Publications
131 results found
Biffi C, Cerrolaza JJ, Tarroni G, et al., 2019, 3D high-resolution cardiac segmentation reconstruction from 2D views using conditional variational autoencoders, 16th IEEE International Symposium on Biomedical Imaging (ISBI), Publisher: IEEE, Pages: 1643-1646, ISSN: 1945-7928
Accurate segmentation of heart structures imaged by cardiac MR is key for the quantitative analysis of pathology. High-resolution 3D MR sequences enable whole-heart structural imaging but are time-consuming, expensive to acquire and they often require long breath holds that are not suitable for patients. Consequently, multiplanar breath-hold 2D cines sequences are standard practice but are disadvantaged by lack of whole-heart coverage and low through-plane resolution. To address this, we propose a conditional variational autoencoder architecture able to learn a generative model of 3D high-resolution left ventricular (LV) segmentations which is conditioned on three 2D LV segmentations of one short-axis and two long-axis images. By only employing these three 2D segmentations, our model can efficiently reconstruct the 3D high-resolution LV segmentation of a subject. When evaluated on 400 unseen healthy volunteers, our model yielded an average Dice score of 87.92 ± 0.15 and outperformed competing architectures (TL-net, Dice score = 82.60 ± 0.23, p = 2.2 · 10 -16 ).
Garcia-Pavia P, Kim Y, Restrepo-Cordoba MA, et al., 2019, Genetic variants associated with cancer therapy-induced cardiomyopathy, Circulation, Vol: 140, Pages: 31-41, ISSN: 0009-7322
BackgroundCancer therapy-induced cardiomyopathy (CCM) is associated with cumulative drug exposures and pre-existing cardiovascular disorders. These parametersincompletely account for substantial inter-individual susceptibility to CCM. We hypothesized that rare variants in cardiomyopathy genes contribute to CCM.MethodsWe studied 213 CCM patients from three cohorts: retrospectively recruited adults with diverse cancers (n=99), prospectively phenotyped breast cancer adults (n=73) and prospectively phenotyped children with acute myeloid leukemia (n=41). Cardiomyopathy genes, including nine pre-specified genes were sequenced. The prevalence of rare variants was compared between CCM cohorts and The Cancer Genome Atlas (TCGA) participants(n=2053), healthy volunteers(n=445), and ancestry-matchedreference population. Clinical characteristics and outcomes were assessed, stratified by genotypes. A prevalent CCM genotype was modeled in anthracycline-treated mice.ResultsCCM was diagnosed 0.4-9 years after chemotherapy; 90% of these patients received anthracyclines. Adult CCM patients had cardiovascular risk factors similar to the U.S. population. Among nine prioritized genes CCM patients had more rare protein-altering variants than comparative cohorts (p≤1.98e-04). Titin-truncating variants (TTNtv) predominated, occurring in 7.5% CCM patients versus 1.1% TCGA participants (p=7.36e-08), 0.7% healthy volunteers (p=3.42e-06), and 0.6% reference population (p=5.87e-14). Adult CCM patients with TTNtv experienced more heart failure and atrial fibrillation (p=0.003)and impaired myocardial recovery (p=0.03) than those without.Consistent with human data, anthracycline-treated TTNtv mice and isolated TTNtv cardiomyocytes showed sustained contractile dysfunction unlike wildtype (p=0.0004 and p<0.002, respectively).ConclusionsUnrecognized rare variants in cardiomyopathy-associated genes, particularly TTNtv, increased the risk for CCM in children and adults, and adverse cardiac events
Pillinger T, Osimo EF, de Marvao A, et al., 2019, Cardiac structure and function in patients with schizophrenia taking antipsychotic drugs: an MRI study, Translational Psychiatry, Vol: 9, ISSN: 2158-3188
Cardiovascular disease (CVD) is a major cause of excess mortality in schizophrenia. Preclinical evidence shows antipsychotics can cause myocardial fibrosis and myocardial inflammation in murine models, but it is not known if this is the case in patients. We therefore set out to determine if there is evidence of cardiac fibrosis and/or inflammation using cardiac MRI in medicated patients with schizophrenia compared with matched healthy controls. Thirty-one participants (14 patients and 17 controls) underwent cardiac MRI assessing myocardial markers of fibrosis/inflammation, indexed by native myocardial T1 time, and cardiac structure (left ventricular (LV) mass) and function (left/right ventricular end-diastolic and end-systolic volumes, stroke volumes, and ejection fractions). Participants were physically fit, and matched for age, gender, smoking, blood pressure, BMI, HbA1c, ethnicity, and physical activity. Compared with controls, native myocardial T1 was significantly longer in patients with schizophrenia (effect size, d = 0.89; p = 0.02). Patients had significantly lower LV mass, and lower left/right ventricular end-diastolic and stroke volumes (effect sizes, d = 0.86-1.08; all p-values < 0.05). There were no significant differences in left/right end-systolic volumes and ejection fractions between groups (p > 0.05). These results suggest an early diffuse fibro-inflammatory myocardial process in patients that is independent of established CVD-risk factors and could contribute to the excess cardiovascular mortality associated with schizophrenia. Future studies are required to determine if this is due to antipsychotic treatment or is intrinsic to schizophrenia.
