Improving abdomen tumor low-dose CT images using a fast dictionary learning based processing

Abstract

In abdomen computed tomography (CT), repeated radiation exposures are often inevitable for cancer patients who receive surgery or radiotherapy guided by CT images. Low-dose scans should thus be considered in order to avoid the harm of accumulative x-ray radiation. This work is aimed at improving abdomen tumor CT images from low-dose scans by using a fast dictionary learning (DL) based processing. Stemming from sparse representation theory, the proposed patch-based DL approach allows effective suppression of both mottled noise and streak artifacts. The experiments carried out on clinical data show that the proposed method brings encouraging improvements in abdomen low-dose CT images with tumors.

Low-dose CT image processing using artifact suppressed dictionary learning

“…The discriminative dictionaries in Fig.2 are used. In the second stage, the general DL algorithm in [10] is applied with a 256 dictionary size and patches 8 8. Sparsity level is set to 5, and 20 iterations (Itern=20) are used in the K-SVD calculation to update the dictionary D and the sparse coefficient  .…”

“…Three novel discriminative dictionaries are introduced in this paper to give significant improvement in LDCT imaging [10]. With 1/4 routine tube current, the proposed ASDL approach leads to CT images with comparable quality to SDCT images.…”

“…Each T is the preset parameter of sparsity level that limits the maximum nonzero entry number in  ij . Based on [9][10], the solution to Eq. (1) includes the following two steps (2) and (3):…”

Patch-based models and algorithms for image processing: a review of the basic principles and methods, and their application in computed tomography

“…The basic denoising algorithm described in "Image processing using learned overcomplete dictionaries" section was used for denoising of abdomen [18], head [7,16], and micro-CT images [42] with promising results. This simple algorithm resulted in effective suppression of noise and artifacts and a marked improvement in the visual and objective image quality.…”

Artifact Suppressed Dictionary Learning for Low-Dose CT Image Processing

“…Some successful applications in medical imaging have been explored for DL approaches. They concerned undersampled MRI image reconstruction [32], resolution enhancement [33], interior tomography [34], DL constrained iterative LDCT reconstruction [35], 3-D medical image denoising [36], few-views tomography [9-11, 15-17, 37], spectral CT [38] and abdomen LDCT image processing [39]. It has been widely accepted that the TV based reconstruction can also be considered a typical tomographic application of compressed sensing theory [9,11,15,37].…”