Automated Quality-Controlled Cardiovascular Magnetic Resonance Pericardial Fat Quantification Using a Convolutional Neural Network in the UK Biobank

Abstract

Background: Pericardial adipose tissue (PAT) may represent a novel risk marker for cardiovascular disease. However, absence of rapid radiation-free PAT quantification methods has precluded its examination in large cohorts.Objectives: We developed a fully automated quality-controlled tool for cardiovascular magnetic resonance (CMR) PAT quantification in the UK Biobank (UKB).Methods: Image analysis comprised contouring an en-bloc PAT area on four-chamber cine images. We created a ground truth manual analysis dataset randomly split into training and test sets. We built a neural network for automated segmentation using a Multi-residual U-net architecture with incorporation of permanently active dropout layers to facilitate quality control of the model's output using Monte Carlo sampling. We developed an in-built quality control feature, which presents predicted Dice scores. We evaluated model performance against the test set (n = 87), the whole UKB Imaging cohort (n = 45,519), and an external dataset (n = 103). In an independent dataset, we compared automated CMR and cardiac computed tomography (CCT) PAT quantification. Finally, we tested association of CMR PAT with diabetes in the UKB (n = 42,928).Results: Agreement between automated and manual segmentations in the test set was almost identical to inter-observer variability (mean Dice score = 0.8). The quality control method predicted individual Dice scores with Pearson r = 0.75. Model performance remained high in the whole UKB Imaging cohort and in the external dataset, with medium–good quality segmentation in 94.3% (mean Dice score = 0.77) and 94.4% (mean Dice score = 0.78), respectively. There was high correlation between CMR and CCT PAT measures (Pearson r = 0.72, p-value 5.3 ×10−18). Larger CMR PAT area was associated with significantly greater odds of diabetes independent of age, sex, and body mass index.Conclusions: We present a novel fully automated method for CMR PAT quantification with good model performance on independent and external datasets, high correlation with reference standard CCT PAT measurement, and expected clinical associations with diabetes.

Deep-Learning Segmentation of Epicardial Adipose Tissue Using Four-Chamber Cardiac Magnetic Resonance Imaging

“…Recently, automatic total pericardial fat quantification has been developed in 4Ch cine MRI. Bard, Raisi-Estabragh et al [ 23 ] obtained segmentation performances (DSC EAT+PAT = 0.8) very similar to ours (DSC EAT+PAT = 0.88) on their respective test-set. In their study, only the end-diastolic frame had been segmented while we segmented the full 4Ch cine MRI and trained on three consecutive cine frames to leverage cine temporal information.…”

“…Second, while the pericardial fascia is not clearly visible on short axis views, which was often reduced to a thin line that may be blurred by partial volume effects, on four-chamber (4Ch) views, the pericardium is generally less affected by partial volume effects resulting in better visualization. As such, the four-chamber view is recommended for evaluating pericarditis [ 21 ] and is a frequent choice of orientation to quantify EAT, PAT, and pericardial fat [ 22 , 23 , 24 , 25 , 26 , 27 ]. Consequently, the EAT analysis in this study were based on quantification of its 2D area representation in 4Ch long-axis cine MRI views.…”

“…Indeed, fully automated methods applied on routine images, such as cine MRI, could be rapidly translated to the clinics. Bard et al [ 23 ] developed a deep learning method to quantify pericardial fat in 4Ch long-axis cine MRI and evaluated it on the UK BioBank dataset. However, the segmentation of pericardial fat (EAT + PAT) limits the evaluation of the distinct roles and clinical implications of epicardial fat compared to paracardial fat.…”

Epicardial Fat Expansion in Diabetic and Obese Patients With Heart Failure and Preserved Ejection Fraction—A Specific HFpEF Phenotype

“…In addition, CMR opens the possibility to a combined evaluation of potential underlying myocardial ischemia, storage diseases like amyloidosis, diffuse myocardial fibrosis (extracellular volume), and epicardial fat. Recently, a growing number of publications also demonstrate the easiness of epicardial fat quantification using artificial intelligence (AI) algorithms (69). Beside the heart, the same approach can also be applied to other organ regions in patients with HFpEF like the abdomen (70).…”

Cardiac MR: From Theory to Practice

“…Some frameworks additionally provide myocardial strain measures (299,342). Automated segmentation methods have also been proposed to quantitatively derive other important markers of cardiovascular disease such as volume of pericardial adipose tissue (343), and scarred tissue areas (from LGE images) (344)(345)(346)(347)(348). Moreover, few DL-based methods have proposed to automatically quantify myocardial tissue from native T 1 mapping (349,350) and myocardial blood flow from contrast-enhanced perfusion CMR (351,352).…”