Peter Forthmann

Penalized maximum-likelihood sinogram restoration for dual focal spot computed tomography

Due to various system non-idealities, the raw data generated by a computed tomography (CT) machine are not readily usable for reconstruction. Although the deterministic nature of corruption effects such as crosstalk and afterglow permits correction by deconvolution, there is a drawback because deconvolution usually amplifies noise. Methods that perform raw data correction combined with noise suppression are commonly termed sinogram restoration methods. The need for sinogram restoration arises, for example, when photon counts are low and non-statistical reconstruction algorithms such as filtered backprojection are used. Many modern CT machines offer a dual focal spot (DFS) mode, which serves the goal of increased radial sampling by alternating the focal spot between two positions on the anode plate during the scan. Although the focal spot mode does not play a role with respect to how the data are affected by the above-mentioned corruption effects, it needs to be taken into account if regularized sinogram restoration is to be applied to the data. This work points out the subtle difference in processing that sinogram restoration for DFS requires, how it is correctly employed within the penalized maximum-likelihood sinogram restoration algorithm and what impact it has on image quality.

Adaptive two-pass cone-beam artifact correction using a FOV-preserving two-source geometry: A simulation study

The evolution to ever wider detector arrays that are able to cover whole organs with a single circular gantry sweep has revitalized the research efforts toward finding improved axial scanning algorithms and protocols. The authors propose a computed tomography scan and reconstruction concept using two sources, a single detector and a two-pass cone-beam correction method, as an integral part of the reconstruction. Compared with standard circular acquisition and reconstruction methods, the new concept excels with improved coverage and very low cone-beam artifact level also for short scan acquisitions, which makes it especially attractive for cardiac applications.

Vector field interpolation for cardiac motion compensated reconstruction

Cardiac CT image reconstruction suffers from motion artifacts due to heart motion during acquisition. In order to mitigate these effects, it is common practice to acquire with fast gantry rotation and do gated reconstruction. In addition, it is possible to estimate heart motion retrospectively and to incorporate that information in a motion compensated reconstruction (MCR). However, if shape tracking algorithms are used for generation of a heart motion vector field (MVF), the number and positions of the resulting motion vectors will not coincide with the number and positions of the voxels in the reconstruction grid. For MCR algorithms that require one motion vector at each voxel location, the MVF must be interpolated. This work examines different interpolation approaches for the MVF interpolation problem and their effects on the MCR results.

Comparison of three sinogram restoration methods

The raw data acquired during a computed tomography (CT) scan carry the unwanted traces of a number of adverse effects connected with the measurement setup and the acquisition process. To name a few, these include systematic errors like detector crosstalk and afterglow, fluctuations in tube power during the scan, but also statistical effects like photon noise. Most systematic effects can be cast into a linear model, providing a way for neutralizing the influence of these errors through deconvolution. However, this deconvolution process inevitably increases the image noise content. For low-dose scans, application of some kind of noise suppression algorithm is mandatory, in order to keep its disturbing influence on the reconstructed images in check. Since resolution and noise are antagonizing properties, noise suppression usually has the side effect of decreasing resolution. The interest in finding an algorithm that deals with this quandary in an optimal way is obvious. This work compares three deconvolution/denoising methods, identifying the one that performs best on a set of simulated data. The tested methods of combined sinogram deconvolution/denoising are based on (1) regularized matrix inversion, (2) straight matrix inversion plus adaptive filtering, and (3) deconvolution by a penalized maximum likelihood approach. In-plane and axial noise/resolution measurements identified the penalized maximum-likelihood method as best suited for low-dose applications. The adaptive filter approach performed well, but did not retain as much resolution when going to higher smoothing levels. The analytic deconvolution, however, could not compete against the other two methods.

Correction of breathing motion in the thorax for helical CT

Abstract This paper investigates a reconstruction method for helical computed tomography which compensates for the motion artifacts in the thorax caused by patient breathing. The method takes into account a motion vector field determined from a four-dimensional (4-D) uncompensated image data set. Surface models of the lung and the ribs are tracked through the 4-D data set to create motion information within the entire thorax. Finally, an image is reconstructed using motion compensated back-projection. The results show that due to the use of shape models for the motion estimation, the method is fast and robust. Furthermore, since the surfaces are tracked individually, reconciling the opposite motion direction of the lung and rib cage is avoided in one motion vector field.

Advanced processing for motion-compensated reconstruction in cardiac cone-beam CT

Cardiac CT image reconstruction suffers from artifacts due to heart motion during acquisition. In order to mitigate these effects, it is common practice to choose a protocol with minimal gating window and fast gantry rotation. In addition, it is possible to estimate heart motion retrospectively and to incorporate the information in a motion-compensated reconstruction (MCR). If shape tracking algorithms are used for generation of the heart motion-vector field (MVF), the number and positions of the motion vectors will not coincide with the number and positions of the voxels in the reconstruction grid. In this case, data interpolation is necessary for MCR algorithms which require one motion vector at each voxel location. This work examines different data interpolation approaches for the MVF interpolation problem and the effects on the MCR results.

Effect of detector time delay and its correction in multiple source computed tomography

In scintillating detectors x-rays are converted to luminescent photons with a time delay. The corresponding time resolution of the detector can have - in contrast to usual multi-slice CT - a deteriorating effect in new CT concepts with multiple sources illuminating one detector, because x-ray intensities measured here in consecutive projections correspond to the absorption along paths through very different regions of the object. A new analytical description of these effects is presented and a correction algorithm is derived. It is also shown that the detector time delay and its correction can lead to a noticeable increase of image noise.

Sinogram restoration of dual focal spot CT data

The raw data generated by a computed tomography (CT) machine are not readily usable for reconstruction. This is the result of various system non-idealities, and although the deterministic nature of corruption effects like crosstalk and afterglow permits removal through deconvolution, there is the drawback that deconvolution increases noise. Methods that perform raw data correction combined with noise suppression are commonly termed sinogram restoration methods. The need for sinogram restoration arises, for example, when photon counts are low and non-statistical reconstruction algorithms like filtered backprojection are used. Many modern CT machines offer a so-called dual focal spot (DFS) mode, which serves the goal of increased radial sampling by switching the focal spot between two positions on the anode plate during the scan. Although the focal spot mode does not play a role with respect to how the data are affected by the above mentioned corruption effects, it needs to be taken into account, if regularized sinogram restoration is to be applied to the data. This work points out the subtle difference in processing that sinogram restoration for DFS requires, how it is correctly employed within the penalized maximum likelihood sinogram restoration algorithm, and what impact that has on image quality.