Department News

Williamson to receive Farrington Daniels Award

Jeffrey F. Williamson, PhD
Professor of Radiation Oncology

Jeffrey Williamson, PhD (Professor, Physics Division) has been selected to receive the prestigious Farrington Daniels Award from the American Association of Physicists in Medicine (AAPM), in recognition of his paper, “A linear separable two-parameter model for dual energy CT imaging of proton stopping power computation,” which was chosen as the best therapy physics paper published in Medical Physics in 2016. The award, which is the highest recognition offered by the journal, will be presented to Dr. Williamson and his co-authors at the 2017 Annual AAPM meeting in Denver, Colorado.

The paper presents a new linear, separable model (Basis Vector Model, or BVM) for extracting stopping power ratios (SPR) from patient x-ray CTs acquired at two dissimilar energies (dual energy CT or DECT). SPRs are needed to predict the depth of the Bragg peak in proton therapy: the location in the patient where the protons stop, release their residual energy in a burst of subnuclear particles. Because DECT provides two independent measurements at each tissue voxel, it can in principle disambiguate SPR dependence on tissue composition and density, supporting more accurate estimation of the Bragg peak depth, increasing confidence that tumor is accurately treated without overdosing critical organs. In comparison to the more complicated nonlinear SPR models, which are limited to vendor-reconstructed images, BVM supports model-based, statistical iterative reconstruction from the raw signals directly measured by the CT scanner, which result in more accurate and artifact-free images.

This figure shows percent error of proton stopping power prediction for three different types of common tissues: adipose-like (top row), muscle-like (middle row), and bone-like (bottom row) tissues by applying two DECT imaging methods: the BVM (basis-vector model) and VCU Torikoshi parametric fit model (VCU tPFM). The black cross indicates the prediction error by using the tissue compositions recommended by ICRU Report 44.

The BVM is a fundamental contribution leading to the award of an NCI grant (R01 CA212638, “Quantitative dual-energy ct imaging for cancer imaging and Radiation therapy applications”) awarded to Washington University. Besides Dr. Williamson, Radiation Oncology co-investigators include Drs. Jeff Bradley and Tianyu Zhao.

Read Dr. Williamson’s paper » 

The work discussed in this publication was completed during Dr. Williamson’s time at Virginia Commonwealth University (VCU) in Richmond, VA, prior to his arrival at WUSTL. The first author of the paper is VCU Medical Physics PhD student Dong Han.