Faculty Bibliography

PURPOSE: To evaluate the feasibility and perform initial comparative evaluations of a 5-minute comprehensive whole-heart magnetic resonance imaging (MRI) protocol with four image acquisition types: perfusion (PERF), function (CINE), coronary artery imaging (CAI), and late gadolinium enhancement (LGE). MATERIALS AND METHODS: This study protocol was Health Insurance Portability and Accountability Act (HIPAA)-compliant and Institutional Review Board-approved. A 5-minute comprehensive whole-heart MRI examination protocol (Accelerated) using 6-8-fold-accelerated volumetric parallel imaging was incorporated into and compared with a standard 2D clinical routine protocol (Standard). Following informed consent, 20 patients were imaged with both protocols. Datasets were reviewed for image quality using a 5-point Likert scale (0 = non-diagnostic, 4 = excellent) in blinded fashion by two readers. RESULTS: Good image quality with full whole-heart coverage was achieved using the accelerated protocol, particularly for CAI, although significant degradations in quality, as compared with traditional lengthy examinations, were observed for the other image types. Mean total scan time was significantly lower for the Accelerated as compared to Standard protocols (28.99 +/- 4.59 min vs. 1.82 +/- 0.05 min, P < 0.05). Overall image quality for the Standard vs. Accelerated protocol was 3.67 +/- 0.29 vs. 1.5 +/- 0.51 (P < 0.005) for PERF, 3.48 +/- 0.64 vs. 2.6 +/- 0.68 (P < 0.005) for CINE, 2.35 +/- 1.01 vs. 2.48 +/- 0.68 (P = 0.75) for CAI, and 3.67 +/- 0.42 vs. 2.67 +/- 0.84 (P < 0.005) for LGE. Diagnostic image quality for Standard vs. Accelerated protocols was 20/20 (100%) vs. 10/20 (50%) for PERF, 20/20 (100%) vs. 18/20 (90%) for CINE, 18/20 (90%) vs. 18/20 (90%) for CAI, and 20/20 (100%) vs. 18/20 (90%) for LGE. CONCLUSION: This study demonstrates the technical feasibility and promising image quality of 5-minute comprehensive whole-heart cardiac examinations, with simplified scan prescription and high spatial and temporal resolution enabled by highly parallel imaging technology. The study also highlights technical hurdles that remain to be addressed. Although image quality remained diagnostic for most scan types, the reduced image quality of PERF, CINE, and LGE scans in the Accelerated protocol remain a concern. J. Magn. Reson. Imaging 2012. (c) 2012 Wiley Periodicals, Inc.

OBJECTIVES: We compared the image quality and diagnostic performance of 2 fat-suppression methods for black-blood T2-weighted fast spin-echo (FSE), which are as follows: (a) short T1 inversion recovery (STIR; FSE-STIR) and (b) spectral adiabatic inversion recovery (SPAIR; FSE-SPAIR), for detection of acute myocardial injury. BACKGROUND: Edema-sensitive T2-weighted FSE cardiac magnetic resonance (CMR) imaging is useful in detecting acute myocardial injury but may experience reduced myocardial signal and signal dropout. The SPAIR pulse aims to eliminate artifacts associated with the STIR pulse. MATERIALS AND METHODS: A total of 65 consecutive patients referred for CMR evaluation of myocardial structure and function underwent FSE-STIR and FSE-SPAIR, in addition to cine and late gadolinium enhancement (LGE) CMR. T2-weighted FSE images were independently evaluated by 2 readers for image quality and artifacts (Likert scale of 1-5; best-worst) and presence of increased myocardial signal suggestive of edema. In addition, clinical CMR interpretation, incorporating all CMR sequences available, was recorded for comparison. Diagnostic performance of each T2-weighted sequence was measured using recent (<30 days) troponin elevation greater than 2 times the upper limit of normal as the reference standard for acute myocardial injury. RESULTS: Of the 65 patients, there were 21 (32%) with acute myocardial injury. Image quality and artifact scores were significantly better with FSE-SPAIR compared with FSE-STIR (2.15 vs 2.68, P < 0.01; 2.62 vs 3.05, P < 0.01, respectively). The sensitivity, specificity, positive predictive value, and negative predictive value for acute myocardial injury were as follows: 29%, 93%, 67%, and 73% for FSE-SPAIR; 38%, 91%, 67%, and 75% for FSE-STIR; 71%, 98%, 94%, and 88% for clinical interpretation including LGE, T2, and wall motion. There was a statistically significant difference in sensitivity between the clinical interpretation and each of the T2-weighted sequences but not between each T2-weighted sequence. CONCLUSIONS: Although FSE-SPAIR demonstrated significantly improved image quality and decreased artifacts, isolated interpretations of each T2-weighted technique demonstrated high specificity but overall low sensitivity for the detection of myocardial injury, with no difference in accuracy between the techniques. However, real-world interpretation in combination with cine and LGE CMR methods significantly improves the overall sensitivity and diagnostic performance.