Attard M, Dawes T, Simoes Monteiro de Marvao A, et al., 2019, Metabolic pathways associated with right ventricular adaptation to pulmonary hypertension: Three dimensional analysis of cardiac magnetic resonance imaging, EHJ Cardiovascular Imaging / European Heart Journal - Cardiovascular Imaging, Vol: 20, Pages: 668-676, ISSN: 2047-2412
AimsWe sought to identify metabolic pathways associated with right ventricular (RV) adaptation to pulmonary hypertension (PH). We evaluated candidate metabolites, previously associated with survival in pulmonary arterial hypertension, and used automated image segmentation and parametric mapping to model their relationship to adverse patterns of remodelling and wall stress.Methods and resultsIn 312 PH subjects (47.1% female, mean age 60.8 ± 15.9 years), of which 182 (50.5% female, mean age 58.6 ± 16.8 years) had metabolomics, we modelled the relationship between the RV phenotype, haemodynamic state, and metabolite levels. Atlas-based segmentation and co-registration of cardiac magnetic resonance imaging was used to create a quantitative 3D model of RV geometry and function—including maps of regional wall stress. Increasing mean pulmonary artery pressure was associated with hypertrophy of the basal free wall (β = 0.29) and reduced relative wall thickness (β = −0.38), indicative of eccentric remodelling. Wall stress was an independent predictor of all-cause mortality (hazard ratio = 1.27, P = 0.04). Six metabolites were significantly associated with elevated wall stress (β = 0.28–0.34) including increased levels of tRNA-specific modified nucleosides and fatty acid acylcarnitines, and decreased levels (β = −0.40) of sulfated androgen.ConclusionUsing computational image phenotyping, we identify metabolic profiles, reporting on energy metabolism and cellular stress-response, which are associated with adaptive RV mechanisms to PH.
Mazzarotto F, Tayal P, Buchan R, et al., 2019, RE-EVALUATING THE GENETIC CONTRIBUTION OF MONOGENIC DILATED CARDIOMYOPATHY, Annual Conference of the British-Cardiovascular-Society (BCS) - Digital Health Revolution, Publisher: BMJ PUBLISHING GROUP, Pages: A100-A100, ISSN: 1355-6037
Tarroni G, Oktay O, Bai W, et al., 2019, Learning-based quality control for cardiac MR images, IEEE Transactions on Medical Imaging, Vol: 38, Pages: 1127-1138, ISSN: 0278-0062
The effectiveness of a cardiovascular magnetic resonance (CMR) scan depends on the ability of the operator to correctly tune the acquisition parameters to the subject being scanned and on the potential occurrence of imaging artefacts such as cardiac and respiratory motion. In the clinical practice, a quality control step is performed by visual assessment of the acquired images: however, this procedure is strongly operatordependent, cumbersome and sometimes incompatible with the time constraints in clinical settings and large-scale studies. We propose a fast, fully-automated, learning-based quality control pipeline for CMR images, specifically for short-axis image stacks. Our pipeline performs three important quality checks: 1) heart coverage estimation, 2) inter-slice motion detection, 3) image contrast estimation in the cardiac region. The pipeline uses a hybrid decision forest method - integrating both regression and structured classification models - to extract landmarks as well as probabilistic segmentation maps from both long- and short-axis images as a basis to perform the quality checks. The technique was tested on up to 3000 cases from the UK Biobank as well as on 100 cases from the UK Digital Heart Project, and validated against manual annotations and visual inspections performed by expert interpreters. The results show the capability of the proposed pipeline to correctly detect incomplete or corrupted scans (e.g. on UK Biobank, sensitivity and specificity respectively 88% and 99% for heart coverage estimation, 85% and 95% for motion detection), allowing their exclusion from the analysed dataset or the triggering of a new acquisition.
