and nonrigid registration tasks. Our implementation is entirely GPU based.
متن کاملAutomatic localization of target vertebrae in spine surgery using fast CT-to-fluoroscopy (3D-2D) image registration
Localization of target vertebrae is an essential step in minimally invasive spine surgery, with conventional methods relying on “level counting” – i.e., manual counting of vertebrae under fluoroscopy starting from readily identifiable anatomy (e.g., the sacrum). The approach requires an undesirable level of radiation, time, and is prone to counting errors due to the similar appearance of verteb...
متن کاملHigh-performance GPU based Rendering for Real-Time, rigid 2D/3D-Image Registration in Radiation Oncology DIPLOMARBEIT
This thesis presents a comparison of high-speed rendering algorithms for the application in 2D/3D-image registration in radiation oncology. Image guided radiation therapy (IGRT) is a technique for improving the treatment of cancer with ionizing radiation by adapting the treatment plan to the current situation using 2D/3D-image registration. To accelerate this procedure, also rendering of Digita...
متن کاملAutomatic localization of vertebral levels in x-ray fluoroscopy using 3D-2D registration: a tool to reduce wrong-site surgery.
Surgical targeting of the incorrect vertebral level (wrong-level surgery) is among the more common wrong-site surgical errors, attributed primarily to the lack of uniquely identifiable radiographic landmarks in the mid-thoracic spine. The conventional localization method involves manual counting of vertebral bodies under fluoroscopy, is prone to human error and carries additional time and dose....
متن کامل