A framework for noise-power spectrum analysis of multidimensional images.

A methodological framework for experimental analysis of the noise-power spectrum (NPS) of multidimensional images is presented that employs well-known properties of the n-dimensional (nD) Fourier transform. The approach is generalized to n dimensions, reducing to familiar cases for n = 1 (e.g., time series) and n = 2 (e.g., projection radiography) and demonstrated experimentally for two cases in which n = 3 (viz., using an active matrix flat-panel imager for x-ray fluoroscopy and cone-beam CT to form three-dimensional (3D) images in spatiotemporal and volumetric domains, respectively). The relationship between fully nD NPS analysis and various techniques for analyzing a "central slice" of the NPS is formulated in a manner that is directly applicable to measured nD data, highlights the effects of correlation, and renders issues of NPS normalization transparent. The spatiotemporal NPS of fluoroscopic images is analyzed under varying conditions of temporal correlation (image lag) to investigate the degree to which the NPS is reduced by such correlation. For first-frame image lag of approximately 5-8%, the NPS is reduced by approximately 20% compared to the lag-free case. A simple model is presented that results in an approximate rule of thumb for computing the effect of image lag on NPS under conditions of spatiotemporal separability. The volumetric NPS of cone-beam CT images is analyzed under varying conditions of spatial correlation, controlled by adjustment of the reconstruction filter. The volumetric NPS is found to be highly asymmetric, exhibiting a ramp characteristic in transverse planes (typical of filtered back-rojection) and a band-limited characteristic in the longitudinal direction (resulting from low-pass characteristics of the imager). Such asymmetry could have implications regarding the detectability of structures visualized in transverse versus sagittal or coronal planes. In all cases, appreciation of the full dimensionality of the image data is essential to obtaining meaningful NPS results. The framework may be applied to NPS analysis of image data of arbitrary dimensionality provided the system satisfies conditions of NPS existence.

[1]  R. Bracewell.  The Fourier Transform and Its Applications , 1966 .

[2]  J. Bendat,et al.  Random Data: Analysis and Measurement Procedures , 1971 .

[3]  J. C. Dainty,et al.  Image Science: Principles, Analysis and Evaluation of Photographic-Type Imaging Processes , 1974 .

[4]  K. Hanson,et al.  Detectability in computed tomographic images. , 1979, Medical physics.

[5]  R. F. Wagner,et al.  Application of information theory to the assessment of computed tomography. , 1979, Medical physics.

[6]  L. Feldkamp,et al.  Practical cone-beam algorithm , 1984 .

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

[8]  John G. Proakis,et al.  Probability, random variables and stochastic processes , 1985, IEEE Trans. Acoust. Speech Signal Process..

[9]  M. Rabbani,et al.  Detective quantum efficiency of imaging systems with amplifying and scattering mechanisms. , 1987, Journal of the Optical Society of America. A, Optics and image science.

[10]  M F Kijewski,et al.  The noise power spectrum of CT images. , 1987, Physics in medicine and biology.

[11]  Larry E. Antonuk,et al.  Considerations for High Frame Rate Operation of Two-Dimensional a-Si:H Imaging Arrays , 1993 .

[12]  A Fenster,et al.  A spatial-frequency dependent quantum accounting diagram and detective quantum efficiency model of signal and noise propagation in cascaded imaging systems. , 1994, Medical physics.

[13]  M J Yaffe,et al.  Analysis of the spatial-frequency-dependent DQE of optically coupled digital mammography detectors. , 1994, Medical physics.

[14]  H. Blume,et al.  DQE(f) of four generations of computed radiography acquisition devices. , 1995, Medical physics.

[15]  Perry Sprawls,et al.  The Expanding role of medical physics in diagnostic imaging , 1997 .

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

[17]  J Yorkston,et al.  Empirical and theoretical investigation of the noise performance of indirect detection, active matrix flat-panel imagers (AMFPIs) for diagnostic radiology. , 1997, Medical physics.

[18]  I. Blevis,et al.  Digital radiology using active matrix readout of amorphous selenium: construction and evaluation of a prototype real-time detector. , 1997, Medical physics.

[19]  Ruola Ning,et al.  Selenium flat-panel detector-based volume tomographic angiography imaging: phantom studies , 1998, Medical Imaging.

[20]  J A Rowlands,et al.  Digital x-ray imaging using amorphous selenium: reduction of aliasing. , 1998, Medical physics.

[21]  Jeffrey H. Siewerdsen,et al.  DQE and system optimization for indirect-detection flat-panel imagers in diagnostic radiology , 1998, Medical Imaging.

[22]  J Yorkston,et al.  Signal, noise power spectrum, and detective quantum efficiency of indirect-detection flat-panel imagers for diagnostic radiology. , 1998, Medical physics.

[23]  Jean-Pierre Moy.  Image quality of scintillator-based x-ray electronic imagers , 1998, Medical Imaging.

[24]  K Stierstorfer,et al.  Self-normalizing method to measure the detective quantum efficiency of a wide range of x-ray detectors. , 1999, Medical physics.

[25]  J H Siewerdsen,et al.  Cone-beam computed tomography with a flat-panel imager: effects of image lag. , 1999, Medical physics.

[26]  John Moore Boudry.  Operation of amorphous silicon detectors for chest radiography within system latency requirements , 1999, Medical Imaging.

[27]  Isaias D. Job,et al.  Characterization of a third-generation multimode sensor panel , 1999, Medical Imaging.

[28]  D. Jaffray,et al.  A ghost story: spatio-temporal response characteristics of an indirect-detection flat-panel imager. , 1999, Medical physics.

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

[30]  P. Granfors,et al.  Performance of a 41X41-cm2 amorphous silicon flat panel x-ray detector for radiographic imaging applications. , 2000, Medical physics.

[31]  J H Siewerdsen,et al.  Strategies to improve the signal and noise performance of active matrix, flat-panel imagers for diagnostic x-ray applications. , 2000, Medical physics.

[32]  J H Siewerdsen,et al.  Cone-beam computed tomography with a flat-panel imager: initial performance characterization. , 2000, Medical physics.

[33]  S Suryanarayanan,et al.  Full breast digital mammography with an amorphous silicon-based flat panel detector: physical characteristics of a clinical prototype. , 2000, Medical physics.

[34]  Harold L. Kundel,et al.  Handbook of Medical Imaging, Volume 1. Physics and Psychophysics , 2000 .

[35]  Hiroshi Aradate,et al.  Large area 2-dimensional detector for real-time 3-dimensional CT (4D-CT) , 2001 .

[36]  Ian A. Cunningham,et al.  Detective quantum efficiency of fluoroscopic systems: the case for a spatial-temporal approach (or, does the ideal observer have infinite patience?) , 2001, SPIE Medical Imaging.

[37]  I A Cunningham,et al.  Parallel cascades: new ways to describe noise transfer in medical imaging systems. , 2001, Medical physics.

[38]  Hiroshi Aradate,et al.  Large-area two-dimensional detector for real-time three-dimensional CT (4D CT) , 2001, SPIE Medical Imaging.

[39]  J A Rowlands,et al.  Effects of characteristic x rays on the noise power spectra and detective quantum efficiency of photoconductive x-ray detectors. , 2001, Medical physics.

[40]  D. Jaffray,et al.  Cone-beam computed tomography with a flat-panel imager: magnitude and effects of x-ray scatter. , 2001, Medical physics.