Interdisciplinary Initiatives Program Round 9 - 2018

Steven Chu, Physics and Molecular & Cellular Physiology
Jeremy Dahl, Radiology (Pediatric Radiology)

Ultrasound (US) imaging is widely used in medical diagnosis. It is the least expensive imaging modality, does not expose ionizing radiation to patients, and offers near-real time imaging speeds. Despite these advantages, anatomic features are frequently concealed in ultrasound images due to insufficient image contrast.

In preliminary data, we showed that various biological tissues including calyces in pig kidney, fat layer in salmon, and glioblastoma tumors in mouse brain are revealed in nonlinear, difference-frequency ultrasound (dfUS) imaging that are not visible in conventional linear contrast images or in conventional second harmonic US.

We also performed dfUS Doppler-flow measurements using FDA-approved microbubbles, which are an ultrasound contrast agent. We demonstrated that the nonlinear Doppler signal is no longer constrained by the trade-off between the axial resolution (small Δz = v_soundΔt) and the Doppler frequency resolution (small Δf) set by the mathematical limit known as the Fouriertransform limit, ΔtΔf ≥ 1/2π.  The simultaneous measurement of slow fluid flow with millimeter spatial resolution is possible. The nonlinear measurement of fast blood flows also eliminates the aliasing ambiguity, analogous to motion picture films where the sampling interval (for example, 30 frames/second) can lead to the appearance of the spokes of a wagon wheel going backwards or forwards with arbitrary speed.

This seed grant application seeks to develop a real-time imaging system for nonlinear ultrasound based on our preliminary nonlinear imaging data. The acquisition speed of the proposed system should be < 0.5 millisecond per directed sound pulse propagating into tissue (often time referred as an A-line scan) for depths up to 20 cm, and an imaging rate of > 30 frames per second. The system should allow us to image cancer tumors and other tissues that generate strong nonlinear contrast with a spatial resolution of ≤ 1 millimeter. We should be able to image both slow blood flows with resolution approaching ~ 1 cm/s, and fast flows > 100 cm/s with the same spatial resolution. Thus, the spatial visualization of plaque buildup in coronary arteries with millimeter resolution may also be possible.

This seed grant will fund the exploration of imaging of cancer tumors with a practical imaging system. We also will pursue the ambitious goal of imaging of coronary arteries in living animals by synchronizing the image acquisition with the heartbeat. This demonstration, if successful, will allow the non-invasive imaging of most of the anterior and posterior coronary arteries without catheters or ionizing radiation. Today, primary care physicians are reluctant to order expensive and invasive procedures until symptoms appear. As a result, noninvasive early diagnosis of coronary artery diseases is currently unavailable.

The availability of a real-time nonlinear ultrasound imaging system would open up new avenues for applications in medical diagnosis and treatment.