Aashish Goela

Left ventricle segmentation in MRI via convex relaxed distribution matching.

A fundamental step in the diagnosis of cardiovascular diseases, automatic left ventricle (LV) segmentation in cardiac magnetic resonance images (MRIs) is still acknowledged to be a difficult problem. Most of the existing algorithms require either extensive training or intensive user inputs. This study investigates fast detection of the left ventricle (LV) endo- and epicardium surfaces in cardiac MRI via convex relaxation and distribution matching. The algorithm requires a single subject for training and a very simple user input, which amounts to a single point (mouse click) per target region (cavity or myocardium). It seeks cavity and myocardium regions within each 3D phase by optimizing two functionals, each containing two distribution-matching constraints: (1) a distance-based shape prior and (2) an intensity prior. Based on a global measure of similarity between distributions, the shape prior is intrinsically invariant with respect to translation and rotation. We further introduce a scale variable from which we derive a fixed-point equation (FPE), thereby achieving scale-invariance with only few fast computations. The proposed algorithm relaxes the need for costly pose estimation (or registration) procedures and large training sets, and can tolerate shape deformations, unlike template (or atlas) based priors. Our formulation leads to a challenging problem, which is not directly amenable to convex-optimization techniques. For each functional, we split the problem into a sequence of sub-problems, each of which can be solved exactly and globally via a convex relaxation and the augmented Lagrangian method. Unlike related graph-cut approaches, the proposed convex-relaxation solution can be parallelized to reduce substantially the computational time for 3D domains (or higher), extends directly to high dimensions, and does not have the grid-bias problem. Our parallelized implementation on a graphics processing unit (GPU) demonstrates that the proposed algorithm requires about 3.87 s for a typical cardiac MRI volume, a speed-up of about five times compared to a standard implementation. We report a performance evaluation over 400 volumes acquired from 20 subjects, which shows that the obtained 3D surfaces correlate with independent manual delineations. We further demonstrate experimentally that (1) the performance of the algorithm is not significantly affected by the choice of the training subject and (2) the shape description we use does not change significantly from one subject to another. These results support the fact that a single subject is sufficient for training the proposed algorithm.

Influence of pacing site characteristics on response to cardiac resynchronization therapy.

Background—Transmural scar occupying left ventricular (LV) pacing regions has been associated with reduced response to cardiac resynchronization therapy (CRT). However, spatial influences of lead tip delivery relative to scar at both pacing sites remain poorly explored. This study evaluated scar distribution relative to LV and right ventricular (RV) lead tip placement through coregistration of late gadolinium enhancement MRI and cardiac computed tomographic (CT) findings. Influences on CRT response were assessed by serial echocardiography. Methods and Results—Sixty patients receiving CRT underwent preimplant late gadolinium enhancement MRI, postimplant cardiac CT, and serial echocardiography. Blinded segmental evaluations of mechanical delay, percentage scar burden, and lead tip location were performed. Response to CRT was defined as a reduction in LV end-systolic volume ≥15% at 6 months. The mean age and LV ejection fraction were 64±9 years and 25±7%, respectively. Mean scar volume was higher among CRT nonresponders for both the LV (23±23% versus 8±14% [P=0.01]) and RV pacing regions (40±32% versus 24±30% [P=0.04]). Significant pacing region scar was identified in 13% of LV pacing regions and 37% of RV pacing regions. Absence of scar in both regions was associated with an 81% response rate compared with 55%, 25%, and 0%, respectively, when the RV, LV, or both pacing regions contained scar. LV pacing region dyssynchrony was not predictive of response. Conclusions—Myocardial scar occupying the LV pacing region is associated with nonresponse to CRT. Scar occupying the RV pacing region is encountered at higher frequency and seems to provide a more intermediate influence on CRT response.

