Motion Correction for 3D Super-resolution Ultrasound Imaging

Before and after motion correction.

Localization-based super-resolution methods rely on physiological flow to relocate or replenish sparsely distributed microbubbles inside. This results in long acquisition times in the order of seconds to generate a single super-resolved frame (excluding the data processing time) due to the slow flow in microvasculature.

A clinical method will require a handheld ultrasound scan. Through multiple frames acquired over a duration of seconds, handheld probe movement, respiration, and cardiac motion will significantly degrade the image quality and resolution. Using retrospective data, it was possible to implement a [2D motion correction algorithm for super-resolution ultrasound imaging] (https://ieeexplore.ieee.org/document/8334280). This will form the basis of a 3D motion correction algorithm to be applied on newly acquired images. This project requires finding most suitable motion estimation and correction method for ultrasound imaging particularly for 3D super-resolution imaging.

Avatar
Sevan Harput
Associate Professor in Electrical and Electronic Engineering

Sevan is a Associate Professor in the at Division of Electrical and Electronic Engineering, London South Bank University, where he leads the SPEED Ultrasound Lab. The ultrasound researchlab aims to develop new imaging and sensing technologies using acoustic waves at ultrasonic frequencies.

Publications

High frame rate 3-D ultrasound imaging technology combined with super-resolution processing method can visualize 3-D microvascular …

The structure of microvasculature cannot be resolved using conventional ultrasound (US) imaging due to the fundamental diffraction …