Pillinger T, Osimo EF, de Marvao A, et al., 2019, Cardiac structure and function in patients with schizophrenia taking antipsychotic drugs: an MRI study
<jats:title>Abstract</jats:title><jats:p>Cardiovascular disease (CVD) is a major cause of excess mortality in schizophrenia. Preclinical evidence shows antipsychotics can cause myocardial fibrosis and myocardial inflammation in murine models, but it is not known if this is the case in patients. We therefore set out to determine if there is evidence of cardiac fibrosis and/or inflammation using cardiac MRI in medicated patients with schizophrenia compared with matched healthy controls. 31 participants (14 patients and 17 controls) underwent cardiac MRI assessing myocardial markers of fibrosis/inflammation, indexed by native myocardial T1 time, and cardiac structure (left ventricular (LV) mass) and function (left/right ventricular end-diastolic and end-systolic volumes, stroke volumes, and ejection fractions). Participants were physically fit, and matched for age, gender, smoking, blood pressure, BMI, HbA1c, ethnicity, and physical activity. Compared with controls, native myocardial T1 was significantly longer in patients with schizophrenia (effect size, d=0.89; p=0.02). Patients had significantly lower LV mass, and lower left/right ventricular end-diastolic and stroke volumes (effect sizes, d=0.86-1.08; all p-values <0.05). There were no significant differences in left/right end-systolic volumes and ejection fractions between groups (p>0.05). These results suggest an early diffuse fibro-inflammatory myocardial process in patients that is independent of established CVD-risk factors and could contribute to the excess cardiovascular mortality associated with schizophrenia. Future studies are required to determine if this is due to antipsychotic treatment or is intrinsic to schizophrenia.</jats:p>
Meyer HV, Dawes TJW, Serrani M, et al., 2019, Genomic analysis reveals a functional role for myocardial trabeculae in adults
<jats:title>ABSTRACT</jats:title><jats:p>Since being first described by Leonardo da Vinci in 1513 it has remained an enigma why the endocardial surfaces of the adult heart retain a complex network of muscular trabeculae – with their persistence thought to be a vestige of embryonic development. For causative physiological inference we harness population genomics, image-based intermediate phenotyping and <jats:italic>in silico</jats:italic> modelling to determine the effect of this complex cardiovascular trait on function. Using deep learning-based image analysis we identified genetic associations with trabecular complexity in 18,097 UK Biobank participants which were replicated in an independently measured cohort of 1,129 healthy adults. Genes in these associated regions are enriched for expression in the fetal heart or vasculature and implicate loci associated with haemodynamic phenotypes and developmental pathways. A causal relationship between increasing trabecular complexity and both ventricular performance and electrical activity are supported by complementary biomechanical simulations and Mendelian randomisation studies. These findings show that myocardial trabeculae are a previously-unrecognised determinant of cardiovascular physiology in adult humans.</jats:p>
Bello G, Dawes T, Duan J, et al., 2019, Deep learning cardiac motion analysis for human survival prediction, Nature Machine Intelligence, Vol: 1, Pages: 95-104, ISSN: 2522-5839
Motion analysis is used in computer vision to understand the behaviour of moving objects in sequences of images. Optimizing the interpretation of dynamic biological systems requires accurate and precise motion tracking as well as efficient representations of high-dimensional motion trajectories so that these can be used for prediction tasks. Here we use image sequences of the heart, acquired using cardiac magnetic resonance imaging, to create time-resolved three-dimensional segmentations using a fully convolutional network trained on anatomical shape priors. This dense motion model formed the input to a supervised denoising autoencoder (4Dsurvival), which is a hybrid network consisting of an autoencoder that learns a task-specific latent code representation trained on observed outcome data, yielding a latent representation optimized for survival prediction. To handle right-censored survival outcomes, our network used a Cox partial likelihood loss function. In a study of 302 patients, the predictive accuracy (quantified by Harrell’s C-index) was significantly higher (P = 0.0012) for our model C = 0.75 (95% CI: 0.70–0.79) than the human benchmark of C = 0.59 (95% CI: 0.53–0.65). This work demonstrates how a complex computer vision task using high-dimensional medical image data can efficiently predict human survival.
Walsh R, Mazzarotto F, Whiffin N, et al., 2019, Quantitative approaches to variant classification increase the yield and precision of genetic testing in Mendelian diseases: The case of hypertrophic cardiomyopathy, Genome Medicine, Vol: 11, ISSN: 1756-994X
BackgroundInternational guidelines for variant interpretation in Mendelian disease set stringent criteria to report a variant as (likely) pathogenic, prioritising control of false-positive rate over test sensitivity and diagnostic yield. Genetic testing is also more likely informative in individuals with well-characterised variants from extensively studied European-ancestry populations. Inherited cardiomyopathies are relatively common Mendelian diseases that allow empirical calibration and assessment of this framework.MethodsWe compared rare variants in large hypertrophic cardiomyopathy (HCM) cohorts (up to 6179 cases) to reference populations to identify variant classes with high prior likelihoods of pathogenicity, as defined by etiological fraction (EF). We analysed the distribution of variants using a bespoke unsupervised clustering algorithm to identify gene regions in which variants are significantly clustered in cases.ResultsAnalysis of variant distribution identified regions in which variants are significantly enriched in cases and variant location was a better discriminator of pathogenicity than generic computational functional prediction algorithms. Non-truncating variant classes with an EF ≥ 0.95 were identified in five established HCM genes. Applying this approach leads to an estimated 14–20% increase in cases with actionable HCM variants, i.e. variants classified as pathogenic/likely pathogenic that might be used for predictive testing in probands’ relatives.ConclusionsWhen found in a patient confirmed to have disease, novel variants in some genes and regions are empirically shown to have a sufficiently high probability of pathogenicity to support a “likely pathogenic” classification, even without additional segregation or functional data. This could increase the yield of high confidence actionable variants, consistent with the framework and recommendations of current guidelines. The techniques outlined offer a consisten
Duan J, Schlemper J, Bai W, et al., 2018, Combining deep learning and shape priors for bi-ventricular segmentation of volumetric cardiac magnetic resonance images, MICCAI ShapeMI Workshop, Publisher: Springer Verlag, Pages: 258-267, ISSN: 0302-9743
In this paper, we combine a network-based method with image registration to develop a shape-based bi-ventricular segmentation tool for short-axis cardiac magnetic resonance (CMR) volumetric images. The method first employs a fully convolutional network (FCN) to learn the segmentation task from manually labelled ground truth CMR volumes. However, due to the presence of image artefacts in the training dataset, the resulting FCN segmentation results are often imperfect. As such, we propose a second step to refine the FCN segmentation. This step involves performing a non-rigid registration with multiple high-resolution bi-ventricular atlases, allowing the explicit shape priors to be inferred. We validate the proposed approach on 1831 healthy subjects and 200 subjects with pulmonary hypertension. Numerical experiments on the two datasets demonstrate that our approach is capable of producing accurate, high-resolution and anatomically smooth bi-ventricular models, despite the artefacts in the input CMR volumes.
Dawes T, Simoes Monteiro de Marvao A, Shi W, et al., 2018, Identifying the optimal regional predictor of right ventricular global function: a high resolution 3D cardiac magnetic resonance study, Anaesthesia, Vol: 74, Pages: 312-320, ISSN: 0003-2409
Right ventricular (RV) function has prognostic value in acute, chronic and peri‐operative disease, although the complex RV contractile pattern makes rapid assessment difficult. Several two‐dimensional (2D) regional measures estimate RV function, however the optimal measure is not known. High‐resolution three‐dimensional (3D) cardiac magnetic resonance cine imaging was acquired in 300 healthy volunteers and a computational model of RV motion created. Points where regional function was significantly associated with global function were identified and a 2D, optimised single‐point marker (SPM‐O) of global function developed. This marker was prospectively compared with tricuspid annular plane systolic excursion (TAPSE), septum‐freewall displacement (SFD) and their fractional change (TAPSE‐F, SFD‐F) in a test cohort of 300 patients in the prediction of RV ejection fraction. RV ejection fraction was significantly associated with systolic function in a contiguous 7.3 cm2 patch of the basal RV freewall combining transverse (38%), longitudinal (35%) and circumferential (27%) contraction and coinciding with the four‐chamber view. In the test cohort, all single‐point surrogates correlated with RV ejection fraction (p < 0.010), but correlation (R) was higher for SPM‐O (R = 0.44, p < 0.001) than TAPSE (R = 0.24, p < 0.001) and SFD (R = 0.22, p < 0.001), and non‐significantly higher than TAPSE‐F (R = 0.40, p < 0.001) and SFD‐F (R = 0.43, p < 0.001). SPM‐O explained more of the observed variance in RV ejection fraction (19%) and predicted it more accurately than any other 2D marker (median error 2.8 ml vs 3.6 ml, p < 0.001). We conclude that systolic motion of the basal RV freewall predicts global function more accurately than other 2D estimators. However, no markers summarise 3D contractile patterns, limiting their predictive accuracy.
