Unencrypted login | home

Program Information

Estimation of the 2-D Presampled MTF of a Digital Flat Panel Detector Using An Edge Test Device


J Wells

J Wells1,2*, J Dobbins1-4, (1) Carl E. Ravin Advanced Imaging Laboratories, Duke University Medical Center, Durham, NC (2) Medical Physics Graduate Program, Duke University Medical Center, Durham, NC (3) Department of Radiology, Duke University Medical Center, Durham, NC (4) Department of Biomedical Engineering, Duke University, Durham, NC

TU-A-218-1 Tuesday 8:00:00 AM - 9:55:00 AM Room: 218

Purpose: In this work, we report on the novel application of an opaque edge test device for practical estimation of the 2-D modulation transfer function (MTF) of a digital flat panel (DFP) detector.

Methods: We estimated the 2-D MTF of a prototype GE Revolution XQ/i DFP detector (GE Healthcare, Waukesha, WI) using an opaque edge test device. The original work by Samei et al. [Med. Phys. 25, 102-113 (1998)] served as the basis of our methodology to which we also contributed further theoretical analysis regarding sub-pixel precision of 1-D MTF measurements at arbitrary angle. The edge was imaged five times at each of ten angles relative to the sampling lattice: 1.33, 7.59, 20.22, 30.29, 37.90, 48.07, 58.85, 69.78, 78.13, and 88.02 degrees. The 1-D MTF was computed from edge data at each angle using Fourier analysis. The 2-D MTF was estimated through a simple surface fit and linear interpolation scheme based on Delaunay triangulation of the collection of 1-D MTF data.

Results: All angles yielded nearly identical 1-D MTFs with relative standard deviations of 0.39%, 1.3%, 4.2%, and 5.6% at 1, 2, 3, and 4 cycles/mm, respectively. These measurements exhibited no discernible patterns with respect to acquisition angle indicating that the 2-D MTF has approximate circular symmetry at frequencies below 4 cycles/mm. Although this assertion agrees with the conclusions of other authors, blur due to the finite rectangular pixel aperture may merit further investigation at the highest spatial frequencies where the MTF may not behave in a circularly symmetric manner.

Conclusions: We present for the first time the practical estimation of the 2-D MTF of a DFP detector using an edge test device. The general availability, acceptance, and ease of implementation of edge-based MTF assessments make this a valuable technique for 2-D MTF measurement in academic, industrial, and clinical settings.

Contact Email