Multicontrast reconstruction using compressed sensing with low rank and spatially varying edge-preserving constraints for high-resolution MR characterization of myocardial infarction.

PURPOSE: To enable robust reconstruction for highly accelerated three-dimensional multicontrast late enhancement imaging to provide improved MR characterization of myocardial infarction with isotropic high spatial resolution. THEORY AND METHODS: A new method using compressed sensing with low rank and spatially varying edge-preserving constraints (CS-LASER) is proposed to improve the reconstruction of fine image details from highly undersampled data. CS-LASER leverages the low rank feature of the multicontrast volume series in MR relaxation and integrates spatially varying edge preservation into the explicit low rank constrained compressed sensing framework using weighted total variation. With an orthogonal temporal basis pre-estimated, a multiscale iterative reconstruction framework is proposed to enable the practice of CS-LASER with spatially varying weights of appropriate accuracy. RESULTS: In in vivo pig studies with both retrospective and prospective undersamplings, CS-LASER preserved fine image details better and presented tissue characteristics with a higher degree of consistency with histopathology, particularly in the peri-infarct region, than an alternative technique for different acceleration rates. An isotropic resolution of 1.5 mm was achieved in vivo within a single breath-hold using the proposed techniques. CONCLUSION: Accelerated three-dimensional multicontrast late enhancement with CS-LASER can achieve improved MR characterization of myocardial infarction with high spatial resolution. Magn Reson Med 78:598-610, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

COVID-19 in the U.S. during pre-vaccination period: Shifting impact of sociodemographic factors and air pollution.

Pandemic "wave" usually refers to the rise and fall of the infections with time, however, for a large country, the variations due to geographical location could be considerable. In this work, we investigated COVID-19 infection and fatality across the U.S. during the pandemic waves in the pre-vaccination period (January 2020-December 2020). Focusing on counties with a population ≥100,000, the data from the entire period were first segmented into two equal phases roughly corresponding to the first pandemic wave and subsequent surge, and each phase was further divided into two zones based on infection rate. We studied the potential influences of six sociodemographic variables (population density, age, poverty, education, and percentage of Hispanic and African American population) and four air pollutants (PM2.5, NO2, SO2, and O3) on the differences in infection and fatality observed among different phases and zones. We noticed a distinct difference in the overall impact of COVID-19 between the two phases of the pre-vaccination period with a substantial decrease in the fatality in the second phase despite an increase in the infection. Analysis using log-linear regression modeling further revealed a shift in the impact of several risk factors considered in this study. For example, population density and lesser education were found to be significant for infection during the first phase of the pandemic alone. Furthermore, population density and lesser education along with poverty and NO2 level had a significant contribution to fatality during the first phase of the pandemic, while age over 65 years was important in both phases. Interestingly, the effects of many of these factors were found to be significant only in the zones with higher infection rates. Our findings indicate that the impacts of several well-known sociodemographic and environmental risk factors for COVID-19 are not constant throughout the course of the pandemic, and therefore, careful considerations should be made about their role when developing preventative and mitigative measures.

Differential Impact of COVID-19 Risk Factors on Ethnicities in the United States.

The coronavirus disease (COVID-19) has revealed existing health inequalities in racial and ethnic minority groups in the US. This work investigates and quantifies the non-uniform effects of geographical location and other known risk factors on various ethnic groups during the COVID-19 pandemic at a national level. To quantify the geographical impact on various ethnic groups, we grouped all the states of the US. into four different regions (Northeast, Midwest, South, and West) and considered Non-Hispanic White (NHW), Non-Hispanic Black (NHB), Hispanic, Non-Hispanic Asian (NHA) as ethnic groups of our interest. Our analysis showed that infection and mortality among NHB and Hispanics are considerably higher than NHW. In particular, the COVID-19 infection rate in the Hispanic community was significantly higher than their population share, a phenomenon we observed across all regions in the US but is most prominent in the West. To gauge the differential impact of comorbidities on different ethnicities, we performed cross-sectional regression analyses of statewide data for COVID-19 infection and mortality for each ethnic group using advanced age, poverty, obesity, hypertension, cardiovascular disease, and diabetes as risk factors. After removing the risk factors causing multicollinearity, poverty emerged as one of the independent risk factors in explaining mortality rates in NHW, NHB, and Hispanic communities. Moreover, for NHW and NHB groups, we found that obesity encapsulated the effect of several other comorbidities such as advanced age, hypertension, and cardiovascular disease. At the same time, advanced age was the most robust predictor of mortality in the Hispanic group. Our study quantifies the unique impact of various risk factors on different ethnic groups, explaining the ethnicity-specific differences observed in the COVID-19 pandemic. The findings could provide insight into focused public health strategies and interventions.

A Regularization Approach to Blind Deblurring and Denoising of QR Barcodes.

