Abstract:Purpose: Four-dimensional cone-beam computed tomography (4D CBCT) has been developed to provide a sequence of phase-resolved reconstructions in image-guided radiation therapy. However, 4D CBCT images are degraded by severe streaking artifacts because the 4D CBCT reconstruction process is an extreme sparse-view CT procedure wherein only under-sampled projections are used for the reconstruction of each phase. To obtain a set of 4D CBCT images achieving both high spatial and temporal resolution, we propose an algorithm by providing a high-quality initial image at the beginning of the iterative reconstruction process for each phase to guide the final reconstructed result toward its optimal solution. Methods: The proposed method consists of three steps to generate the initial image. First, a prior image is obtained by an iterative reconstruction method using the measured projections of the entire set of 4D CBCT images. The prior image clearly shows the appearance of structures in static regions, although it contains blurring artifacts in motion regions. Second, the robust principal component analysis (RPCA) model is adopted to extract the motion components corresponding to each phase-resolved image. Third, a set of initial images are produced by the proposed linear estimation model that combines the prior image and the RPCA-decomposed motion components. The final 4D CBCT images are derived from the simultaneous algebraic reconstruction technique (SART) equipped with the initial images. Qualitative and quantitative evaluations were performed by using two extended cardiac-torso (XCAT) phantoms and two sets of patient data. Several state-of-the-art 4D CBCT algorithms were performed for comparison to validate the performance of the proposed method. Results: The image quality of phase-resolved images is greatly improved by the proposed method in both phantom and patient studies. The results show an outstanding spatial resolution, in which streaking artifacts are suppressed to a large extent, while detailed structures such as tumors and blood vessels are well restored. Meanwhile, the proposed method depicts a high temporal resolution with a distinct respiratory motion change at different phases. For simulation phantom, quantitative evaluations of the simulation data indicate that an average of 36.72% decrease at EI phase and 42% decrease at EE phase in terms of root-mean-square error (RMSE) are achieved by our method when comparing with PICCS algorithm in Phantom 1 and Phantom 2. In addition, the proposed method has the lowest entropy and the highest normalized mutual information compared with the existing methods in simulation experiments, such as PRI, RPCA-4DCT, SMART, and PICCS. And for real patient cases, the proposed method also achieves the lowest entropy value compared with the competitive method. Conclusions: The proposed algorithm can generate an optimal initial image to improve iterative reconstruction performance. The final sequence of phase-resolved volumes guided by the initial image achieves high spatiotemporal resolution by eliminating motion-induced artifacts. This study presents a practical 4D CBCT reconstruction method with leading image quality. (C) 2020 American Association of Physicists in Medicine

High resolution reconstruction for optimization of imaging quality of GE Discovery CT 750 HD

A Phantom Study Comparing Low-Dose CT Physical Image Quality from Five Different CT Scanners

Ct Image Resolution Improvement By Geometric Modelling In Iterative Reconstruction

Comparison of image quality based on deep-learning image reconstruction and iterative reconstruction algorithm for dual-energy CT: a phantom and animal-model study

Application of High-Resolution CT Local Augmentation Reconstruction to Improve the Accurate Diagnosis of Small Pulmonary Nodule Structure

High-Resolution Head CTA: A Prospective Patient Study Comparing Image Quality of Photon-Counting Detector CT and Energy-Integrating Detector CT

A Preliminary Investigation of Image Reconstruction with Variable Resolutions in Diagnostic CT

Super-resolution deep-learning reconstruction for cardiac CT: impact of radiation dose and focal spot size on task-based image quality

Influent factors in thorax image quality on sensation 64-slice spiral CT:a phantom study

Image Quality of Coronary CT Angiography at Ultra Low Tube Voltage Reconstructed with a Deep-Learning Image Reconstruction Algorithm in Patients of Different Weight

Improvements in dose efficiency with high resolution scan modes in photon counting CT

A Preliminary Study Of Image Reconstruction From Low-Dose Data In Dedicated Breast Ct

Noise and spatial resolution properties of a commercially available deep-learning based CT reconstruction algorithm.

Fully automated image quality evaluation on patient CT: Multi-vendor and multi-reconstruction study

Assessment of Noise Reduction Potential and Image Quality Improvement of a New Generation Adaptive Statistical Iterative Reconstruction (ASIR-V) in Chest CT

High-quality Initial Image-Guided 4D CBCT Reconstruction

Evaluation of dose reduction and image quality in abdominal CT using adaptive statistical iterative reconstruction in a phantom study

Super Resolution in CT

Low dose 4D-CT super-resolution reconstruction via inter-plane motion estimation based on optical flow

Optimization of Detector Thickness for Single Slice Helical CT with ROC Study

Optimization-based Design for Artifact Reduction in Advanced Diagnostic CT