Regional Assessment of Cardiac Left Ventricular Myocardial Function via MRI Statistical Features

Automating the detection and localization of segmental (regional) left ventricle (LV) abnormalities in magnetic resonance imaging (MRI) has recently sparked an impressive research effort, with promising performances and a breadth of techniques. However, despite such an effort, the problem is still acknowledged to be challenging, with much room for improvements in regard to accuracy. Furthermore, most of the existing techniques are labor intensive, requiring delineations of the endo- and/or epi-cardial boundaries in all frames of a cardiac sequence. The purpose of this study is to investigate a real-time machine-learning approach which uses some image features that can be easily computed, but that nevertheless correlate well with the segmental cardiac function. Starting from a minimum user input in only one frame in a subject dataset, we build for all the regional segments and all subsequent frames a set of statistical MRI features based on a measure of similarity between distributions. We demonstrate that, over a cardiac cycle, the statistical features are related to the proportion of blood within each segment. Therefore, they can characterize segmental contraction without the need for delineating the LV boundaries in all the frames. We first seek the optimal direction along which the proposed image features are most descriptive via a linear discriminant analysis. Then, using the results as inputs to a linear support vector machine classifier, we obtain an abnormality assessment of each of the standard cardiac segments in real-time. We report a comprehensive experimental evaluation of the proposed algorithm over 928 cardiac segments obtained from 58 subjects. Compared against ground-truth evaluations by experienced radiologists, the proposed algorithm performed competitively, with an overall classification accuracy of 86.09% and a kappa measure of 0.73.

Regional heart motion abnormality detection: An information theoretic approach

Tracking regional heart motion and detecting the corresponding abnormalities play an essential role in the diagnosis of cardiovascular diseases. Based on functional images, which are subject to noise and segmentation/registration inaccuracies, regional heart motion analysis is acknowledged as a difficult problem and, therefore, incorporation of prior knowledge is desirable to enhance accuracy. Given noisy data and a nonlinear dynamic model to describe myocardial motion, an unscented Kalman smoother is proposed in this study to estimate the myocardial points. Due to the similarity between the statistical information of normal and abnormal heart motions, detecting and classifying abnormality is a challenging problem. We use the Shannons differential entropy of the distributions of potential classifier features to detect and locate regional heart motion abnormality. A naive Bayes classifier algorithm is constructed from the Shannons differential entropy of different features to automatically detect abnormal functional regions of the myocardium. Using 174 segmented short-axis magnetic resonance cines obtained from 58 subjects (21 normal and 37 abnormal), the proposed method is quantitatively evaluated by comparison with ground truth classifications by radiologists over 928 myocardial segments. The proposed method performed significantly better than other recent methods, and yielded an accuracy of 86.5% (base), 89.4% (mid-cavity) and 84.5% (apex). The overall classification accuracy was 87.1%. Furthermore, standard kappa statistic comparisons between the proposed method and visual wall motion scoring by radiologists showed that the proposed algorithm can yield a kappa measure of 0.73.

Direct Estimation of Cardiac Biventricular Volumes With an Adapted Bayesian Formulation

Accurate estimation of the ventricular volumes is essential to the assessment of global cardiac functions. The existing estimation methods are mostly restricted to the left ventricle (LV), and often require segmentation which is challenging and computationally expensive. This paper proposes to estimate the volumes of both LV and right ventricle (RV) jointly with an efficient segmentation-free method. The proposed method employs an adapted Bayesian formulation. It introduces a novel likelihood function to exploit multiple appearance features, and a novel prior probability model to incorporate the area correlation between LV and RV cavities. The method is validated on a comprehensive dataset containing 56 clinical subjects (3360 images in total). The experimental results demonstrate that the estimated biventricular volumes are highly correlated to their independent ground truth. As a result, the proposed method enables a direct, efficient, and accurate assessment of global cardiac functions.

Global assessment of cardiac function using image statistics in MRI

The cardiac ejection fraction (EF) depends on the volume variation of the left ventricle (LV) cavity during a cardiac cycle, and is an essential measure in the diagnosis of cardiovascular diseases. It is often estimated via manual segmentation of several images in a cardiac sequence, which is prohibitively time consuming, or via automatic segmentation, which is a challenging and computationally expensive task that may result in high estimation errors. In this study, we propose to estimate the EF in real-time directly from image statistics using machine learning technique. From a simple user input in only one image, we build for all the images in a subject dataset (200 images) a statistic based on the Bhattacharyya coefficient of similarity between image distributions. We demonstrate that these statistics are non-linearly related to the LV cavity areas and, therefore, can be used to estimate the EF via an Artificial Neural Network (ANN) directly. A comprehensive evaluation over 20 subjects demonstrated that the estimated EFs correlate very well with those obtained from independent manual segmentations.

