A new scheme for curved needle segmentation in three-dimensional ultrasound images

Abstract

Ultrasound image guided needle insertion is the method of choice for a wide variety of medical diagnostic and therapeutic procedures. When flexible needles are inserted in soft tissue, these needles generally follow a curved path. Segmenting the trajectory of the needles in ultrasound images will facilitate guiding them within the tissue. In this paper, a novel algorithm for curved needle segmentation in three-dimensional (3D) ultrasound images is presented. The algorithm is based on the projection of a filtered 3D image onto a two-dimensional (2D) image. Detection of the needle in the resulting 2D image determines a surface on which the needle is located. The needle is then segmented on the surface. The proposed technique is able to detect needles without any previous assumption about the needle shape, or any a priori knowledge about the needle insertion axis line.

Line filtering for surgical tool localization in 3D ultrasound images

“…Most tool localization techniques for 3D ultrasound are based on projections and Hough transform or Radon transform variants [9,10,11,12,13,14,15]. One general disadvantage of projection methods is their computational complexity, even though almost real-time speed can be achieved by using multiresolution [16,17] and Graphics Processing Unit (GPU) implementation [18].…”

Calibration of EM Sensors for Spatial Tracking of 3D Ultrasound Probes

“…The images were processed to enhance the needle's appearance. Following the needle segmentation method described in Aboofazeli et al [2], anisotropic diffusion [18] is first used for speckle reduction and smoothing with the subsequent application of an exponential contrast enhancement filter. The contrast adjustment filter has the form…”

Robotic Needle Steering: Design, Modeling, Planning, and Image Guidance

“…US images tend to be noisy, due to reflections, reverberations, shadows, air pockets, and biological speckle, which makes needle localization challenging. Some needle localization methods use 2D images [15,20,53], while others compound a 3D volume from a tracked sweep of 2D images [2,21]. For completeness, we note that, due to current limitations on voxel resolution and transfer speed, 3D US probes have not been practical for image-based needle guidance.…”

“…They segmented the needle in the projections with the Hough transform and then reconstructed the needle from its 2D projections as a straight line. Aboofazeli et al recently localized curved non-planar needles in 3D space [2]. They pre-filtered a compounded 3D US volume and produced series of 2D images by ray casting.…”