Kye Young Jeong

Reduction of artifacts due to multiple metallic objects in computed tomography

An X-ray computed tomography (CT) image including metallic objects suffers from annoying metal artifacts such as shades and streaks. In this paper, we propose a novel algorithm for reducing metal artifacts via a reprojectionreconstruction process. In the proposed algorithm, we first reconstruct a CT image from the original projection data. We then remove metallic object regions and replace them with the value of soft-tissue; apply total-variation-based smoothing to the image in order to reduce streak artifacts while preserving the shapes of non-metallic objects; and obtain the reprojection data of the smoothened image. Even though the reprojection data do not contain metallic objects, remaining shade artifacts (especially in the regions between metal objects) still affect the reprojection data. Those artifacts are found to be mainly concentrated in overlapping regions of metal-traces in the reprojection data. Hence, we horizontally interpolate the overlapping region with intensity values of its boundary pixels. We then replace whole metal trace regions in the original projection data with the processed reprojection data. The completed projection data are then used for the reconstruction of the final image. The proposed algorithm can reduce streak and shade artifacts while preserving the shape of non-metallic objects. It is proved that the proposed algorithm provides noticeably better performance in metal artifact reduction compared with the algorithms based on linear interpolation and the model image reprojection.

Metal artifact reduction based on sinogram correction in CT

A CT image including metallic objects suffers from strong metal artifacts such as shades and streaks. Most metal artifact reduction algorithms based on filtered back-projection have tried to make a metal-free projection data by modifying the original data on the metal-traces in the sinogram. Although those algorithms are able to reduce metal artifacts, they tend to produce new artifacts and distort non-metallic regions. In this paper, we propose a novel algorithm for reducing metal artifacts while preserving non-metallic region with minimum distortion. In the algorithm, we first generate a metal-free image by removing metallic regions in the initially reconstructed image and interpolating the intensities of metallic regions with those of non-metallic surrounding pixels. Based on the metal-free image, we obtain a reprojection data. We then apply a high-pass filter (HPF) for the reprojection data and interpolate overlapping regions along the metal-trace direction. We merge the processed reprojection data with the low-pass filtered projection data after replacing metal-trace regions with linearly interpolated values. The final reconstructed image is then obtained by using this merged projection data. The proposed algorithm provides much better performance compared with the existing algorithms by effectively reducing shade artifacts between metallic objects without introducing shape distortion of non-metallic objects.

Sinogram-based super-resolution in PET

Spatial resolution is intrinsically limited in positron emission tomography (PET) systems, mainly due to the crystal width. To increase the spatial resolution for a given crystal width, mechanical movements such as wobble and dichotomic motions are introduced to the PET systems. However, multiple sinograms obtained through such movements provide oversampled data. In this paper, to increase the spatial resolution, we present a novel super-resolution (SR) scheme that employs multiple sinograms. For SR, we first propose a blur kernel estimation scheme through a Monte Carlo simulation. Based on the estimated blur kernel, we adopt a maximum a posteriori expectation maximization method in estimating a high-resolution sinogram from multiple low-resolution sinograms. The proposed algorithm provides noticeable improvement of the spatial resolution in real PET images.

Sinogram Super-Resolution Using a Space-Variant Blur Matrix in PET

Positron emission tomography (PET) images suffer from low spatial resolution. To improve the spatial resolution, we previously proposed a sinogram-based super-resolution (SR) algorithm for a whole-body PET scanner, by assuming space invariant blur. However, since the spatial resolution of a sinogram varies along the radial direction due to parallax error, this algorithm is not appropriate for providing a high-resolution sinogram with reduction of parallax error. In this paper, we propose a novel and efficient sinogram-based SR algorithm that is suitable even for a small animal PET scanner by using space variant blur matrices. In the algorithm, we estimate the space variant blur matrices through a Monte Carlo simulation and use them for the SR process to obtain a high-resolution sinogram. Using a Derenzo phantom and a line source, we demonstrate in a real PET scanner, microPET R4, that the proposed SR algorithm noticeably improves the spatial resolution while alleviating its space variance. By applying the proposed SR algorithm, the full width at half-maximum (FWHM) value reaches 1.2 mm at the system center and 1.63 mm with a considerable parallax error reduction at a radial position of 4 cm.

GPU-based fast projection-backprojection algorithm for 3-D PET image reconstruction

Iterative image reconstruction algorithms based on a stochastic model of emission tomography have been widely studied, because they can provide better image quality than analytic reconstruction algorithms. However, their long reconstruction time has been a major bottle neck for further developments of high resolution PET scanners and their applications. In recent years, there have been several attempts to reduce the PET image reconstruction time by using a graphic processing unit (GPU). To obtain high computational performance on a massive parallel GPU, however, global memory coalescing and branching diversity are to be carefully considered, which are not considered in most existing GPU-based algorithms. To increase global memory coalescing, we propose the image-rotation-based (IR-based) projection and frame-rotation-based (FR-based) backprojection schemes. We then successfully incorporate the geometrical symmetry property into the proposed schemes to reduce the branching diversity. Thereby, we effectively reduce the total image reconstruction time from many hours to a few seconds. Experimental results show that the proposed algorithm reduces the computation time by a factor of about 539 compared with a CPU-based straightforward implementation.