QR bar codes are prototypical images for which part of the image is a priori known (required patterns). Open source bar code readers, such as ZBar, are readily available. We exploit both these facts to provide and assess purely regularization-based methods for blind deblurring of QR bar codes in the presence of noise.

Multiscale properties of weighted total variation flow with applications to denoising and registration.

Images consist of structures of varying scales: large scale structures such as flat regions, and small scale structures such as noise, textures, and rapidly oscillatory patterns. In the hierarchical (BV, L(2)) image decomposition, Tadmor, et al. (2004) start with extracting coarse scale structures from a given image, and successively extract finer structures from the residuals in each step of the iterative decomposition. We propose to begin instead by extracting the finest structures from the given image and then proceed to extract increasingly coarser structures. In most images, noise could be considered as a fine scale structure. Thus, starting the image decomposition with finer scales, rather than large scales, leads to fast denoising. We note that our approach turns out to be equivalent to the nonstationary regularization in Scherzer and Weickert (2000). The continuous limit of this procedure leads to a time-scaled version of total variation flow. Motivated by specific clinical applications, we introduce an image depending weight in the regularization functional, and study the corresponding weighted TV flow. We show that the edge-preserving property of the multiscale representation of an input image obtained with the weighted TV flow can be enhanced and localized by appropriate choice of the weight. We use this in developing an efficient and edge-preserving denoising algorithm with control on speed and localization properties. We examine analytical properties of the weighted TV flow that give precise information about the denoising speed and the rate of change of energy of the images. An additional contribution of the paper is to use the images obtained at different scales for robust multiscale registration. We show that the inherently multiscale nature of the weighted TV flow improved performance for registration of noisy cardiac MRI images, compared to other methods such as bilateral or Gaussian filtering. A clinical application of the multiscale registration algorithm is also demonstrated for aligning viability assessment magnetic resonance (MR) images from 8 patients with previous myocardial infarctions.

Feasibility Study of Respiratory Motion Modeling Based Correction for MRI-Guided Intracardiac Interventional Procedures.

GOAL: The purpose of this study is to improve the accuracy of interventional catheter guidance during intracardiac procedures. Specifically, the use of preprocedural magnetic resonance roadmap images for interventional guidance has limited anatomical accuracy due to intraprocedural respiratory motion of the heart. Therefore, we propose to build a novel respiratory motion model to compensate for this motion-induced error during magnetic resonance imaging (MRI)-guided procedures. METHODS: We acquire 2-D real-time free-breathing images to characterize the respiratory motion, and build a smooth motion model via registration of 3-D prior roadmap images to the real-time images within a novel principal axes frame of reference. The model is subsequently used to correct the interventional catheter positions with respect to the anatomy of the heart. RESULTS: We demonstrate that the proposed modeling framework can lead to smoother motion models, and potentially lead to more accurate motion estimates. Specifically, MRI-guided intracardiac ablations were performed in six preclinical animal experiments. Then, from retrospective analysis, the proposed motion modeling technique showed the potential to achieve a 27% improvement in ablation targeting accuracy. CONCLUSION: The feasibility of a respiratory motion model-based correction framework has been successfully demonstrated. SIGNIFICANCE: The improvement in ablation accuracy may lead to significant improvements in success rate and patient outcomes for MRI-guided intracardiac procedures.

Multiscale registration of real-time and prior MRI data for image-guided cardiac interventions.

Recently, there is a growing interest in using magnetic resonance imaging (MRI) to guide interventional procedures due to its excellent soft tissue contrast and lack of ionizing radiation compared to traditional radiographic guidance. One of these applications is the use of MRI to guide radio frequency ablation of anatomic substrates, within the left ventricle, responsible for ventricular tachycardia. However, different MRI acquisition schemes have significant tradeoffs between image quality and acquisition time. Guidance using high-quality preoperative 3-D MR images is limited in the case of cardiac interventions because the heart moves dynamically during the procedure. On the other hand, 2-D real-time MR images acquired during the intervention sacrifice image quality for shorter image acquisition time, leading to real-time positional updates of cardiac anatomy. Ideally, we wish to combine the advantages of live feedback from real-time images and accurate visualization of anatomical structures from preoperative images. Therefore, to improve the MRI guidance capabilities for cardiac interventions, we describe a novel multiscale rigid registration framework to correct for respiratory motion between the prior and real-time datasets. In the proposed approach, we use a weighted total variation flow algorithm to extract coarse-to-fine features from the input images and subsequently register the corresponding scales in a hierarchical manner. Registration experiments were performed with in vivo human imaging data, and the target registration error achieved was 1.51 mm. Thus, the feasibility of motion correction in an interventional setting has been demonstrated, which may lead to significant improvements in the guidance of cardiac interventions.