Encrypted login | home

Program Information

Iterative Reconstruction for Carbon Computed Tomography with Accurate Boundary Detection

D Shrestha

D Shrestha*, N Qin , Y Zhang , F Kalantari , X Jia , A Pompos , S Jiang , J Wang , The University of Texas Southwestern Medical Center, Dallas, TX


TU-C3-GePD-JT-5 (Tuesday, August 1, 2017) 10:30 AM - 11:00 AM Room: Joint Imaging-Therapy ePoster Theater

Purpose: In heavy ion radiation therapy, an accurate localization of the Bragg peak reduces range uncertainties and provides greater dose conformity while sparing the healthy tissues. Computed tomography using carbon ions (carbon CT) can reduce the range uncertainty by reconstructing the 3D relative electron density map of a patient body directly. Our goal is to improve the carbon CT quality by incorporating an accurate boundary detection strategy into an iterative reconstruction algorithm.

Methods: The Geant4 toolkit was used to simulate carbon CT projections of two different phantoms with inserts of various materials and line pair (lp) densities from 1.67 lp/cm through 5 lp/cm. For each phantom, 60 projections were obtained with ions/projection using 430 MeV/u ions. Each carbon ion was first projected onto a virtual plane close to the object surface along the ion tracking direction. Ions collected at the virtual plane were then used for an approximate reconstruction of the phantom. The phantom boundary was extracted from this approximate reconstruction and individual ions were tracked onto the surface of the phantom for a more accurate estimate of ion trajectory. Subsequent reconstruction was performed by the iterative algebraic reconstruction technique coupled with total variation minimization (ART-TV).

Results: The straight line path approximation for the carbon ions inside the phantom was found to be reasonably good. The standard deviation of the lateral displacement being about 0.7 mm at a depth of 20 mm. Using ART-TV along with the boundary detection technique, high quality carbon CT images were reconstructed with good contrast and high spatial resolution of more than 5 lp/cm with an imaging dose of 3mGy per scan to the isocenter.

Conclusion: Our phantom boundary detection technique coupled with ART-TV can be used to reconstruct carbon CT images with good contrast and high spatial resolution with a relatively low imaging dose.

Contact Email: