Improved Quantification of Myocardium Scar in Late Gadolinium Enhancement Images: Deep Learning Based Image Fusion Approach.

BACKGROUND: Quantification of myocardium scarring in late gadolinium enhanced (LGE) cardiac magnetic resonance imaging can be challenging due to low scar-to-background contrast and low image quality. To resolve ambiguous LGE regions, experienced readers often use conventional cine sequences to accurately identify the myocardium borders. PURPOSE: To develop a deep learning model for combining LGE and cine images to improve the robustness and accuracy of LGE scar quantification. STUDY TYPE: Retrospective. POPULATION: A total of 191 hypertrophic cardiomyopathy patients: 1) 162 patients from two sites randomly split into training (50%; 81 patients), validation (25%, 40 patients), and testing (25%; 41 patients); and 2) an external testing dataset (29 patients) from a third site. FIELD STRENGTH/SEQUENCE: 1.5T, inversion-recovery segmented gradient-echo LGE and balanced steady-state free-precession cine sequences ASSESSMENT: Two convolutional neural networks (CNN) were trained for myocardium and scar segmentation, one with and one without LGE-Cine fusion. For CNN with fusion, the input was two aligned LGE and cine images at matched cardiac phase and anatomical location. For CNN without fusion, only LGE images were used as input. Manual segmentation of the datasets was used as reference standard. STATISTICAL TESTS: Manual and CNN-based quantifications of LGE scar burden and of myocardial volume were assessed using Pearson linear correlation coefficients (r) and Bland-Altman analysis. RESULTS: Both CNN models showed strong agreement with manual quantification of LGE scar burden and myocardium volume. CNN with LGE-Cine fusion was more robust than CNN without LGE-Cine fusion, allowing for successful segmentation of significantly more slices (603 [95%] vs. 562 (89%) of 635 slices; P < 0.001). Also, CNN with LGE-Cine fusion showed better agreement with manual quantification of LGE scar burden than CNN without LGE-Cine fusion (%ScarLGE-cine = 0.82 × %Scarmanual , r = 0.84 vs. %ScarLGE = 0.47 × %Scarmanual , r = 0.81) and myocardium volume (VolumeLGE-cine = 1.03 × Volumemanual , r = 0.96 vs. VolumeLGE = 0.91 × Volumemanual , r = 0.91). DATA CONCLUSION: CNN based LGE-Cine fusion can improve the robustness and accuracy of automated scar quantification. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: 1.

Three-dimensional Deep Convolutional Neural Networks for Automated Myocardial Scar Quantification in Hypertrophic Cardiomyopathy: A Multicenter Multivendor Study.

Background Cardiac MRI late gadolinium enhancement (LGE) scar volume is an important marker for outcome prediction in patients with hypertrophic cardiomyopathy (HCM); however, its clinical application is hindered by a lack of measurement standardization. Purpose To develop and evaluate a three-dimensional (3D) convolutional neural network (CNN)-based method for automated LGE scar quantification in patients with HCM. Materials and Methods We retrospectively identified LGE MRI data in a multicenter (n = 7) and multivendor (n = 3) HCM study obtained between November 2001 and November 2011. A deep 3D CNN based on U-Net architecture was used for LGE scar quantification. Independent CNN training and testing data sets were maintained with a 4:1 ratio. Stacks of short-axis MRI slices were split into overlapping substacks that were segmented and then merged into one volume. The 3D CNN per-site and per-vendor performances were evaluated with respect to manual scar quantification performed in a core laboratory setting using Dice similarity coefficient (DSC), Pearson correlation, and Bland-Altman analyses. Furthermore, the performance of 3D CNN was compared with that of two-dimensional (2D) CNN. Results This study included 1073 patients with HCM (733 men; mean age, 49 years ± 17 [standard deviation]). The 3D CNN-based quantification was fast (0.15 second per image) and demonstrated excellent correlation with manual scar volume quantification (r = 0.88, P < .001) and ratio of scar volume to total left ventricle myocardial volume (%LGE) (r = 0.91, P < .001). The 3D CNN-based quantification strongly correlated with manual quantification of scar volume (r = 0.82-0.99, P < .001) and %LGE (r = 0.90-0.97, P < .001) for all sites and vendors. The 3D CNN identified patients with a large scar burden (>15%) with 98% accuracy (202 of 207) (95% confidence interval [CI]: 95%, 99%). When compared with 3D CNN, 2D CNN underestimated scar volume (r = 0.85, P < .001) and %LGE (r = 0.83, P < .001). The DSC of 3D CNN segmentation was comparable among different vendors (P = .07) and higher than that of 2D CNN (DSC, 0.54 ± 0.26 vs 0.48 ± 0.29; P = .02). Conclusion In the hypertrophic cardiomyopathy population, a three-dimensional convolutional neural network enables fast and accurate quantification of myocardial scar volume, outperforms a two-dimensional convolutional neural network, and demonstrates comparable performance across different vendors. © RSNA, 2019 Online supplemental material is available for this article.

The relationship between the quantitative extent of late gadolinium enhancement and burden of nonsustained ventricular tachycardia in hypertrophic cardiomyopathy: A delayed contrast-enhanced magnetic resonance study.

OBJECTIVES: To examine the relationship between late gadolinium enhancement (LGE) extent and nonsustained ventricular tachycardia (NSVT) characteristics in patients with hypertrophic cardiomyopathy (HCM). BACKGROUND: NSVT has been shown to be independently associated with sudden cardiac death (SCD) in HCM. Previous studies have found LGE on cardiac magnetic resonance (CMR) to be independently associated with NSVT. METHODS: Seventy-three patients who had 14-day Holter monitoring for either risk stratification for SCD (n = 62) or evaluation of atrial fibrillation (n = 11) on a CMR study were included. Areas of LGE in left ventricle (LV) were visually identified and analyzed quantitatively for both high (≥6 SD above the mean signal intensity of normal myocardium) and intermediate (≥4 but <6 SD) LGE signal intensity. RESULTS: Patients with more extensive LGE had longer (P = 0.0028) and more frequent (P = 0.02) episodes of NSVT. In univariate analyses, frequency of NSVT was associated with LGE extent (rs = 0.43, P = 0.001), LV ejection fraction (rs = -0.38, P < 0.001), LV mass (rs = 0.32, P = 0.005), LV maximal wall thickness (rs = 0.28, P = 0.016), and left atrium diameter (rs = 0.29, P = 0.001); maximal length of NSVT was associated with LGE extent (rs = 0.52, P < 0.001), LV ejection fraction (rs = -0.44, P < 0.001), LV mass (rs = 0.37, P = 0.001), and left atrium diameter (rs = 0.3, P < 0.001). In multivariable analyses, LGE extent remained the sole variable independently associated with frequency (P = 0.001) and maximal length of episodes of NSVT (P = 0.001). No significant association was found between the rate of NSVT and LGE extent. CONCLUSIONS: LGE extent is independently associated with a greater burden and longer episodes of NSVT in HCM. These findings support the association between myocardial fibrosis as represented by LGE and ventricular tachyarrhythmias in HCM.

Prognostic value of quantitative contrast-enhanced cardiovascular magnetic resonance for the evaluation of sudden death risk in patients with hypertrophic cardiomyopathy.

BACKGROUND: Hypertrophic cardiomyopathy (HCM) is the most common cause of sudden death in the young, although not all patients eligible for sudden death prevention with an implantable cardioverter-defibrillator are identified. Contrast-enhanced cardiovascular magnetic resonance with late gadolinium enhancement (LGE) has emerged as an in vivo marker of myocardial fibrosis, although its role in stratifying sudden death risk in subgroups of HCM patients remains incompletely understood. METHODS AND RESULTS: We assessed the relation between LGE and cardiovascular outcomes in 1293 HCM patients referred for cardiovascular magnetic resonance and followed up for a median of 3.3 years. Sudden cardiac death (SCD) events (including appropriate defibrillator interventions) occurred in 37 patients (3%). A continuous relationship was evident between LGE by percent left ventricular mass and SCD event risk in HCM patients (P=0.001). Extent of LGE was associated with an increased risk of SCD events (adjusted hazard ratio, 1.46/10% increase in LGE; P=0.002), even after adjustment for other relevant disease variables. LGE of ≥15% of LV mass demonstrated a 2-fold increase in SCD event risk in those patients otherwise considered to be at lower risk, with an estimated likelihood for SCD events of 6% at 5 years. Performance of the SCD event risk model was enhanced by LGE (net reclassification index, 12.9%; 95% confidence interval, 0.3-38.3). Absence of LGE was associated with lower risk for SCD events (adjusted hazard ratio, 0.39; P=0.02). Extent of LGE also predicted the development of end-stage HCM with systolic dysfunction (adjusted hazard ratio, 1.80/10% increase in LGE; P<0.03). CONCLUSIONS: Extensive LGE measured by quantitative contrast enhanced CMR provides additional information for assessing SCD event risk among HCM patients, particularly patients otherwise judged to be at low risk.

Significance of left ventricular apical-basal muscle bundle identified by cardiovascular magnetic resonance imaging in patients with hypertrophic cardiomyopathy.

AIMS: Cardiovascular magnetic resonance (CMR) has improved diagnostic and management strategies in hypertrophic cardiomyopathy (HCM) by expanding our appreciation for the diverse phenotypic expression. We sought to characterize the prevalence and clinical significance of a recently identified accessory left ventricular (LV) muscle bundle extending from the apex to the basal septum or anterior wall (i.e. apical-basal). METHODS AND RESULTS: CMR was performed in 230 genotyped HCM patients (48 ± 15 years, 69% male), 30 genotype-positive/phenotype-negative (G+/P-) family members (32 ± 15 years, 30% male), and 126 controls. Left ventricular apical-basal muscle bundle was identified in 145 of 230 (63%) HCM patients, 18 of 30 (60%) G+/P- family members, and 12 of 126 (10%) controls (G+/P- vs. controls; P < 0.01). In HCM patients, the prevalence of an apical-basal muscle bundle was similar among those with disease-causing sarcomere mutations compared with patients without mutation (64 vs. 62%; P = 0.88). The presence of an LV apical-basal muscle bundle was not associated with LV outflow tract obstruction (P = 0.61). In follow-up, 33 patients underwent surgical myectomy of whom 22 (67%) were identified to have an accessory LV apical-basal muscle bundle, which was resected in all patients. CONCLUSION: Apical-basal muscle bundles are a unique myocardial structure commonly present in HCM patients as well as in G+/P- family members and may represent an additional morphologic marker for HCM diagnosis in genotype-positive status.

Risk stratification and outcome of patients with hypertrophic cardiomyopathy >=60 years of age.

BACKGROUND: Hypertrophic cardiomyopathy (HCM) is prominently associated with risk for sudden death and disease progression, largely in young patients. Whether patients of more advanced age harbor similar risks is unresolved, often creating clinical dilemmas, particularly in decisions for primary prevention of sudden death with implantable defibrillators. METHODS AND RESULTS: We studied 428 consecutive HCM patients presenting at ≥60 years of age and followed for 5.8±4.8 years; 53% were women. Of the 428 patients, 279 (65%) survived to 73±7 years of age (range, 61-96 years), most (n=245, 88%) with no/mild symptoms, including 135 with ≥1 conventional sudden death risk factors and 50 (37%) with late gadolinium enhancement. Over follow-up, 149 (35%) died at 80±8 years of age, mostly from non-HCM-related causes (n=133, 31%), including a substantial proportion from noncardiac disease (n=54). Sixteen patients (3.7%) had HCM-related mortality events (0.64%/y), including embolic stroke (n=6), progressive heart failure or transplantation (n=3), postoperative complications (n=2), and arrhythmic sudden death events (n=5, 1.2% [0.20%/y]). All-cause mortality was increased in HCM patients ≥60 years of age compared with an age-matched US general population, predominantly as a result of non-HCM-related diseases (P<0.001; standard mortality ratio, 1.5). CONCLUSIONS: HCM patients surviving into the seventh decade of life are at low risk for disease-related morbidity/mortality, including sudden death, even with conventional risk factors. These data do not support aggressive prophylactic defibrillator implantation at advanced ages in HCM. Other cardiac or noncardiac comorbidities have a greater impact on survival than HCM in older patients.

Accelerated isotropic sub-millimeter whole-heart coronary MRI: compressed sensing versus parallel imaging.