Tarroni G, Oktay O, Sinclair M, et al., 2018, A comprehensive approach for learning-based fully-automated inter-slice motion correction for short-axis cine cardiac MR image stacks, 21st International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) / 8th Eurographics Workshop on Visual Computing for Biology and Medicine (VCBM), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 268-276, ISSN: 0302-9743
In the clinical routine, short axis (SA) cine cardiac MR (CMR) image stacks are acquired during multiple subsequent breath-holds. If the patient cannot consistently hold the breath at the same position, the acquired image stack will be affected by inter-slice respiratory motion and will not correctly represent the cardiac volume, introducing potential errors in the following analyses and visualisations. We propose an approach to automatically correct inter-slice respiratory motion in SA CMR image stacks. Our approach makes use of probabilistic segmentation maps (PSMs) of the left ventricular (LV) cavity generated with decision forests. PSMs are generated for each slice of the SA stack and rigidly registered in-plane to a target PSM. If long axis (LA) images are available, PSMs are generated for them and combined to create the target PSM; if not, the target PSM is produced from the same stack using a 3D model trained from motion-free stacks. The proposed approach was tested on a dataset of SA stacks acquired from 24 healthy subjects (for which anatomical 3D cardiac images were also available as reference) and compared to two techniques which use LA intensity images and LA segmentations as targets, respectively. The results show the accuracy and robustness of the proposed approach in motion compensation.
Biffi C, Oktay O, Tarroni G, et al., 2018, Learning interpretable anatomical features through deep generative models: Application to cardiac remodeling, International Conference On Medical Image Computing & Computer Assisted Intervention, Publisher: Springer, Pages: 464-471, ISSN: 0302-9743
Alterations in the geometry and function of the heart define well-established causes of cardiovascular disease. However, current approaches to the diagnosis of cardiovascular diseases often rely on subjective human assessment as well as manual analysis of medical images. Both factors limit the sensitivity in quantifying complex structural and functional phenotypes. Deep learning approaches have recently achieved success for tasks such as classification or segmentation of medical images, but lack interpretability in the feature extraction and decision processes, limiting their value in clinical diagnosis. In this work, we propose a 3D convolutional generative model for automatic classification of images from patients with cardiac diseases associated with structural remodeling. The model leverages interpretable task-specific anatomic patterns learned from 3D segmentations. It further allows to visualise and quantify the learned pathology-specific remodeling patterns in the original input space of the images. This approach yields high accuracy in the categorization of healthy and hypertrophic cardiomyopathy subjects when tested on unseen MR images from our own multi-centre dataset (100%) as well on the ACDC MICCAI 2017 dataset (90%). We believe that the proposed deep learning approach is a promising step towards the development of interpretable classifiers for the medical imaging domain, which may help clinicians to improve diagnostic accuracy and enhance patient risk-stratification.
Duan J, Schlemper J, Bai W, et al., 2018, Deep Nested Level Sets: Fully Automated Segmentation of Cardiac MR Images in Patients with Pulmonary Hypertension, International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Pages: 595-603, ISSN: 0302-9743
Bhuva A, Treibel TA, De Marvao A, et al., 2018, Septal hypertrophy in aortic stenosis and its regression after valve replacement is more plastic in males than females: insights from 3D machine learning approach, European-Society-of-Cardiology Congress, Publisher: European Society of Cardiology, Pages: 1132-1132, ISSN: 0195-668X
Background: Evaluation of left ventricular non-compaction (LVNC) is an increasingly common indication for cardiac magnetic resonance imaging (MRI). Fractal dimension (FD) is a unitless measure of geometrical complexity which can be used to quantify LV trabeculation. FD is increased in LVNC, but there have been few studies on FD in normal subjects. The aim of the study was to establish reference ranges for FD in a healthy population, and identify covariates which are associated with FD.Methods: MRI was performed in 1,913 volunteers without hypertension, diabetes, or heart disease (1055 female, 858 male; median age 40, range 19-82). FD was derived from LV short-axis images, using a custom MATLAB box-counting algorithm. The maximal FD in the apical half of the LV was used for all analyses, as previously described.Results: Normal ranges (2.5-97.5th percentile) for female and male subjects were 1.154 - 1.367 and 1.179 - 1.392, respectively. FD was significantly correlated with age, gender, ethnicity, body surface area (BSA), activity score, and systolic blood pressure. In multivariable analysis, FD was independently correlated with increased age (β 0.11, p<0.001), male gender (β 0.09, p<0.001), African/Afro-Caribbean ethnicity (β 0.18, p<0.001), increased BSA (β 0.27, p<0.001), and increased activity score (β 0.07, p=0.002). Since ethnicity was found to significantly affect FD, normal ranges were calculated for each subgroup (see table).Conclusions: This is the largest study on FD in healthy subjects, and the first to present gender- and race-specific normal ranges. The association between FD and age suggests that LV trabeculation is a dynamic phenotype which may change with age.
