Truncation effect reduction for fast iterative reconstruction in cone-beam CT

Abstract

Background: Iterative reconstruction for cone-beam computed tomography (CBCT) has been applied to improve image quality and reduce radiation dose. In a case where an object is larger than a flat panel detector, most CBCT images contain truncated data or incomplete projections, which degrade image quality. In this work, we propose the truncation effect reduction for fast iterative reconstruction in CBCT imaging inside the field of view (FOV).Methods: The volume matrix size of FOV and the height of projection images were extrapolated to a suitable size. These extended projections were reconstructed by fast iterative reconstruction. Moreover, a smoothing parameter for noise regularization in iterative reconstruction was also modified to reduce the accumulated error while processing. The proposed work was evaluated by image quality measurements and compared with the conventional filtered back projection (FBP) method. To validate the proposed method, we used a head phantom for evaluation and tested on real human head data. Results: In the experimental results, the reconstructed images from the head phantom can be enhanced apparently. In addition, fast iteration reconstruction can be run continuously while remaining the consistent mean-percentage-error (MPE) value with a large number of iterations. The CNR of the soft-tissue images was improved by the increased contrast and the decreased noise. Visualization of low contrast in the ventricle and soft-tissue images can be observed much clearer compared to those from FBP using the same effective radiation dose of 5 mGy. Conclusions: Our proposed work has satisfactory performance to reduce the truncation effect, especially inside the FOV with better image quality for soft-tissue imaging. The convergence of fast iterative reconstruction tends to be stable for many iterations.