PURPOSE: To enable accelerated isotropic sub-millimeter whole-heart coronary MRI within a 6-min acquisition and to compare this with a current state-of-the-art accelerated imaging technique at acceleration rates beyond what is used clinically. METHODS: Coronary MRI still faces major challenges, including lengthy acquisition time, low signal-to-noise-ratio (SNR), and suboptimal spatial resolution. Higher spatial resolution in the sub-millimeter range is desirable, but this results in increased acquisition time and lower SNR, hindering its clinical implementation. In this study, we sought to use an advanced B1-weighted compressed sensing technique for highly accelerated sub-millimeter whole-heart coronary MRI, and to compare the results to parallel imaging, the current-state-of-the-art, where both techniques were used at acceleration rates beyond what is used clinically. Two whole-heart coronary MRI datasets were acquired in seven healthy adult subjects (30.3 ± 12.1 years; 3 men), using prospective 6-fold acceleration, with random undersampling for the proposed compressed sensing technique and with uniform undersampling for sensitivity encoding reconstruction. Reconstructed images were qualitatively compared in terms of image scores and perceived SNR on a four-point scale (1 = poor, 4 = excellent) by an experienced blinded reader. RESULTS: The proposed technique resulted in images with clear visualization of all coronary branches. Overall image quality and perceived SNR of the compressed sensing images were significantly higher than those of parallel imaging (P = 0.03 for both), which suffered from noise amplification artifacts due to the reduced SNR. CONCLUSION: The proposed compressed sensing-based reconstruction and acquisition technique for sub-millimeter whole-heart coronary MRI provides 6-fold acceleration, where it outperforms parallel imaging with uniform undersampling.

3D late gadolinium enhancement in a single prolonged breath-hold using supplemental oxygenation and hyperventilation.

PURPOSE: To evaluate the feasibility of three-dimensional (3D) single breath-hold late gadolinium enhancement (LGE) of the left ventricle (LV) using supplemental oxygen and hyperventilation and compressed-sensing acceleration. METHODS: Breath-hold metrics [breath-hold duration, diaphragmatic/LV position drift, and maximum variation of R wave to R wave (RR) interval] without and with supplemental oxygen and hyperventilation were assessed in healthy adult subjects using a real-time single shot acquisition. Ten healthy subjects and 13 patients then underwent assessment of the proposed 3D breath-hold LGE acquisition (field of view = 320 × 320 × 100 mm(3) , resolution = 1.6 × 1.6 × 5.0 mm(3) , acceleration rate of 4) and a free-breathing acquisition with right hemidiaphragm navigator (NAV) respiratory gating. Semiquantitative grading of overall image quality, motion artifact, myocardial nulling, and diagnostic value was performed by consensus of two blinded observers. RESULTS: Supplemental oxygenation and hyperventilation increased the breath-hold duration (35 ± 11 s to 58 ± 21 s; P < 0.0125) without significant impact on diaphragmatic/LV position drift or maximum variation of RR interval (both P > 0.01). LGE images were of similar quality when compared with free-breathing acquisitions, but with reduced total scan time (85 ± 22 s to 35 ± 6 s; P < 0.001). CONCLUSIONS: Supplemental oxygenation and hyperventilation allow for prolonged breath-holding and enable single breath-hold 3D accelerated LGE with similar image quality as free breathing with NAV.

Radiomics of Late Gadolinium Enhancement Reveals Prognostic Value of Myocardial Scar Heterogeneity in Hypertrophic Cardiomyopathy.

BACKGROUND: Late gadolinium enhancement (LGE) scar burden by cardiac magnetic resonance is a major risk factor for sudden cardiac death (SCD) in hypertrophic cardiomyopathy (HCM). However, there is currently limited data on the incremental prognostic value of integrating myocardial LGE radiomics (ie, shape and texture features) into SCD risk stratification models. OBJECTIVES: The purpose of this study was to investigate the incremental prognostic value of myocardial LGE radiomics beyond current European Society of Cardiology (ESC) and American College of Cardiology (ACC)/American Heart Association (AHA) models for SCD risk prediction in HCM. METHODS: A total of 1,229 HCM patients (62% men; age 52 ± 16 years) from 3 medical centers were included. Left ventricular myocardial radiomic features were calculated from LGE images. Principal component analysis was used to reduce the radiomic features and calculate 3 principal radiomics (PrinRads). Cox and logistic regression analyses were then used to evaluate the significance of the extracted PrinRads of LGE images, alone or in combination with ESC or ACC/AHA models, to predict SCD risk. The ACC/AHA risk markers include LGE burden using a dichotomized 15% threshold of LV scar. RESULTS: SCD events occurred in 30 (2.4%) patients over a follow-up period of 49 ± 28 months. Risk prediction using PrinRads resulted in higher c-statistics than the ESC (0.69 vs 0.57; P = 0.02) and the ACC/AHA (0.69 vs 0.67; P = 0.75) models. Risk predictions were improved by combining the 3 PrinRads with ESC (0.73 vs 0.57; P < 0.01) or ACC/AHA (0.76 vs 0.67; P < 0.01) risk scores. The net reclassification index was improved by combining the PrinRads with ESC (0.25 [95% CI: 0.08-0.43]; P = 0.005) or ACC/AHA (0.05 [95% CI: -0.07 to 0.16]; P = 0.42) models. One PrinRad was a significant predictor of SCD risk (HR: 0.57 [95% CI: 0.39-0.84]; P = 0.01). LGE heterogeneity was a major component of PrinRads and a significant predictor of SCD risk (HR: 0.07 [95% CI: 0.01-0.75]; P = 0.03). CONCLUSIONS: Myocardial LGE radiomics are strongly associated with SCD risk in HCM and provide incremental risk stratification beyond current ESC or AHA/ACC risk models. Our proof-of-concept study warrants further validation.

