The projects page is currently (Dec 2016) being rebuilt. - Please revisit as updates are coming online.

Quantitative Blood Volume Flow Imaging

Investigators: Drs. J. Brian Fowlkes, Oliver D. Kripfgans, Stephen Z. Pinter, and Jonathan M. Rubin

Traditional Doppler methods often rely on knowledge of the Doppler angle and geometry assumptions. The here investigated 3D Volume Flow Technique does not. We are developing a blood volume flow biomarker that measures blood volume per unit time (milliliters per minute), commonly confused with blood flow velocity (centimeters per second). The general principle of the technique is the measurement of blood volume flowing through a surface intersecting the blood vessel of interest. Two significant steps enable Quantitative Blood Volume Flow Imaging:

Computation of flux, not velocity: 3D volume flow (Q) is computed by multiplying blood flow velocity (vn in the shown lumen), as measured by color flow, by the surface area of the intersected lumen (An). Given that Q = A0 × v0 = A1 × v1 = A2 × v2, 3D volume flow is independent of angle. Specifically, An = A0 / cos(αn) and vn = v0 × cos(αn), therefore, the cosine factor cancels when A is multiplied by v. Note that the lumen is sampled with many voxels, here only one contiguous surface is shown for simplicity.

Computation of flux on the native surface: Gauss’ Theorem states that volume flow (Q) can be obtained by integrating the product Ai × vi over the surface area (S), i.e., the c-surface. In ultrasound the c-surface is the surface perpendicular to the imaging beams. Power mode data is used to weight each area (Ai) in order to correct for partial volume effects. Partial volume effects arise when an image voxel is partially inside a given lumen and partially outside.

Breast Imaging – Limited Angle Tomography

PI: Dr. Paul L. Carson

Team: Dr. H.-P. Chan, Dr. J.B. Fowlkes, Dr. M. Goodsitt, Dr. M. Roubidoux and A. Almazroa, C. A. Green, R. Jintamethasawat, X. Zhang

Breast ultrasound is an important adjunct to mammography for screening women with dense breast tissue. However, ultrasound screening leads to an increased number of visits for further diagnosis, including some unnecessary biopsies. Our goals are to improve breast ultrasound screening by: 1) performing it in a system integrated with mammography in order to increase the correlation between lesions found in the x-ray and ultrasound images and 2) using a dual-sided (top/bottom) technique to increase breast coverage and to permit speed of sound imaging for better cancer detection and discrimination. We expect the resulting 3D ultrasound in the same compression as 3D mammography will be especially helpful for younger women, who often have cancers diagnosed late and are frequently called back for unnecessary imaging studies and biopsies.