9 March 2017 Quantitative 1D diffraction signatures during dual detector scatter VOI breast CBCT
Author Affiliations +
Abstract
Dual detector VOI scatter CBCT is similar to dual detector VOI CBCT except that during the high resolution scan, the low resolution flat panel detector is also used to capture the scattered photons. Simulations show a potential use of scatter to diagnose suspicious VOIs. Energy integrated signals due to scatter (EISs) were computed for a specific imaging task involving a malignant lesion and was labelled as a hypothetical experiment (expt) result. The signal was compared to predictions (pred) using benign and malignant lesions. The ΔEISs=EISs|expt - EISs|pred displayed eye catching diffraction structure when the prediction calculation used a benign lesion. The structure occurred even when the phantom compositions were different for prediction and experiment calculations. Since the diffraction structure has a circularly symmetric behaviour because the tissues are amorphous in nature, the 2D ΔEISs patterns were transformed to 1D signals. The 1D signals were obtained by calculating the mean ΔEISs signals in rings. The mean pixel values were a function of the momentum transfer argument q = 4π sin(θ/2)/λ which ranged from 12 to 46 nm-1. The 1D signals correlated well with the 2D profiles. Of particular interest were scatter signatures between q = 20 and 30 nm-1 where malignant tissue is predicted to scatter more than benign fibroglandular tissue. The 1D diffraction signatures could allow a better method to diagnose a suspicious lesion during dual detector scatter VOI CBCT.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Robert J. LeClair, "Quantitative 1D diffraction signatures during dual detector scatter VOI breast CBCT", Proc. SPIE 10132, Medical Imaging 2017: Physics of Medical Imaging, 101324Q (9 March 2017); doi: 10.1117/12.2254480; https://doi.org/10.1117/12.2254480
PROCEEDINGS
14 PAGES


SHARE
Back to Top