Dawes T, Cai J, Quinlan M, et al., 2018, Fractal analysis of right ventricular trabeculae in pulmonary hypertension, Radiology, Vol: 288, Pages: 386-395, ISSN: 0033-8419
Purpose: To measure right ventricular (RV) trabecular complexity by its fractal dimension (FD) in healthy subjects and patients with pulmonary hypertension (PH) and assess its relationship to hemodynamic and functional parameters, and future cardiovascular events. Materials and methods: This retrospective study used data acquired from May 2004 to October 2013 for 256 patients with newly-diagnosed PH that underwent cardiac magnetic resonance (CMR) imaging, right heart catheterization and six-minute walk distance testing with a median follow-up of 4.0 years. 256 healthy controls underwent CMR only. Biventricular FD, volumes and function were assessed on short-axis cine images. Reproducibility was assessed by intraclass correlation coefficient, correlation between variables was assessed by Pearson’s correlation test, and mortality prediction compared by univariable and multivariable Cox regression analysis. Results: RV-FD reproducibility had an intraclass correlation coefficient of 0.97 (95% confidence interval [CI]: 0.96, 0.98).RV-FD was higher in PH patients than healthy subjects (median 1.310, inter-quartile range [IQR] 1.281-1.341 vs 1.264, 1.242-1.295, P<.001) with the greatest difference near the apex. RV-FD was associated with pulmonary vascular resistance (r=0.30, P<.001). In univariable Cox regression analysis, RV-FD was a significant predictor of death (hazards ratio [HR]: 1.256, CI: 1.011, 1.560, P=.04), but in a multivariable analysis did not predict survival independently of conventional parameters of RV remodeling (HR: 1.179, CI: 0.871, 1.596, P=0.29). Conclusion: Fractal analysis of RV trabecular complexity is a highly reproducible measure of remodeling in PH associated with afterload, although the gain in survival prediction over traditional markers is not significant.
Walsh R, Mazzarotto F, Whiffin N, et al., 2018, Quantitative approaches to variant classification increase the yield and precision of genetic testing in Mendelian diseases: The case of hypertrophic cardiomyopathy, Publisher: bioRxiv
ABSTRACT Background International guidelines for variant interpretation in Mendelian disease set stringent criteria to report a variant as (likely) pathogenic, prioritising control of false positive rate over test sensitivity and diagnostic yield. Genetic testing is also more likely informative in individuals with well-characterised variants from extensively studied European-ancestry populations. Inherited cardiomyopathies are relatively common Mendelian diseases that allow empirical calibration and assessment of this framework. Results We compared rare variants in large hypertrophic cardiomyopathy (HCM) cohorts to reference populations to identify variant classes with high prior likelihoods of pathogenicity, as defined by etiological fraction (EF). Analysis of variant distribution identified regions in which variants are significantly enriched in cases and variant location was a better discriminator of pathogenicity than generic computational functional prediction algorithms. Non-truncating variant classes with an EF≥0.95, and therefore clinically actionable, were identified in 5 established HCM genes. Applying this approach leads to an estimated 14-20% increase in cases with actionable HCM variants. Conclusions When found in a patient confirmed to have disease, novel variants in some genes and regions are empirically shown to have a sufficiently high probability of pathogenicity to support a “likely pathogenic” classification, even without additional segregation or functional data. This could increase the yield of high confidence actionable variants, consistent with the framework and recommendations of current guidelines. The techniques outlined offer a consistent, unbiased and equitable approach to variant interpretation for Mendelian disease genetic testing. We propose adaptations to ACMG/AMP guidelines to incorporate such evidence in a quantitative and transparent manner.