Relationship Between Genotype Status and Clinical Outcome in Hypertrophic Cardiomyopathy.

BACKGROUND: The genetic basis of hypertrophic cardiomyopathy (HCM) is complex, and the relationship between genotype status and clinical outcome is incompletely resolved. METHODS AND RESULTS: We assessed a large international HCM cohort to define in contemporary terms natural history and clinical consequences of genotype. Consecutive patients (n=1468) with established HCM diagnosis underwent genetic testing. Patients with pathogenic (or likely pathogenic) variants were considered genotype positive (G+; n=312; 21%); those without definite disease-causing mutations (n=651; 44%) or variants of uncertain significance (n=505; 35%) were considered genotype negative (G-). Patients were followed up for a median of 7.8 years (interquartile range, 3.5-13.4 years); HCM end points were examined by cumulative event incidence. Over follow-up, 135 (9%) patients died, 33 from a variety of HCM-related causes. After adjusting for age, all-cause and HCM-related mortality did not differ between G- versus G+ patients (hazard ratio [HR], 0.78 [95% CI, 0.46-1.31]; P=0.37; HR, 0.93 [95% CI, 0.38-2.30]; P=0.87, respectively). Adverse event rates, including heart failure progression to class III/IV, heart transplant, or heart failure death, did not differ (G- versus G+) when adjusted for age (HR, 1.20 [95% CI, 0.63-2.26]; P=0.58), nor was genotype independently associated with sudden death event risk (HR, 1.39 [95% CI, 0.88-2.21]; P=0.16). In multivariable analysis, age was the only independent predictor of all-cause and HCM-related mortality, heart failure progression, and sudden death events. CONCLUSIONS: In this large consecutive cohort of patients with HCM, genotype (G+ or G-) was not a predictor of clinical course, including all-cause and HCM-related mortality and risk for heart failure progression or sudden death. G+ status should not be used to dictate clinical management or predict outcome in HCM.

Free-breathing 3D cardiac MRI using iterative image-based respiratory motion correction.

Respiratory motion compensation using diaphragmatic navigator gating with a 5 mm gating window is conventionally used for free-breathing cardiac MRI. Because of the narrow gating window, scan efficiency is low resulting in long scan times, especially for patients with irregular breathing patterns. In this work, a new retrospective motion compensation algorithm is presented to reduce the scan time for free-breathing cardiac MRI that increasing the gating window to 15 mm without compromising image quality. The proposed algorithm iteratively corrects for respiratory-induced cardiac motion by optimizing the sharpness of the heart. To evaluate this technique, two coronary MRI datasets with 1.3 mm(3) resolution were acquired from 11 healthy subjects (seven females, 25 ± 9 years); one using a navigator with a 5 mm gating window acquired in 12.0 ± 2.0 min and one with a 15 mm gating window acquired in 7.1 ± 1.0 min. The images acquired with a 15 mm gating window were corrected using the proposed algorithm and compared to the uncorrected images acquired with the 5 and 15 mm gating windows. The image quality score, sharpness, and length of the three major coronary arteries were equivalent between the corrected images and the images acquired with a 5 mm gating window (P-value > 0.05), while the scan time was reduced by a factor of 1.7.

