Development of an alias-free color flow imaging technique using Doppler tolerant coding for improved diagnostic ultrasound

University Of Illinois At Urbana-Champaign, Urbana IL

Investigators

Abstract

PROJECT SUMMARY Color Flow Imaging (CFI) is a ubiquitous mode on ultrasound scanners and has been around for several decades. Several limitations still exist for CFI: 1) the highest velocity that can be estimated depends on the pulse repetition frequency (PRF) of the imaging system, i.e., the Nyquist frequency, and high flow velocities lead to aliasing of the estimates, 2) for deeper structures, the PRF must be reduced, which makes aliasing artifacts even worse for fast flow, 3) in many scanners, the PRF depends on the size of the color flow map, and so for fast moving flows, e.g., the heart, smaller CFI windows are needed to maintain a higher PRF, 4) the signal-to-noise ratio (SNR) in deeper structures can be low, resulting in poorer results, and 5) tissue motion can also reduce the clarity of CFI as the flow signal can sometimes be difficult to separate from tissue signal. Clutter filters attempt to remove the tissue signal, but for fast moving tissues, like the heart, the task is difficult. Even with these limitations, CFI is an invaluable tool for diagnostic ultrasound. However, alleviating these limitations would result in better diagnostic outcomes and open the door for additional applications of this technology. The basic approach to CFI has not changed over the last few decades. In this research, we are proposing a highly innovative and fundamentally new approach for CFI that does not face the same limitations as traditional methods. To achieve this, we propose the use of Doppler-tolerant codes for estimating flow parameters. Specifically, we will use up and down hyperbolic chirps to estimate flow parameters from ultrasound echoes. We call this approach ‘chirpCFI’. Hyperbolic chirps have a unique property that scattering of the chirp from moving objects, e.g., a red blood cell, results in a time shift of the compressed waveform. The faster the object is moving, the more the compressed chirp is shifted in time. The time shift of the compressed up chirp will be in the opposite direction of the time shift of the compressed down chirp. By comparing the shifts in the compressed up and down chirps, the flow speed and direction for a scatterer can be estimated. As a result of the unique nature of the hyperbolic chirps, flow can be estimated without the worry of aliasing because the flow estimates do not depend on the PRF. The SNR is improved because the energy in the chirp is higher. Deeper structures can be imaged because the approach does not depend on PRF and the SNR is higher. Tissue clutter signal is reduced because the flow signal does not depend on multiple moving frames to provide an estimate. In this proposal, we will further optimize and validate chirpCFI in phantoms and in vivo, under challenging scenarios such as imaging the heart in motion. To successfully meet the goals of research, two specific aims are proposed. The first aim is to further optimize the hyperbolic chirp-based chirpCFI approach to provide the best image quality. The second aim is to validate and quantify image quality of chirpCFI in vivo with applications in the heart. The scientific premise of the research is that a novel flow technique can be engineered, based on using up and down hyperbolic chirps, that overcomes current limitations of long held methods for CFI.