Theoretical analysis of high-resolution digital mammography

Phys Med Biol. 2006 Jun 21;51(12):3041-55. doi: 10.1088/0031-9155/51/12/003. Epub 2006 May 31.

Abstract

The performance of a high-resolution charge coupled device-based full-field digital mammography imager was analysed using a mathematical framework based on an adaptation of cascaded linear systems theory described by other investigators. This work has been conducted in order to understand the impact of various design parameters on the physical performance characteristics of the imager. Specifically, the effect of pixel size, scintillator thickness and packing density, x-ray spectra, air kerma, dark current, charge integration time, and pixel fill-factor on the frequency dependent detective quantum efficiency was studied using a charge-coupled device as a reference platform. The imaging system was modelled as a series of physical processes with gain and spatial spreading. For each stage, the signal and noise power spectra were computed and propagated through the imaging chain as inputs to subsequent stages. Good agreement between experimental and theoretical predictions was obtained for various x-ray spectral conditions that were investigated. The modulation transfer function, MTF(f) and detective quantum efficiency DQE(f) characteristics obtained in this study are encouraging and comparable to other digital mammography systems. The results of this study strongly suggest the feasibility of large area scintillator-based digital mammography imagers with pixel sizes below 100 microm.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms
  • Computer Simulation
  • Computer-Aided Design*
  • Equipment Design
  • Equipment Failure Analysis
  • Feasibility Studies
  • Information Storage and Retrieval / methods
  • Mammography / methods*
  • Models, Theoretical*
  • Radiographic Image Enhancement / methods*
  • Radiographic Image Interpretation, Computer-Assisted / methods*
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Signal Processing, Computer-Assisted*