Search

We perform unsupervised analysis of image-derived shape and motion features extracted from 3,822 cardiac Magnetic resonance imaging (MRIs) of the UK Biobank. First, with a feature extraction method previously published based on deep learning models, we extract from each case 9 feature values characterizing both the cardiac shape and motion. Second, a feature selection is performed to remove highly correlated feature pairs. Third, clustering is carried out using a Gaussian mixture model on the selected features. After analysis, we identify 2 small clusters that probably correspond to 2 pathological categories. Further confirmation using a trained classification model and dimensionality reduction tools is carried out to support this finding. Moreover, we examine the differences between the other large clusters and compare our measures with the ground truth.

Cardiovascular magnetic resonance (CMR) radiomics is a novel technique for advanced cardiac image phenotyping by analyzing multiple quantifiers of shape and tissue texture. In this paper, we assess, in the largest sample published to date, the performance of CMR radiomics models for identifying changes in cardiac structure and tissue texture due to cardiovascular risk factors. We evaluated five risk factor groups from the first 5,065 UK Biobank participants: hypertension (n = 1,394), diabetes (n = 243), high cholesterol (n = 779), current smoker (n = 320), and previous smoker (n = 1,394). Each group was randomly matched with an equal number of healthy comparators (without known cardiovascular disease or risk factors). Radiomics analysis was applied to short axis images of the left and right ventricles at end-diastole and end-systole, yielding a total of 684 features per study. Sequential forward feature selection in combination with machine learning (ML) algorithms (support vector machine, random forest, and logistic regression) were used to build radiomics signatures for each specific risk group. We evaluated the degree of separation achieved by the identified radiomics signatures using area under curve (AUC), receiver operating characteristic (ROC), and statistical testing. Logistic regression with L1-regularization was the optimal ML model. Compared to conventional imaging indices, radiomics signatures improved the discrimination of risk factor vs. healthy subgroups as assessed by AUC [diabetes: 0.80 vs. 0.70, hypertension: 0.72 vs. 0.69, high cholesterol: 0.71 vs. 0.65, current smoker: 0.68 vs. 0.65, previous smoker: 0.63 vs. 0.60]. Furthermore, we considered clinical interpretation of risk-specific radiomics signatures. For hypertensive individuals and previous smokers, the surface area to volume ratio was smaller in the risk factor vs. healthy subjects; perhaps reflecting a pattern of global concentric hypertrophy in these conditions. In the diabetes subgroup, the most discriminatory radiomics feature was the median intensity of the myocardium at end-systole, which suggests a global alteration at the myocardial tissue level. This study confirms the feasibility and potential of CMR radiomics for deeper image phenotyping of cardiovascular health and disease. We demonstrate such analysis may have utility beyond conventional CMR metrics for improved detection and understanding of the early effects of cardiovascular risk factors on cardiac structure and tissue.

Aim: Left ventricular non-compaction (LVNC) is perceived as a rare high-risk cardiomyopathy characterized by excess left ventricular (LV) trabeculation. However, there is increasing evidence contesting the clinical significance of LV hyper-trabeculation and the existence of LVNC as a distinct cardiomyopathy. The aim of this study is to assess the association of LV trabeculation extent with cardiovascular morbidity and all-cause mortality in patients undergoing clinical cardiac magnetic resonance (CMR) scans across 57 European centers from the EuroCMR registry.Methods and Results: We studied 822 randomly selected cases from the EuroCMR registry. Image acquisition was according to international guidelines. We manually segmented images for LV chamber quantification and measurement of LV trabeculation (as per Petersen criteria). We report the association between LV trabeculation extent and important cardiovascular morbidities (stroke, atrial fibrillation, heart failure) and all-cause mortality prospectively recorded over 404 ± 82 days of follow-up. Maximal non-compaction to compaction ratio (NC/C) was mean (standard deviation) 1.81 ± 0.67, from these, 17% were above the threshold for hyper-trabeculation (NC/C > 2.3). LV trabeculation extent was not associated with increased risk of the defined outcomes (morbidities, mortality, LV CMR indices) in the whole cohort, or in sub-analyses of individuals without ischaemic heart disease, or those with NC/C > 2.3.Conclusion: Among 882 patients undergoing clinical CMR, excess LV trabeculation was not associated with a range of important cardiovascular morbidities or all-cause mortality over ~12 months of prospective follow-up. These findings suggest that LV hyper-trabeculation alone is not an indicator for worse cardiovascular prognosis.

Large population studies such as the UK Biobank provide great opportunities for understanding the pathophysiology, health impact and prognostic factors associated with COVID-19, a condition that has had significant impact on almost everyone around the world. We highlight the vast opportunities, challenges and limitations for research and collaboration from the UK Biobank and other large population studies in helping us better understand and manage both current and potential future pandemics.

Background: Cardiometabolic morbidity and medications, specifically Angiotensin Converting Enzyme inhibitors (ACEi) and Angiotensin Receptor Blockers (ARBs), have been linked with adverse outcomes from coronavirus disease 2019 (COVID-19). This study aims to investigate, factors associated with COVID-19 positivity in hospital for 1,436 UK Biobank participants; compared with individuals who tested negative, and with the untested, presumed negative, rest of the cohort.Methods: We studied 7,099 participants from the UK Biobank who had been tested for COVID-19 in hospital. We considered the following exposures: age, sex, ethnicity, body mass index (BMI), diabetes, hypertension, hypercholesterolaemia, ACEi/ARB use, prior myocardial infarction (MI), and smoking. We undertook comparisons between (1) COVID-19 positive and COVID-19 negative tested participants; and (2) COVID-19 tested positive and the remaining participants (tested negative plus untested, n = 494,838). Logistic regression models were used to investigate univariate and mutually adjusted associations.Results: Among participants tested for COVID-19, Black, Asian, and Minority ethnic (BAME) ethnicity, male sex, and higher BMI were independently associated with a positive result. BAME ethnicity, male sex, greater BMI, diabetes, hypertension, and smoking were independently associated with COVID-19 positivity compared to the remaining cohort (test negatives plus untested). However, similar associations were observed when comparing those who tested negative for COVID-19 with the untested cohort; suggesting that these factors associate with general hospitalization rather than specifically with COVID-19.Conclusions: Among participants tested for COVID-19 with presumed moderate to severe symptoms in a hospital setting, BAME ethnicity, male sex, and higher BMI are associated with a positive result. Other cardiometabolic morbidities confer increased risk of hospitalization, without specificity for COVID-19. ACE/ARB use did not associate with COVID-19 status.

Background: Convolutional neural network (CNN) based segmentation methods provide an efficient and automated way for clinicians to assess the structure and function of the heart in cardiac MR images. While CNNs can generally perform the segmentation tasks with high accuracy when training and test images come from the same domain (e.g., same scanner or site), their performance often degrades dramatically on images from different scanners or clinical sites.Methods: We propose a simple yet effective way for improving the network generalization ability by carefully designing data normalization and augmentation strategies to accommodate common scenarios in multi-site, multi-scanner clinical imaging data sets. We demonstrate that a neural network trained on a single-site single-scanner dataset from the UK Biobank can be successfully applied to segmenting cardiac MR images across different sites and different scanners without substantial loss of accuracy. Specifically, the method was trained on a large set of 3,975 subjects from the UK Biobank. It was then directly tested on 600 different subjects from the UK Biobank for intra-domain testing and two other sets for cross-domain testing: the ACDC dataset (100 subjects, 1 site, 2 scanners) and the BSCMR-AS dataset (599 subjects, 6 sites, 9 scanners).Results: The proposed method produces promising segmentation results on the UK Biobank test set which are comparable to previously reported values in the literature, while also performing well on cross-domain test sets, achieving a mean Dice metric of 0.90 for the left ventricle, 0.81 for the myocardium, and 0.82 for the right ventricle on the ACDC dataset; and 0.89 for the left ventricle, 0.83 for the myocardium on the BSCMR-AS dataset.Conclusions: The proposed method offers a potential solution to improve CNN-based model generalizability for the cross-scanner and cross-site cardiac MR image segmentation task.

The burden of pregnancy-related heart disease has dramatically increased over the last decades due to the increasing age at first pregnancy and higher prevalence of cardiovascular risk factors such as diabetes, hypertension, and obesity. Pregnancy is associated with physiological changes in the cardiovascular system, including hemodynamic, metabolic, and hormonal adaptations to meet the increased metabolic demands of the mother and fetus. It has been postulated that pregnancy may act as a cardiovascular stress test to identify women at high risk for heart disease, where the inability to adequately adapt to the physiologic stress of pregnancy may reveal the presence of genetic susceptibility to cardiovascular disease or accelerate the phenotypic expression of both inherited and acquired heart diseases, such as peripartum cardiomyopathy (PPCM). PPCM is a rare and incompletely understood clinical condition. Despite recent advances in the understanding of its pathogenesis, PPCM is not attributable to a well-defined pathological mechanism, and therefore, its diagnosis still relies on the exclusion of overlapping dilated phenotypes. Cardiac imaging plays a key role in any peripartum woman with signs and symptoms of heart failure in establishing the diagnosis, ruling out life-threatening complications, guiding therapy and conveying prognostic information. Echocardiography represents the first-line imaging technique, given its robust diagnostic yield and its favorable cost-effectiveness. Cardiovascular magnetic resonance is a biologically safe high-throughput modality that allows accurate morpho-functional assessment of the cardiovascular system in addition to the unique asset of myocardial tissue characterization as a pivotal piece of information in the pathophysiological puzzle of PPCM. In this review, we will highlight current evidence on the role of multimodality imaging in the differential diagnosis, prognostic assessment, and understanding of the pathophysiological basis of PPCM.

Cardiac imaging plays an important role in the diagnosis of cardiovascular disease (CVD). Until now, its role has been limited to visual and quantitative assessment of cardiac structure and function. However, with the advent of big data and machine learning, new opportunities are emerging to build artificial intelligence tools that will directly assist the clinician in the diagnosis of CVDs. This paper presents a thorough review of recent works in this field and provide the reader with a detailed presentation of the machine learning methods that can be further exploited to enable more automated, precise and early diagnosis of most CVDs.

Abstract Cardiovascular disease remains a leading cause of morbidity and mortality globally. Changing natural history of the disease due to improved care of acute conditions and ageing population necessitates new strategies to tackle conditions which have more chronic and indolent course. These include an increased deployment of safe screening methods, life-long surveillance, and monitoring of both disease activity and tailored-treatment, by way of increasingly personalized medical care. Cardiovascular magnetic resonance (CMR) is a non-invasive, ionising radiation-free method, which can support a significant number of clinically relevant measurements and offers new opportunities to advance the state of art of diagnosis, prognosis and treatment. The objective of the SCMR Clinical Trial Taskforce was to summarizes the evidence to emphasize where currently CMR-guided clinical care can indeed translate into meaningful use and efficient deployment of resources results in meaningful and efficient use. The objective of the present initiative was to provide an appraisal of evidence on analytical validation, including the accuracy and precision, and clinical qualification of parameters in disease context, clarifying the strengths and weaknesses of the state of art, as well as the gaps in the current evidence This paper is complementary to the existing position papers on standardized acquisition and post-processing ensuring robustness and transferability for widespread use. Themed imaging-endpoint guidance on trial design to support drug-discovery or change in clinical practice (part II), will be presented in a follow-up paper in due course. As CMR continues to undergo rapid development, regular updates of the present recommendations are foreseen.

Abstract Background Cardiovascular resonance (CMR) imaging is a standard imaging modality for assessing cardiovascular diseases (CVDs), the leading cause of death globally. CMR enables accurate quantification of the cardiac chamber volume, ejection fraction and myocardial mass, providing information for diagnosis and monitoring of CVDs. However, for years, clinicians have been relying on manual approaches for CMR image analysis, which is time consuming and prone to subjective errors. It is a major clinical challenge to automatically derive quantitative and clinically relevant information from CMR images. Methods Deep neural networks have shown a great potential in image pattern recognition and segmentation for a variety of tasks. Here we demonstrate an automated analysis method for CMR images, which is based on a fully convolutional network (FCN). The network is trained and evaluated on a large-scale dataset from the UK Biobank, consisting of 4,875 subjects with 93,500 pixelwise annotated images. The performance of the method has been evaluated using a number of technical metrics, including the Dice metric, mean contour distance and Hausdorff distance, as well as clinically relevant measures, including left ventricle (LV) end-diastolic volume (LVEDV) and end-systolic volume (LVESV), LV mass (LVM); right ventricle (RV) end-diastolic volume (RVEDV) and end-systolic volume (RVESV). Results By combining FCN with a large-scale annotated dataset, the proposed automated method achieves a high performance in segmenting the LV and RV on short-axis CMR images and the left atrium (LA) and right atrium (RA) on long-axis CMR images. On a short-axis image test set of 600 subjects, it achieves an average Dice metric of 0.94 for the LV cavity, 0.88 for the LV myocardium and 0.90 for the RV cavity. The mean absolute difference between automated measurement and manual measurement is 6.1 mL for LVEDV, 5.3 mL for LVESV, 6.9 gram for LVM, 8.5 mL for RVEDV and 7.2 mL for RVESV. On long-axis image test sets, the average Dice metric is 0.93 for the LA cavity (2-chamber view), 0.95 for the LA cavity (4-chamber view) and 0.96 for the RA cavity (4-chamber view). The performance is comparable to human inter-observer variability. Conclusions We show that an automated method achieves a performance on par with human experts in analysing CMR images and deriving clinically relevant measures.

Artificial intelligence (AI) using machine learning techniques will change healthcare as we know it. While healthcare AI applications are currently trailing behind popular AI applications, such as personalized web-based advertising, the pace of research and deployment is picking up and about to become disruptive. Overcoming challenges such as patient and public support, transparency over the legal basis for healthcare data use, privacy preservation, technical challenges related to accessing large-scale data from healthcare systems not designed for Big Data analysis, and deployment of AI in routine clinical practice will be crucial. Cardiac imaging and imaging of other body parts is likely to be at the frontier for the development of applications as pattern recognition and machine learning are a significant strength of AI with practical links to image processing. Many opportunities in cardiac imaging exist where AI will impact patients, medical staff, hospitals, commissioners and thus, the entire healthcare system. This perspective article will outline our vision for AI in cardiac imaging with examples of potential applications, challenges and some lessons learnt in recent years.

Objectives: To assess whether electronic (e-) coaching, using personalized web-based lifestyle and risk factor counselling with additional email prompts, provides additional risk reduction when added to standard of care (SOC) in individuals at increased risk. Methods: Between June 2013 and May 2015, 402 participants were allocated 1:1 to e-coaching and SOC versus SOC. Participants free of manifest cardiovascular disease, with internet access, and a 10-year QRISK2 cardiovascular risk of ≥10% were enrolled. Change in oscillometric carotid-femoral pulse wave velocity (PWV) from baseline to six months was the primary endpoint. Secondary outcomes included change in blood pressure (BP), weight, and risk scores. Analysis was by intention to treat. Results: Mean (±SD) age was 65.5 (5.6) years with 37% females. Primary outcome data were available for 94%. There was no difference in PWV reductions between e-coaching and standard of care groups (–0.16 m/s vs. –0.25 m/s, 95% confidence interval –0.39 to 0.22, p = 0.56). There were no differences in the improvement between groups for BP, weight, Framingham, or QRISK2 scores. Pulse wave velocity change was more favorable in those with a higher level of education (p = 0.04), but was not associated with age, gender, presence of diabetes, baseline QRISK2 score, or logins to the website. Conclusions: In individuals at increased cardiovascular risk, a comprehensive ‘health check’ program modestly reduced future risk. Personalized e-coaching did not provide added risk reduction. Currently there is no evidence to routinely recommend e-coaching in cardiovascular health check programs. Trial registration: HAPPY London ClinicalTrials.gov: NCT01911910

Background and Objectives: Obstructive sleep apnea (OSA) is a common disorder with an increased risk for left ventricular and right ventricular dysfunction. Most studies to date have examined populations with manifest cardiovascular disease using echocardiography to analyze ventricular dysfunction with little or no reference to ventricular volumes or myocardial mass. Our aim was to explore these parameters with cardiac MRI. We hypothesized that there would be stepwise increase in left ventricular mass and right ventricular volumes from the unaffected, to the snoring and the OSA group. Materials and Methods: We analyzed cardiac MRI data from 4978 UK Biobank participants free from cardiovascular disease. Participants were allocated into three cohorts: with OSA, with self-reported snoring and without OSA or snoring (n = 118, 1886 and 2477). We analyzed cardiac parameters from balanced cine-SSFP sequences and indexed them to body surface area. Results: Patients with OSA were mostly males (47.3% vs. 79.7%; p < 0.001) with higher body mass index (25.7 ± 4.0 vs. 31.3 ± 5.3 kg/m²; p < 0.001) and higher blood pressure (135 ± 18 vs. 140 ± 17 mmHg; p = 0.012) compared to individuals without OSA or snoring. Regression analysis showed a significant effect for OSA in left ventricular end-diastolic index (LVEDVI) (β = −4.9 ± 2.4 mL/m²; p = 0.040) and right ventricular end-diastolic index (RVEDVI) (β = −6.2 ± 2.6 mL/m²; p = 0.016) in females and for right ventricular ejection fraction (RVEF) (β = 1.7 ± 0.8%; p = 0.031) in males. A significant effect was discovered in snoring females for left ventricular mass index (LVMI) (β = 3.5 ± 0.9 g/m²; p < 0.001) and in males for left ventricular ejection fraction (LVEF) (β = 1.0 ± 0.3%; p = 0.001) and RVEF (β = 1.2 ± 0.3%; p < 0.001). Conclusion: Our study suggests that OSA is highly underdiagnosed and that it is an evolving process with gender specific progression. Females with OSA show significantly lower ventricular volumes while males with snoring show increased ejection fractions which may be an early sign of hypertrophy. Separate prospective studies are needed to further explore the direction of causality.