Dawes TJW, Serrani M, Bai W, et al., 2018, Myocardial trabeculae improve left ventricular function: a combined UK Biobank and computational analysis, GAT Annual Scientific Meeting 2018, Publisher: Association of Anaesthetists of Great Britain and Ireland
De Marvao A, Hadjiphilippou S, Escudier A, et al., 2018, A rare presentation of heart failure, World Congress on Acute Heart Failure, Publisher: Oxford University Press (OUP), Pages: 17-18, ISSN: 1388-9842
de Marvao A, Biffi C, Walsh R, et al., 2018, Defining The Effects Of Genetic Variation Using Machine Learning Analysis Of CMRs: A Study In Hypertrophic Cardiomyopathy And In A Healthy Population, Joint Meeting of the British-Society-of-Cardiovascular-Imaging/British-Society-of-Cardiovascular-CT, British-Society-of-Cardiovascular-Magnetic-Resonance and British-Nuclear-Cardiac-Society on British Cardiovascular Imaging, Publisher: BMJ PUBLISHING GROUP, Pages: A7-A8, ISSN: 1355-6037
Ostrowski PJ, de Marvao A, Quinlan M, et al., 2018, Fractal dimension as an index of left ventricular trabeculation: normal values in healthy subjects, and association with age, Quality of Care and Outcomes Research Scientific Sessions, Publisher: Lippincott, Williams & Wilkins, Pages: 1-8, ISSN: 0009-7322
Background: Evaluation of left ventricular non-compaction (LVNC) is an increasingly common indication for cardiac magnetic resonance imaging (MRI). Fractal dimension (FD) is a unitless measure of geometrical complexity which can be used to quantify LV trabeculation. FD is increased in LVNC, but there have been few studies on FD in normal subjects. The aim of the study was to establish reference ranges for FD in a healthy population, and identify covariates which are associated with FD.Methods: MRI was performed in 1,913 volunteers without hypertension, diabetes, or heart disease (1055 female, 858 male; median age 40, range 19-82). FD was derived from LV short-axis images, using a custom MATLAB box-counting algorithm. The maximal FD in the apical half of the LV was used for all analyses, as previously described.Results: Normal ranges (2.5-97.5th percentile) for female and male subjects were 1.154 - 1.367 and 1.179 - 1.392, respectively. FD was significantly correlated with age, gender, ethnicity, body surface area (BSA), activity score, and systolic blood pressure. In multivariable analysis, FD was independently correlated with increased age (β 0.11, p<0.001), male gender (β 0.09, p<0.001), African/Afro-Caribbean ethnicity (β 0.18, p<0.001), increased BSA (β 0.27, p<0.001), and increased activity score (β 0.07, p=0.002). Since ethnicity was found to significantly affect FD, normal ranges were calculated for each subgroup (see table).Conclusions: This is the largest study on FD in healthy subjects, and the first to present gender- and race-specific normal ranges. The association between FD and age suggests that LV trabeculation is a dynamic phenotype which may change with age.
Oktay O, Ferrante E, Kamnitsas K, et al., 2018, Anatomically Constrained Neural Networks (ACNN): application to cardiac image enhancement and segmentation, IEEE Transactions on Medical Imaging, Vol: 37, Pages: 384-395, ISSN: 0278-0062
Incorporation of prior knowledge about organ shape and location is key to improve performance of image analysis approaches. In particular, priors can be useful in cases where images are corrupted and contain artefacts due to limitations in image acquisition. The highly constrained nature of anatomical objects can be well captured with learning based techniques. However, in most recent and promising techniques such as CNN based segmentation it is not obvious how to incorporate such prior knowledge. State-of-the-art methods operate as pixel-wise classifiers where the training objectives do not incorporate the structure and inter-dependencies of the output. To overcome this limitation, we propose a generic training strategy that incorporates anatomical prior knowledge into CNNs through a new regularisation model, which is trained end-to-end. The new framework encourages models to follow the global anatomical properties of the underlying anatomy (e.g. shape, label structure) via learnt non-linear representations of the shape. We show that the proposed approach can be easily adapted to different analysis tasks (e.g. image enhancement, segmentation) and improve the prediction accuracy of the state-of-the-art models. The applicability of our approach is shown on multi-modal cardiac datasets and public benchmarks. Additionally, we demonstrate how the learnt deep models of 3D shapes can be interpreted and used as biomarkers for classification of cardiac pathologies.