LOR-Based Reconstruction for Super-Resolved 3D PET Image on GPU

Positron emission tomography (PET) images usually suffer from low spatial resolution mainly because of the finite dimension of crystals. To improve the spatial resolution based on wobble scanning, we previously proposed a sinogram-based super-resolution (SR) algorithm using a space-variant blur matrix. However, the algorithm may cause unwanted resolution loss owing to an inevitable interpolation process for preparing evenly spaced projections. In this article, we propose a novel one-step line of response (LOR)-based SR framework for 3D PET images. In the framework, we efficiently determine a large number of space-variant point spread functions (PSFs) in the image space by using the scanner symmetries and the proposed PSF interpolation scheme based on nonrigid registration. Furthermore, to minimize the resolution degradation in the evenly spaced parallel-beam rebinning and to reduce the computational time in the graphics processing unit (GPU) implementation, we introduce parallel-friendly LOR reconstruction based on cone-beam reordering. We then obtain a high resolution image via a one-step super-resolved 3D PET image reconstruction with the determined PSFs. The proposed framework provides noticeable improvement on the spatial resolution of PET images with a considerable reduction of computational time.

Image resolution improvement based on sinogram super-resolution in PET

One of the limits of PET imaging is the low spatial resolution due to a predetermined detector width. To overcome this limit, we may increase the number of samples by using the wobbling motion. Since the line spread function (LSF) of the sinogram is determined by the detector width, however, the increase of the number of samples is not sufficient to improve the sinogram resolution. In this paper, based on oversampled data obtained from the wobbling motion, we propose a novel and efficient super-resolution (SR) scheme for the sinogram. Since the proposed SR scheme adopts the penalized expectation maximization (EM) algorithm, it guarantees non-negative values of the super-resolved sinogram data. Through the experiments, we demonstrate that the proposed SR scheme can noticeably improve the spatial image resolution.

Phased attenuation correction and respiratory motion compensation of pet image by using a ct images and multiple respiratory-phase mr images

Respiratory motion blurs positron emission tomography (PET) images in thorax and abdominal regions, and can cause attenuation-related errors in quantitative parameters. In addition, this motion can cause localization error and disappearance of tumor(s) near the diaphragm. Respiratory gated PET imaging, or 4D PET imaging, is an attractive solution. For 4D PET, a surrogate measurement of the patients breathing is made during the scan, and based on this measurement the list-mode PET data are sorted and reconstructed into multiple images. In several studies, it has been reported that 4D PET imaging recovers tumor volume more accurately than 3D PET imaging, and yields more accurate standard uptake values. In spite of its promise, 4D PET images suffer from statistical noise because only a fraction of the acquired data can be used in each image. Moreover, they also suffer from inconsistent attenuation correction due to mismatches between PET and CT images. In this paper, we propose a novel framework for phased attenuation correction and respiratory motion compensation for a 3D PET image by using a 3D CT image and multiple respiratory-phase 3D MR images. For phased attenuation correction, we generate and utilize 4D CT image obtained via non-rigid registrations among a 3D CT image and multiple respiratory-phase 3D MR images. A 3D PET image at a selected respiratory phase is then obtained by adding attenuation corrected multiple 3D PET images through non-rigid transformation. Experimental result for a clinical dataset shows that the proposed algorithm can provide much clearer organ boundary with improved quality than the conventional method in a 3D PET image.

Non-rigid registration between 3D MR and CT images of the liver based on intensity and edge orientation information

The multi-modal image registration is essential to integrate complementary image information from different modalities of the same organ. In order to find a reliable geometric transformation for registration, a proper choice of similarity measure is required. Even though normalized mutual information (NMI) is popular for multi-modal image registration, it limits the registration performance since the NMI measure only takes intensity values into account without considering any useful spatial information. In this paper, we attempt to enhance the non-rigid registration accuracy between MR and CT images of the liver, by jointly using intensity and spatial information. In order to effectively manipulate the relationship of the intensity and spatial information between MR and CT images, we adopt our previously proposed similarity measure based on 3D joint histogram where edge orientation coincidence is utilized as the spatial information. Since we need non-rigid registration due to the easy deformation of liver, the B-spline free-form deformation (FFD) is adopted as the transformation model. The experimental results for clinical datasets show that the proposed method provides more accurate registration results than a conventional NMI-basedregistration.

PET image reconstruction based on several respiratory-phase low-dose CT images

Positron emission tomography (PET) can provide functional information in the human body. However, thoracic and abdominal PET images suffer from attenuation-related errors due to respiratory motion as well as blurs due to respiratory motion itself. In our previous study, to obtain respiratory-gated PET images, we proposed a novel method for respiratory phase matched attenuation correction (AC) and motion compensation (MC). That method utilizes a virtual 4D CT image generated via non-rigid registrations among a 3D CT image and a couple of respiratory-phase 3D MR images. In this paper, for respiratory phase matched attenuation correction and motion compensation, we propose to use a couple of respiratory-phase low-dose CT images rather than MR images. Experimental result for a clinical dataset shows that the proposed method can also provide much clearer organ boundaries as well as more accurate lesion information than the conventional method in a 3D PET image.