Sex Differences in Quadriceps Strength in OA

The purposes of this study were to determine 1) whether sex differences in quadriceps torque and isotonic power persist when controlling for muscle volume (i. e., torque/muscle volume and power/muscle volume) in participants with knee osteoarthritis (OA) and 2) the factors responsible for potential sex differences. Isometric torque, isotonic power (the product of torque and velocity, measured at 10, 20, 30, 40 and 50% maximal voluntary contraction; MVC) and maximal unloaded velocity were assessed in men (n=16, mean age=62.1 ± 7.2) and women (n=17, mean age=60.4 ± 4.3) with knee OA. Torque and power were normalized to muscle volume. The interpolated twitch technique was used to measure voluntary activation (VA) and evoked twitch and torque-frequency characteristics were measured to obtain information about muscle fibre distribution. Torque and power at all loads were significantly lower in women (p<0.05). Sex differences in power were reduced by 50% when controlling for muscle volume but were still significant at 10-40% MVC (p<0.05). No differences in VA, torque-frequency properties or time-to-peak tension of the evoked twitch were observed (p>0.05). These results suggest that only minor sex differences in torque and power persist when controlling for muscle volume. As VA and contractile property differences were not observed, other factors seem to be responsible.

A parameterization of deformation fields for diffeomorphic image registration and its application to myocardial delineation

This study investigates a new parameterization of deformation fields for image registration. Instead of standard displacements, this parameterization describes a deformation field with its transformation Jacobian and curl of end velocity field. It has two important features which make it appealing to image registration: 1) it relaxes the need of an explicit regularization term and the corresponding ad hoc weight in the cost functional; 2) explicit constraints on transformation Jacobian such as topology preserving and incompressibility constraints are straightforward to impose in a unified framework. In addition, this parameterization naturally describes a deformation field in terms of radial and rotational components, making it especially suited for processing cardiac data. We formulate diffeomorphic image registration as a constrained optimization problem which we solve with a step-then-correct strategy. The effectiveness of the algorithm is demonstrated with several examples and a comprehensive evaluation of myocardial delineation over 120 short-axis cardiac cine MRIs acquired from 20 subjects. It shows competitive performance in comparison to two recent segmentation based approaches.

Extraspinal Findings at Lumbar Spine CT Examinations: Prevalence and Clinical Importance

PURPOSE To prospectively determine the prevalence and clinical importance of extraspinal abnormalities in adult outpatients undergoing computed tomography (CT) of the lumbar spine. MATERIALS AND METHODS Institutional review board approval was obtained for this prospective study. Informed consent was obtained from 400 consecutive adult outpatients (mean age, 49 years; 212 male and 188 female patients) undergoing lumbar spine CT for low back pain and/or radiculopathy. Those with known malignancy were excluded. Dedicated spinal and abdominal full-field-of-view (FOV) images for each patient were reviewed by at least one neuroradiologist and two body radiologists. Extraspinal abnormalities were classified according to the CT Colonography Reporting and Data System (C-RADS). The electronic medical record of the patients with C-RADS E3 and E4 extraspinal findings were reviewed to assess how many of these findings were previously unknown, and the patients were followed up 24-36 months after the initial CT to determine their work-up and outcome. RESULTS Extraspinal findings were present on images in 162 (40.5%) of 400 lumbar spine CT examinations; 59 (14.8%) patients had indeterminate or clinically important findings requiring clinical correlation or further evaluation. After review of the electronic medical record, the prevalence of clinically important findings was 4.3%, comprising an early-stage renal cell carcinoma and transitional cell carcinoma, chronic lymphocytic leukemia, sarcoidosis, and 13 abdominal aortic aneurysms. Excluding anatomic variants, the full FOV was required to best visualize extraspinal abnormalities in 127 (79.4%) of 160 patients. CONCLUSION Reviewing the full-FOV images from lumbar spine CT examinations will result in the detection of a small number of substantial extraspinal pathologic findings in addition to many benign incidental findings.

Disease severity and knee extensor force in knee osteoarthritis: Data from the Osteoarthritis Initiative

To determine whether the method of disease severity measurement influences the magnitude of knee extensor force deficits in knee osteoarthritis (OA).