Performance Of Iterative Reconstruction Research Articles

Performance of Iterative Reconstruction in Image Space Algorithm in Combination with Automatic Tube Current Modulation Compared to Filtered Back Projection in Brain CT Scan

Performance of Iterative Reconstruction in Image Space Algorithm in Combination with Automatic Tube Current Modulation Compared to Filtered Back Projection in Brain CT Scan

Effect of Unmatched System Models on Iterative Reconstruction in Computed Tomography: A Phantom Study

System models used for iterative reconstruction affect the quality of computed tomography (CT) images. The imaging performance of the system models depends on the projectors, which implements forward- and back-projection. The system model based on a single projector is limited to improve image quality because each projector has both benefits and drawbacks. In this study, the effect of unmatched system models on iterative reconstruction was investigated by evaluating the image quality of reconstructed images. The unmatched system models composed of different projectors were used to perform iterative reconstruction. The image quality was evaluated in order to investigate the effect of the unmatched system models, and the imaging performance of the unmatched system models was compared with that of the matched system models by simulations and experiments with digital and anthropomorphic head phantoms. The results showed that the characteristics of CT images depended on the combinations of forward- and back-projectors, and the unmatched system models are able to improve the quality of the images obtained by using iterative reconstruction. The unmatched system models can be potentially used to improve the imaging performance of iterative reconstruction in CT systems.

Penalized PET Reconstruction Using Deep Learning Prior and Local Linear Fitting

Motivated by the great potential of deep learning in medical imaging, we propose an iterative positron emission tomography reconstruction framework using a deep learning-based prior. We utilized the denoising convolutional neural network (DnCNN) method and trained the network using full-dose images as the ground truth and low dose images reconstructed from downsampled data by Poisson thinning as input. Since most published deep networks are trained at a predetermined noise level, the noise level disparity of training and testing data is a major problem for their applicability as a generalized prior. In particular, the noise level significantly changes in each iteration, which can potentially degrade the overall performance of iterative reconstruction. Due to insufficient existing studies, we conducted simulations and evaluated the degradation of performance at various noise conditions. Our findings indicated that DnCNN produces additional bias induced by the disparity of noise levels. To address this issue, we propose a local linear fitting function incorporated with the DnCNN prior to improve the image quality by preventing unwanted bias. We demonstrate that the resultant method is robust against noise level disparities despite the network being trained at a predetermined noise level. By means of bias and standard deviation studies via both simulations and clinical experiments, we show that the proposed method outperforms conventional methods based on total variation and non-local means penalties. We thereby confirm that the proposed method improves the reconstruction result both quantitatively and qualitatively.

Performance of Iterative Reconstruction and Automated Tube Voltage Selection on the Image Quality and Radiation Dose in Abdominal CT Scans

To study the impact of sinogram-affirmed iterative reconstruction (SAFIRE) and concurrent application of automated tube voltage selection (ATVS) on image quality (IQ) and radiation dose. A phantom was scanned using various computed tomography (CT) parameters (kV, 80-120; mAs, 50-200). Abdomen contrast-enhanced CT (CECTs) in 170 adults were performed using dose-modified protocols: in 145 patients (group I), ATVS was applied (mAs, 111-649); in 25 (group II), the kV was fixed at 120 (reference mAs, 150). In 95 patients, standard-dose (SD) scan was available. Two readers evaluated the IQ of filtered back projection (FBP) and SAFIRE (levels 1, 3, and 5) images. In phantom, nonlinear drop in noise with increasing strengths of IR (levels S1-S5) was noted. The dose-modified IR scan was rated diagnostic in all 170 patients, with IQ score comparable to that of SD-FBP (P = 0.3). Lower kV (100/80) was prescribed by ATVS in 70% examinations in group I. In comparison with SD-FBP, the mean dose in CT dose index in group I (IR, 3.2 mGy; SD-FBP, 13.02 mGy; P < 0.0001) and in group II (IR, 4.8 mGy; SD-FBP, 11.8 mGy; P < 0.001) was 75.4% and 59.3% lower. Use of SAFIRE and ATVS provides diagnostic quality images at 59.3% to 75.4% reduced dose compared with SD-FBP scan.

GPU based iterative cone-beam CT reconstruction using empty space skipping technique

Iterative reconstruction of high-resolution cone-beam CT data is still a difficult task due to the demand for vast amounts of computer cycles and associated memory. In order to improve the performance of iterative algorithms for cone-beam CT reconstruction, an acceleration approach integrating GPU acceleration, empty space skipping and multi-resolution technique is proposed. The approach divides the reconstructed volume into equally sized blocks, and empty blocks are identified by reconstructing an initial low-resolution volume and segmenting it with threshold method. Then all non-empty blocks are packed into a new volume, which is initialized by interpolating the low resolution volume and reconstructed at full resolution using iterative algorithms. Finally these non-empty blocks are rearranged to get the reconstructed high-resolution volume. The whole process is implemented in parallel based on GPU. Since only the voxels in non-empty blocks are calculated, the number of considered voxels is greatly reduced, which translates directly into substantial computation, memory requirements and data transfer savings. The method is evaluated by reconstructing images from simulated projection data of phantom and CT datasets. The results indicate that our approach significantly improves the performance of iterative reconstruction while maintaining a high image quality, compared to conventional GPU-based approaches.