Contrast-enhanced specific absorption rate-efficient 3D cardiac cine with respiratory-triggered radiofrequency gating.

PURPOSE: To investigate the use of radiofrequency (RF) gating in conjunction with a paramagnetic contrast agent to reduce the specific absorption rate (SAR) and increase the blood-myocardium contrast in balanced steady-state free precession (bSSFP) 3D cardiac cine. MATERIALS AND METHODS: RF gating was implemented by synchronizing the RF-excitation with an external respiratory sensor (bellows), which could additionally be used for respiratory gating. For reference, respiratory-gated 3D cine images were acquired without RF gating. Free-breathing 3D cine images were acquired in eight healthy subjects before and after contrast injection (Gd-BOPTA) and compared to breath-hold 2D cine. RESULTS: RF-gated 3D cine reduced the SAR by nearly 40% without introducing significant artifacts while providing left ventricle (LV) measurements similar to those obtained with 2D cine. The contrast-to-noise ratio (CNR) was significantly higher for 3D cine compared to 2D cine, both before and after contrast injection; however, no statistically significant CNR increase was observed for the postcontrast 3D cine compared to the precontrast acquisitions. CONCLUSION: Respiratory-triggered RF gating significantly reduces SAR in 3D cine acquisitions, which may enable a more widespread clinical use of 3D cine. Furthermore, CNR of 3D bSSFP cine is higher than of 2D and administration of Gd-BOPTA does not improve the CNR of 3D cine.

Scar heterogeneity on cardiovascular magnetic resonance as a predictor of appropriate implantable cardioverter defibrillator therapy.

BACKGROUND: Despite the survival benefit of implantable-cardioverter-defibrillators (ICDs), the vast majority of patients receiving an ICD for primary prevention do not receive ICD therapy. We sought to assess the role of heterogeneous scar area (HSA) identified by late gadolinium enhancement cardiovascular magnetic resonance (LGE-CMR) in predicting appropriate ICD therapy for primary prevention of sudden cardiac death (SCD). METHODS: From September 2003 to March 2011, all patients who underwent primary prevention ICD implantation and had a pre-implantation LGE-CMR were identified. Scar size was determined using thresholds of 4 and 6 standard deviations (SD) above remote normal myocardium; HSA was defined using 3 different criteria; as the region between 2 SD and 4 SD (HSA2-4SD), between 2SD and 6SD (HSA2-6SD), and between 4SD and 6SD (HSA4-6SD). The end-point was appropriate ICD therapy. RESULTS: Out of 40 total patients followed for 25 ± 24 months, 7 had appropriate ICD therapy. Scar size measured by different thresholds was similar in ICD therapy and non-ICD therapy groups (P = NS for all). However, HSA2-4SD and HSA4-6SD were significantly larger in the ICD therapy group (P = 0.001 and P = 0.03, respectively). In multivariable model HSA2-4SD was the only significant independent predictor of ICD therapy (HR = 1.08, 95%CI: 1.00-1.16, P = 0.04). Kaplan-Meier analysis showed that patients with greater HSA2-4SD had a lower survival free of appropriate ICD therapy (P = 0.026). CONCLUSIONS: In primary prevention ICD implantation, LGE-CMR HSA identifies patients with appropriate ICD therapy. If confirmed in larger series, HSA can be used for risk stratification in primary prevention of SCD.

Interpretable machine learning for automated left ventricular scar quantification in hypertrophic cardiomyopathy patients.

Scar quantification on cardiovascular magnetic resonance (CMR) late gadolinium enhancement (LGE) images is important in risk stratifying patients with hypertrophic cardiomyopathy (HCM) due to the importance of scar burden in predicting clinical outcomes. We aimed to develop a machine learning (ML) model that contours left ventricular (LV) endo- and epicardial borders and quantifies CMR LGE images from HCM patients.We retrospectively studied 2557 unprocessed images from 307 HCM patients followed at the University Health Network (Canada) and Tufts Medical Center (USA). LGE images were manually segmented by two experts using two different software packages. Using 6SD LGE intensity cutoff as the gold standard, a 2-dimensional convolutional neural network (CNN) was trained on 80% and tested on the remaining 20% of the data. Model performance was evaluated using the Dice Similarity Coefficient (DSC), Bland-Altman, and Pearson's correlation. The 6SD model DSC scores were good to excellent at 0.91 ± 0.04, 0.83 ± 0.03, and 0.64 ± 0.09 for the LV endocardium, epicardium, and scar segmentation, respectively. The bias and limits of agreement for the percentage of LGE to LV mass were low (-0.53 ± 2.71%), and correlation high (r = 0.92). This fully automated interpretable ML algorithm allows rapid and accurate scar quantification from CMR LGE images. This program does not require manual image pre-processing, and was trained with multiple experts and software, increasing its generalizability.

