MO-DE-207A-05: Dictionary Learning Based Reconstruction with Low-Rank Constraint for Low-Dose Spectral CT

2016 ◽  
Vol 43 (6Part30) ◽  
pp. 3701-3701 ◽  
Author(s):  
Q Xu ◽  
H Liu ◽  
H Yu ◽  
G Wang ◽  
L Xing
Keyword(s):  
Dictionary Learning ◽  
Low Dose ◽  
Low Rank ◽  
Spectral Ct ◽  
Rank Constraint

Discriminative-shared dictionary learning for class-specific fabric texture characterization

2020 ◽  
Vol 90 (21-22) ◽  
pp. 2478-2491 ◽  
Author(s):  
Zhu Zhan ◽  
Liqing Li ◽  
Xia Chen ◽  
Jun Wang
Keyword(s):  
Dictionary Learning ◽  
Low Rank ◽  
Specific Class ◽  
Similarity Matrix ◽  
Intra Class Correlation ◽  
Rank Constraint ◽  
Learned Dictionary ◽  
Shared Component ◽  
Texture Characterization ◽  
Discriminative Dictionary Learning

In this paper we introduced a novel discriminative-shared dictionary learning (DSDL) model to explicitly extract a set of class-specific features as well as shared features for simultaneous fabric texture characterization. For the discriminative component, we imposed the constraints of minimizing inter-class correlation as well as maximizing intra-class correlation on them. For the shared component, we enforced a low-rank constraint and a similarity constraint on it. To demonstrate the characterization performance of the learned dictionary on multi-class fabric textures, we weaved eight fabric textures in the laboratory and then reconstructed each class of sample with each class of specific dictionary. Preliminary experiments with four-class samples demonstrate that the specific class of dictionary can only reconstruct the corresponding class of fabric texture, but has weak reconstructive ability for the rest class, and vice versa. In addition, we illustrated the contribution of shared dictionary by comparing the convergence rate of DSDL with that of Fisher Discriminative Dictionary Learning. Further, we developed a new indicator based on a similarity matrix for evaluating the discriminative of class-specific dictionaries, and we validated the effectiveness of the indicator by comparing the discriminative indicators of eight group sets. In general, the proposed DSDL model can effectively extract discriminative features and shared features simultaneously of multi-class fabric textures.

Virtual samples based robust block-diagonal dictionary learning for face recognition

2021 ◽  
Vol 25 (5) ◽  
pp. 1273-1290
Author(s):  
Shuangxi Wang ◽  
Hongwei Ge ◽  
Jinlong Yang ◽  
Shuzhi Su
Keyword(s):  
Face Recognition ◽  
Dictionary Learning ◽  
Recognition Performance ◽  
Small Sample Size ◽  
Small Sample ◽  
Low Rank ◽  
Negative Effects ◽  
Linear Classifier ◽  
Virtual Samples ◽  
Block Diagonal

It is an open question to learn an over-complete dictionary from a limited number of face samples, and the inherent attributes of the samples are underutilized. Besides, the recognition performance may be adversely affected by the noise (and outliers), and the strict binary label based linear classifier is not appropriate for face recognition. To solve above problems, we propose a virtual samples based robust block-diagonal dictionary learning for face recognition. In the proposed model, the original samples and virtual samples are combined to solve the small sample size problem, and both the structure constraint and the low rank constraint are exploited to preserve the intrinsic attributes of the samples. In addition, the fidelity term can effectively reduce negative effects of noise (and outliers), and the ε-dragging is utilized to promote the performance of the linear classifier. Finally, extensive experiments are conducted in comparison with many state-of-the-art methods on benchmark face datasets, and experimental results demonstrate the efficacy of the proposed method.

scholarly journals Medical Image Denoising Algorithm Based on Sparse Nonlocal Regularized Weighted Coding and Low Rank Constraint

2021 ◽  
Vol 2021 ◽  
pp. 1-6
Author(s):  
Quan Yuan ◽  
Zhenyun Peng ◽  
Zhencheng Chen ◽  
Yanke Guo ◽  
Bin Yang ◽  
...  
Keyword(s):  
Image Denoising ◽  
Gaussian Noise ◽  
Medical Image ◽  
Impulse Noise ◽  
Medical Images ◽  
Noise Removal ◽  
White Gaussian Noise ◽  
Low Rank ◽  
Rank Constraint ◽  
Nonlocal Regularization