We introduce a novel algorithm for segmenting the high resolution CT images of the left ventricle (LV), particularly the papillary muscles and the trabeculae. High quality segmentations of these structures are necessary in order to better understand the anatomical function and geometrical properties of LV. These fine structures, however, are extremely challenging to capture due to their delicate and complex nature in both geometry and topology. Our algorithm computes the potential missing topological structures of a given initial segmentation. Using techniques from computational topology, e.g. persistent homology, our algorithm find topological handles which are likely to be the true signal. To further increase accuracy, these proposals are measured by the saliency and confidence from a trained classifier. Handles with high scores are restored in the final segmentation, leading to high quality segmentation results of the complex structures.

Deformable models have been widely used with success in medical image analysis. They combine bottom-up information derived from image appearance cues, with top-down shape-based constraints within a physics-based formulation. However, in many real world problems the observations extracted from the image data often contain gross errors, which adversely affect the deformation accuracy. To alleviate this issue, we introduce a new family of deformable models that are inspired from compressed sensing, a technique for efficiently reconstructing a signal based on its sparseness in some domain. In this problem, we employ sparsity to represent the outliers or gross errors, and combine it seamlessly with deformable models. The proposed new formulation is applied to the analysis of cardiac motion, using tagged magnetic resonance imaging (tMRI), where the automated tagging line tracking results are very noisy due to the poor image quality. Our new deformable models track the heart motion robustly, and the resulting strains are consistent with those calculated from manual labels.

BACKGROUND: We sought to determine regional myofiber stress after Coapsys device (Myocor, Inc, Maple Grove, MN) implantation using a finite element model of the left ventricle (LV). Chronic ischemic mitral regurgitation is caused by LV remodeling after posterolateral myocardial infarction. The Coapsys device consists of a single trans-LV chord placed below the mitral valve such that when tensioned it alters LV shape and decreases chronic ischemic mitral regurgitation. METHODS: Finite element models of the LV were based on magnetic resonance images obtained before (preoperatively) and after (postoperatively) coronary artery bypass grafting with Coapsys implantation in a single patient. To determine the effect of Coapsys and LV before stress, virtual Coapsys was performed on the preoperative model. Diastolic and systolic material variables in the preoperative, postoperative, and virtual Coapsys models were adjusted so that model LV volume agreed with magnetic resonance imaging data. Chronic ischemic mitral regurgitation was abolished in the postoperative models. In each case, myofiber stress and pump function were calculated. RESULTS: Both postoperative and virtual Coapsys models shifted end-systolic and end-diastolic pressure-volume relationships to the left. As a consequence and because chronic ischemic mitral regurgitation was reduced after Coapsys, pump function was unchanged. Coapsys decreased myofiber stress at end-diastole and end-systole in both the remote and infarct regions of the myocardium. However, knowledge of Coapsys and LV prestress was necessary for accurate calculation of LV myofiber stress, especially in the remote zone. CONCLUSIONS: Coapsys decreases myofiber stress at end-diastole and end-systole. The improvement in myofiber stress may contribute to the long-term effect of Coapsys on LV remodeling.

Deformable models and graph cuts are two standard image segmentation techniques. Combining some of their benefits, we introduce a new segmentation system for (semi-) automatic delineation of epicardium and endocardium of Left Ventricle of the heart in Magnetic Resonance Images (MRI). Specifically, a temporal information among consecutive phases is exploited via a coupling between deformable models and graph cuts which provides automated accurate cues for graph cuts and also good initialization scheme for deformable model that ultimately leads to more accurate and smooth segmentation results with lower interaction costs than using only graph cut segmentation. In addition, we define deformable model as a region defined by two nested contours and segment epicardium and endocardium in an unified way by optimizing single energy functional. This approach provides inherent coherency among the two contours thus leads to more accurate results than deforming separate contours for each target. We show promising results on the challenging problems of left ventricle segmentation.

This paper proposes an efficient algorithm to simultaneously reconstruct multiple T1/T2-weighted images of the same anatomical cross section from partially sampled k-space data. The simultaneous reconstruction problem is formulated as minimizing a linear combination of three terms corresponding to a least square data fitting, joint total-variation (TV) and group wavelet-sparsity regularization. It is rooted in two observations: (1) the variance of image gradients should be similar for the same spatial position across multiple contrasts; (2) the wavelet coefficients of all images from the same anatomical cross section should have similar sparse modes. To efficiently solve this formulation, we decompose it into group sparsity and joint TV regularization subproblems, respectively. Finally, the reconstructed image is obtained from the weighted average of solutions from two subproblems in an iterative framework. We compare the proposed algorithm with previous methods on SRT24 multi-channel Brain Atlas Data. Experiments demonstrate its superior performance for multi-contrast MR image reconstruction.