Figures of merit for detectors in digital radiography. I. Flat background and deterministic blurring.

Digital radiography systems can be thought of as continuous linear shift-invariant systems followed by sampling. This view, along with the large number of pixels used for flat-panel systems, has motivated much of the work which attempts to extend figures of merit developed for analog systems, in particular noise equivalent quanta (NEQ) and detective quantum efficiency (DQE). A more general approach looks at the system as a continuous-to-discrete mapping and evaluates the signal-to-noise ratio (SNR) completely from the discrete data. In this paper, we study the effect of presampling blur on these figures of merit for a simple model that assumes that the background fluence is constant and that the blurring of the signal is deterministic. We find that for small signals, even in this idealized model, commonly used DQE/NEQ formulations do not accurately track the behavior of the fully digital SNR. Using these NEQ-based figures of merit would lead to different design decisions than using the ideal SNR. This study is meant to bring attention to the assumptions implicitly made when using Fourier methods.

[1]  M. Giger,et al.  Investigation of basic imaging properties in digital radiography. 2. Noise Wiener spectrum. , 1984, Medical physics.

[2]  Alan V. Oppenheim,et al.  Discrete-Time Signal Pro-cessing , 1989 .

[3]  H. V. Trees.  Detection, Estimation, And Modulation Theory , 2001 .

[4]  Harrison H. Barrett,et al.  Analytical approximations to the Hotelling trace for digital x-ray detectors , 2001 .

[5]  I. Cunningham.  Applied Linear-Systems Theory , 2000 .

[6]  K Doi,et al.  Investigation of basic imaging properties in digital radiography. I. Modulation transfer function. , 1984, Medical physics.

[7]  Harrison H. Barrett,et al.  What does DQE say about lesion detectability in digital radiography? , 2001, SPIE Medical Imaging.

[8]  Harrison H Barrett,et al.  Figures of merit for detectors in digital radiography. II. Finite number of secondaries and structured backgrounds. , 2004, Medical physics.

[9]  John A. Rowlands,et al.  Direct-conversion flat-panel x-ray imaging: reduction of noise by presampling filtration , 2000, Medical Imaging.

[10]  Keinosuke Fukunaga,et al.  Introduction to Statistical Pattern Recognition , 1972 .

[11]  Athanasios Papoulis,et al.  Probability, Random Variables and Stochastic Processes , 1965 .

[12]  Kyle J. Myers,et al.  Lesion detectability in digital radiography , 2001, SPIE Medical Imaging.

[13]  Kyle J. Myers,et al.  Megalopinakophobia: its symptoms and cures , 2001, SPIE Medical Imaging.

[14]  J T Dobbins,et al.  Effects of undersampling on the proper interpretation of modulation transfer function, noise power spectra, and noise equivalent quanta of digital imaging systems. , 1995, Medical physics.

[15]  R. F. Wagner,et al.  Unified SNR analysis of medical imaging systems , 1985, Physics in medicine and biology.

[16]  Brandon Dominic Gallas.  Signal detection in lumpy backgrounds , 2001 .

[17]  D. Jaffray,et al.  Optimization of x-ray imaging geometry (with specific application to flat-panel cone-beam computed tomography). , 2000, Medical physics.

[18]  J A Rowlands,et al.  Digital radiology using active matrix readout of amorphous selenium: theoretical analysis of detective quantum efficiency. , 1997, Medical physics.

[19]  H. Barrett,et al.  Objective assessment of image quality. III. ROC metrics, ideal observers, and likelihood-generating functions. , 1998, Journal of the Optical Society of America. A, Optics, image science, and vision.

[20]  Andrew D. A. Maidment,et al.  Linear response theory for detectors consisting of discrete arrays. , 2000, Medical physics.

[21]  R. F. Wagner,et al.  Objective assessment of image quality. II. Fisher information, Fourier crosstalk, and figures of merit for task performance. , 1995, Journal of the Optical Society of America. A, Optics, image science, and vision.