Medical image information may be polluted by noise in the process of generation and transmission, which will seriously hinder the follow-up image processing and medical diagnosis. In medical images, there is a typical mixed noise composed of additive white Gaussian noise (AWGN) and impulse noise. In the conventional denoising methods, impulse noise is first removed, followed by the elimination of white Gaussian noise (WGN). However, it is difficult to separate the two kinds of noises completely in practical application. The existing denoising algorithm of weight coding based on sparse nonlocal regularization, which can simultaneously remove AWGN and impulse noise, is plagued by the problems of incomplete noise removal and serious loss of details. The denoising algorithm based on sparse representation and low rank constraint can preserve image details better. Thus, a medical image denoising algorithm based on sparse nonlocal regularization weighted coding and low rank constraint is proposed. The denoising effect of the proposed method and the original algorithm on computed tomography (CT) image and magnetic resonance (MR) image are compared. It is revealed that, under different σ and ρ values, the PSNR and FSIM values of CT and MRI images are evidently superior to those of traditional algorithms, suggesting that the algorithm proposed in this work has better denoising effects on medical images than traditional denoising algorithms.

Abstract P105: Identification of Cardioembolic Sources and Myocardial Injury in Acute Ischemic Stroke With Delayed-Phase, Low-Dose Chest Ct by Means of a Dual-Layer Spectral Ct Scanner

Stroke ◽  
2021 ◽  
Vol 52 (Suppl_1) ◽  
Author(s):  
Gaston Rodriguez Granillo ◽  
Juan José Cirio ◽  
Ivan Lylyk ◽  
Nicolas Perez ◽  
Maria L Caballero ◽  
...  
Keyword(s):  
Myocardial Injury ◽  
Low Dose ◽  
Late Enhancement ◽  
Chest Ct ◽  
Ct Scans ◽  
Ct Scanner ◽  
Spectral Ct ◽  
Thrombotic Complications ◽  
The Mean ◽  
Delayed Phase

Background: The COVID-19 pandemic has promoted adaptations in diagnostic algorithms. We explored the feasibility and accuracy of delayed phase (DP) chest computed tomography (CT) performed immediately after brain CT perfusion (CTP) for the identification of thrombotic complications and myocardial fibrosis among patients admitted with acute ischemic stroke (AIS). Methods: Since July, we have incorporated the use of low dose chest CT scans using a spectral CT scanner in all patients admitted with AIS, encouraging acquisitions, five min after brain CTP. All scans were non gated and comprised low dose chest CT scans, without additional contrast. Using virtual monochromatic imaging and iodine maps, we evaluated the presence of thrombotic complications, myocardial late enhancement, and myocardial extracellular volume (ECV), as a surrogate of edema and interstitial fibrosis. Results: We included 22 patients. The mean age was 66.2±19.6 years. In 5 patients, a cardioembolic (CE) source was later identified by transesophageal echocardiogram (TEE), [left atrial appendage (LAA) thrombus, n=1], transthoracic echocardiogram with agitated saline injection (patent foramen ovale n=2), or by EKG (atrial fibrillation). Seven patients further underwent either TEE or cardiac CT to identify CE sources. DP non gated chest CT had a sensitivity and specificity of 100% to identify CE sources, 1 LAA thrombus correctly detected. Chest CT identified pulmonary thromboembolism (PE), later confirmed with CT angiography. Chest CT identified myocardial late enhancement in 16 patients (80% in CE vs. 71% in non CE, p=0.68), myocardial fat in 1, and coronary calcification in 77% [with 2.6±2.2 vs 3.8±3.6 coronary calcified segments in CE vs. non CE strokes, p=0.36). The mean ECV was 35±4% in CE vs 32±6% in non CE strokes (p=0.17). The 2 patients with a positive PCR test for COVID-19 showed evidence of myocardial late iodine enhancement, and incremented ECV of the septal wall (38% and 40%, respectively). Conclusions: In this pilot study, DP, non ECG gated, low dose chest CT scan performed 5 min after brain CTP with a spectral scanner; enabled straightforward identification of CE sources among patients with AIS. This approach allowed detection of PE and myocardial injury.

Sign in / Sign up

Export Citation Format

Share Document