 |
Samuel Richard Ph.D. student Medical Biophysics University of Toronto |
||
|
Research
Academic
Pretty fun
stuff
Coffee break stuff
|
Cascaded Systems Analysis Fourier-Based Metrics of Detector Performance A practical, prevalent approach to imaging performance characterization focuses on the measurement and modeling of spatial resolution and noise in terms of the modulation transfer function (MTF), noise-power spectrum (NPS), noise-equivalent quanta (NEQ), and detective quantum efficiency (DQE). Such metrics provide a valuable approach to system characterization and optimization, with widespread application in the development of novel imaging systems. Introduction Cascaded systems analysis (CSA) provides a powerful analytical tool that describes the signal and noise transfer characteristics, in terms of the Fourier-based metrics mentioned above, for various imaging systems (i.e., Image intensifiers, portal imagers, Flat-panel detectors and so on) in a manner that is physically intuitive and identifies the factors that limit imaging performance. Assumptions inherent to CSA include linearity's mean response i.e., signal is linear as a function of input quanta and obeys the superposition of two or more inputs, shift invariance i.e., image is independent of the location of the input, and stationarity, first- and second-order statistics—i.e., the mean and variance—are spatially and temporally invariant. Although real physical imaging systems never completely satisfy these requirements, they are typically assumed to hold over a range of relevant conditions.
CSA represents each physical process in the imaging chain as a gain stage e.g., conversion of x rays into optical photons, a spatial spreading (blurring) stage e.g., spreading of optical photons in the scintillator, or a sampling stage e.g., readout of the detector signal at locations according to the pixel matrix.
References:
|
||