Thin, soft, wearable system for continuous wireless monitoring of artery blood pressure.

Continuous monitoring of arterial blood pressure (BP) outside of a clinical setting is crucial for preventing and diagnosing hypertension related diseases. However, current continuous BP monitoring instruments suffer from either bulky systems or poor user-device interfacial performance, hampering their applications in continuous BP monitoring. Here, we report a thin, soft, miniaturized system (TSMS) that combines a conformal piezoelectric sensor array, an active pressure adaptation unit, a signal processing module, and an advanced machine learning method, to allow real wearable, continuous wireless monitoring of ambulatory artery BP. By optimizing the materials selection, control/sampling strategy, and system integration, the TSMS exhibits improved interfacial performance while maintaining Grade A level measurement accuracy. Initial trials on 87 volunteers and clinical tracking of two hypertension individuals prove the capability of the TSMS as a reliable BP measurement product, and its feasibility and practical usability in precise BP control and personalized diagnosis schemes development.

Accelerated late gadolinium enhancement cardiac MR imaging with isotropic spatial resolution using compressed sensing: initial experience.

PURPOSE: To evaluate the use of low-dimensional-structure self-learning and thresholding (LOST) compressed sensing acquisition and reconstruction in the assessment of left atrial (LA) and left ventricular (LV) scar by using late gadolinium enhancement (LGE) magnetic resonance (MR) imaging with isotropic spatial resolution. MATERIALS AND METHODS: The study was approved by the local institutional review board and was compliant with HIPAA. All subjects provided written informed consent. Twenty-eight patients (eight women; mean age, 58.0 years ± 10.1) with a history of atrial fibrillation were recruited for the LA LGE study, and 14 patients (five women; mean age, 54.2 years ± 18.6) were recruited for assessment of LV myocardial infarction. With use of a pseudorandom k-space undersampling pattern, threefold accelerated three-dimensional (3D) LGE data were acquired with isotropic spatial resolution and reconstructed off-line by using LOST. For comparison, subjects were also imaged by using standard 3D LGE protocols with nonisotropic spatial resolution. Images were compared qualitatively by three cardiologists with regard to diagnostic value, presence of enhancement, and image quality. The signed rank test and Wilcoxon unpaired two-sample test were used to test the hypothesis that there would be no significant difference in image quality ratings with different resolutions. RESULTS: Interpretable images were obtained in 26 of the 28 patients (93%) in the LA LGE study. LGE was seen in 17 of 30 cases (57%) with nonisotropic resolution and in 18 cases (60%) with isotropic resolution. Diagnostic quality scores of isotropic images were significantly higher than those of nonisotropic images with coronal views (median, 3 vs 2, respectively [25th and 75th percentiles: 3, 3 vs 2, 3]; P < .001) and sagittal views (median, 3 vs 2 [25th and 75th percentiles: 3, 4 vs 2, 3]; P < .001) but lower with axial views (median, 4 vs 3 [25th and 75th percentiles: 3, 4 vs 3, 3]; P < .001). For the LV LGE study, all patients had interpretable images. LGE was seen in six of 14 patients (43%), with 100% agreement between both data sets. Diagnostic quality scores of high-isotropic-resolution LV images were higher than those of nonisotropic images with short-axis views (median, 4 vs 3 [25th and 75th percentiles: 3, 4 vs 2, 3]; P = .014) and two-chamber views (median, 4 vs 3 [25th and 75th percentiles: 3, 4 vs 2, 3]; P = .001). CONCLUSION: An accelerated LGE acquisition with LOST enables imaging with high isotropic spatial resolution for improved assessment of LV, LA, and pulmonary vein scar.

Accelerated contrast-enhanced whole-heart coronary MRI using low-dimensional-structure self-learning and thresholding.

