Temporal bone volumetric image deblurring in spiral computed tomography scanning.

RATIONALE AND OBJECTIVES We developed a method for volumetric image deblurring in spiral (helical) computed tomography (CT) scanning with a three-dimensional (3D) Gaussian point spread function (PSF) to improve the quality of temporal bone spiral CT images for assessing the position of cochlear implants electrodes. METHODS A patient was scanned after cochlear implantation, and the temporal bone was reconstructed into a volume with 128 voxels per dimension, 0.1 mm per voxel side, and x 10 gray-scale expansion. The 3D PSF in spiral CT imaging was assumed to be Gaussian separable transversely and longitudinally. Standard deviations of the PSF were derived and subjectively adjusted. The image was then deconvolved using Wiener filtering and maximum-likelihood deconvolution methods. Image quality was assessed both visually and quantitatively using cross-sectional area at half of the maximum (CAHM) of the implanted array as the figure of merit. RESULTS Substantial image deblurring was achieved via deconvolution. Subjectively, anatomic structures were more clearly shown. Deconvolution reduced the CAHM by approximately one third, on average. Three-dimensional deconvolution had better image quality than two-dimensional deconvolution. The maximum-likelihood method produced superior image quality but took longer to process relative to Wiener filtering. CONCLUSION Volumetric image deblurring is practical with a Gaussian PSF. The maximum-likelihood method is preferred if time permits. Deconvolution facilitates the study of fine details of the temporal bone and cochlear implant.

[1]  G Wang,et al.  Theoretical FWTM values in helical CT. , 1994, Medical physics.

[2]  I. Csiszár.  Why least squares and maximum entropy? An axiomatic approach to inference for linear inverse problems , 1991 .

[3]  Joseph A. O'Sullivan,et al.  Deblurring subject to nonnegativity constraints , 1992, IEEE Trans. Signal Process..

[4]  G Wang,et al.  Longitudinal image deblurring in spiral CT. , 1994, Radiology.

[5]  G Wang,et al.  Spatial variation of section sensitivity profile in spiral computed tomography. , 1994, Medical physics.

[6]  G Wang,et al.  Longitudinal resolution in volumetric x-ray computerized tomography--analytical comparison between conventional and helical computerized tomography. , 1994, Medical physics.

[7]  K F King,et al.  Computed tomography scanning with simultaneous patient translation. , 1990, Medical physics.

[8]  William H. Richardson,et al.  Bayesian-Based Iterative Method of Image Restoration , 1972 .

[9]  W. Kalender,et al.  Spiral volumetric CT with single-breath-hold technique, continuous transport, and continuous scanner rotation. , 1990, Radiology.

[10]  L. Shepp,et al.  Maximum Likelihood Reconstruction for Emission Tomography , 1983, IEEE Transactions on Medical Imaging.

[11]  T. Holmes.  Expectation-maximization restoration of band-limited, truncated point-process intensities with application in microscopy , 1989 .

[12]  Timothy J. Holmes.  Maximum-likelihood image restoration adapted for noncoherent optical imaging , 1988 .

[13]  W A Kalender,et al.  A Comparison of Conventional and Spiral CT: An Experimental Study on the Detection of Spherical Lesions , 1994, Journal of computer assisted tomography.

[14]  W. Kalender,et al.  Evaluation of section sensitivity profiles and image noise in spiral CT. , 1992, Radiology.

[15]  L. Lucy.  An iterative technique for the rectification of observed distributions , 1974 .

[16]  G Wang,et al.  Helical CT image noise--analytical results. , 1993, Medical physics.

[17]  W. Kalender,et al.  Determination of the position of nucleus cochlear implant electrodes in the inner ear. , 1994, The American journal of otology.