3D ultrasound imaging with active instrumented needle tip tracking: An ex vivo study

Javad Rostami, Christian Baker, Weidong Liang, Simeon J. West, Tom Vercauteren, Sunish Mathews, Adrien E. Desjardins, Edward Z. Zhang, Paul Beard, Sebastien Ourselin, Laura Peralta Pereira, Wenfeng Xia

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

96 Downloads (Pure)

Abstract

Ultrasound (US) imaging is widely used for guiding percutaneous needle procedures, but accurate visualisation of the needle tip and patient anatomy in 3D and in real-time is challenging. We introduce a system that integrates 3D US imaging with a needle tip tracking modality for guiding minimally invasive procedures such as nerve blocks and tumour biopsies. Using a 256-element sparse spiral array for real-time 3D imaging, the system also provides precise 3D tracking information of the needle tip with an integrated fibre-optic US sensor (FOS) that communicates with the US imaging probe. FOS data were beamformed to generate an image of the needle tip, precisely pinpointing its location on the B-mode images. The tracking accuracy of the systems was tested in a water tank using a 3D translation stage. The accuracy in the vertical direction (Z, depth) was 0.32 mm ± 0.16 mm (RMSE ± SD). In the lateral and elevational directions, the accuracies were 1.21 mm ± 0.55 mm and 0.63 mm ± 0.15 mm, respectively. The clinical translational potential of the system was further demonstrated with needle insertions into ex vivo chicken tissue. Future development aims to enhance image quality and tracking speed, offering significant advancements in the efficiency and safety of needle-based interventions.
Original languageEnglish
Title of host publication2024 IEEE International Ultrasonics Symposium (IUS)
PublisherIEEE
Number of pages4
Publication statusPublished - 30 Sept 2024

Fingerprint

Dive into the research topics of '3D ultrasound imaging with active instrumented needle tip tracking: An ex vivo study'. Together they form a unique fingerprint.

Cite this