We sought to evaluate the efficacy of prospective random undersampling and low-dimensional-structure self-learning and thresholding reconstruction for highly accelerated contrast-enhanced whole-heart coronary MRI. A prospective random undersampling scheme was implemented using phase ordering to minimize artifacts due to gradient switching and was compared to a randomly undersampled acquisition with no profile ordering. This profile-ordering technique was then used to acquire contrast-enhanced whole-heart coronary MRI in 10 healthy subjects with 4-fold acceleration. Reconstructed images and the acquired zero-filled images were compared for depicted vessel length, vessel sharpness, and subjective image quality on a scale of 1 (poor) to 4 (excellent). In a pilot study, contrast-enhanced whole-heart coronary MRI was also acquired in four patients with suspected coronary artery disease with 3-fold acceleration. The undersampled images were reconstructed using low-dimensional-structure self-learning and thresholding, which showed significant improvement over the zero-filled images in both objective and subjective measures, with an overall score of 3.6 ± 0.5. Reconstructed images in patients were all diagnostic. Low-dimensional-structure self-learning and thresholding reconstruction allows contrast-enhanced whole-heart coronary MRI with acceleration as high as 4-fold using clinically available five-channel phased-array coil.

Free-breathing cardiac MR with a fixed navigator efficiency using adaptive gating window size.

A respiratory navigator with a fixed acceptance gating window is commonly used to reduce respiratory motion artifacts in cardiac MR. This approach prolongs the scan time and occasionally yields an incomplete dataset due to respiratory drifts. To address this issue, we propose an adaptive gating window approach in which the size and position of the gating window are changed adaptively during the acquisition based on the individual's breathing pattern. The adaptive gating window tracks the breathing pattern of the subject throughout the scan and adapts the size and position of the gating window such that the gating efficiency is always fixed at a constant value. To investigate the image quality and acquisition time, free breathing cardiac MRI, including both targeted coronary MRI and late gadolinium enhancement imaging, was performed in 67 subjects using the proposed navigator technique. Targeted coronary MRI was acquired from eleven healthy adult subjects using both the conventional and proposed adaptive gating window techniques. Fifty-six patients referred for cardiac MRI were also imaged using late gadolinium enhancement with the proposed adaptive gating window technique. Subjective and objective image assessments were used to evaluate the proposed method. The results demonstrate that the proposed technique allows free-breathing cardiac MRI in a relatively fixed time without compromising imaging quality due to respiratory motion artifacts.

Virtual histology findings and effects of varying doses of atorvastatin on coronary plaque volume and composition in statin-naive patients: the VENUS study.

BACKGROUND: While statin induces plaque regression, its effects, particularly with different doses on plaque virtual histology composition, remain unknown. METHODS AND RESULTS: In this prospective, randomized, double-blinded study, 40 consecutive statin-naive patients with stable angina requiring percutaneous coronary intervention (PCI) were randomized to 2 arms (20 patients each) receiving 6 months of atorvastatin 10 mg or 40 mg daily. The primary end-point was (VH-IVUS) changes from baseline to 6 months, as assessed by a core laboratory. Fifty-four VH-IVUS lesions were analyzed from the 10 mg group and 57 from the 40 mg group. Overall, plaque volume was reduced by 4.28% (-5.10±14.93 mm(3), P<0.001), absolute VH-IVUS fibrous volume by 10.54% (-4.87±10.74 mm(3), P<0.001), and relative percentage fibrous component by 3.29±7.84% (P<0.001), while relative percentage dense calcium increased by 1.50±3.08% (P<0.001), and necrotic core by 3.19±7.82% (P<0.001). Beneficial changes were more substantial in the higher dose (40 mg) group, with significantly more percentage plaque volume regression (-1.50±3.85% vs. 0.38±4.05% increase in the 10 mg group, P=0.014), less relative percentage necrotic core expansion (1.68±7.57% vs. 4.78±7.82% in the 10 mg group, P=0.037), and without occurrence of major adverse cardiac events (vs. 6 patients in the 10 mg group, P=0.020). CONCLUSIONS: In statin-naive patients requiring PCI, 6 months of atorvastatin induced a significant percentage of plaque volume reduction and substantial modification of VH-IVUS composition. In addition, these effects appeared to vary with different doses of atorvastatin, showing significantly better limitation of relative percentage necrotic core expansion at a higher dose.