Cai J, Bryant JA, Thu-Thao L, et al., 2017, Fractal analysis of left ventricular trabeculations is associated with impaired myocardial deformation in healthy Chinese, Journal of Cardiovascular Magnetic Resonance, Vol: 19, ISSN: 1097-6647
Background:Left ventricular (LV) non-compaction (LVNC) is defined by extreme LV trabeculation, but is measured variably. Here we examined the relationship between quantitative measurement in LV trabeculation and myocardial deformation in health and disease and determined the clinical utility of semi-automated assessment of LV trabeculations.Methods:Cardiovascular magnetic resonance (CMR) was performed in 180 healthy Singaporean Chinese (age 20–69 years; males, n = 91), using balanced steady state free precession cine imaging at 3T. The degree of LV trabeculation was assessed by fractal dimension (FD) as a robust measure of trabeculation complexity using a semi-automated technique. FD measures were determined in healthy men and women to derive normal reference ranges. Myocardial deformation was evaluated using feature tracking. We tested the utility of this algorithm and the normal ranges in 10 individuals with confirmed LVNC (non-compacted/compacted; NC/C ratio > 2.3 and ≥1 risk factor for LVNC) and 13 individuals with suspected disease (NC/C ratio > 2.3).Results:Fractal analysis is a reproducible means of assessing LV trabeculation extent (intra-class correlation coefficient: intra-observer, 0.924, 95% CI [0.761–0.973]; inter-observer, 0.925, 95% CI [0.821–0.970]). The overall extent of LV trabeculation (global FD: 1.205 ± 0.031) was independently associated with increased indexed LV end-diastolic volume and mass (sβ = 0.35; p < 0.001 and sβ = 0.13; p < 0.01, respectively) after adjusting for age, sex and body mass index. Increased LV trabeculation was independently associated with reduced global circumferential strain (sβ = 0.17, p = 0.013) and global diastolic circumferential and radial strain rates (sβ = 0.25, p < 0.001 and sβ = −0.15, p
Biffi C, Simoes Monteiro de Marvao A, Attard M, et al., 2017, Three-dimensional Cardiovascular Imaging-Genetics: A Mass Univariate Framework, Bioinformatics, ISSN: 1367-4803
Motivation: Left ventricular (LV) hypertrophy is a strong predictor of cardiovascular outcomes, but its genetic regulation remains largely unexplained. Conventional phenotyping relies on manual calculation of LV mass and wall thickness, but advanced cardiac image analysis presents an opportunity for highthroughput mapping of genotype-phenotype associations in three dimensions (3D).Results: High-resolution cardiac magnetic resonance images were automatically segmented in 1,124 healthy volunteers to create a 3D shape model of the heart. Mass univariate regression was used to plot a 3D effect-size map for the association between wall thickness and a set of predictors at each vertex in the mesh. The vertices where a significant effect exists were determined by applying threshold-free cluster enhancement to boost areas of signal with spatial contiguity. Experiments on simulated phenotypic signals and SNP replication show that this approach offers a substantial gain in statistical power for cardiac genotype-phenotype associations while providing good control of the false discovery rate. This framework models the effects of genetic variation throughout the heart and can be automatically applied to large population cohorts.Availability: The proposed approach has been coded in an R package freely available at https://doi.org/10.5281/zenodo.834610 together with the clinical data used in this work.
Dawes T, de Marvao A, Shi W, et al., 2017, Systolic motion of the basal right ventricular freewall is the strongest predictor of global function: a high resolution 3D imaging study, Association-of-Anaesthetists-of-Great-Britain-and-Ireland (AAGBI) GAT Annual Scientific Meeting, Publisher: Wiley, Pages: 77-77, ISSN: 0003-2409
Tarroni G, Oktay O, Bai W, et al., 2017, Learning-based heart coverage estimation for short-axis cine cardiac MR images, Functional Imaging and Modelling of the Heart (FIMH), Publisher: Springer, Pages: 73-82
The correct acquisition of short axis (SA) cine cardiac MRimage stacks requires the imaging of the full cardiac anatomy betweenthe apex and the mitral valve plane via multiple 2D slices. While in theclinical practice the SA stacks are usually checked qualitatively to en-sure full heart coverage, visual inspection can become infeasible for largeamounts of imaging data that is routinely acquired, e.g. in populationstudies such as the UK Biobank (UKBB). Accordingly, we propose alearning-based technique for the fully-automated estimation of the heartcoverage for SA image stacks. The technique relies on the identificationof cardiac landmarks (i.e. the apex and the mitral valve sides) on twochamber view long axis images and on the comparison of the landmarks’positions to the volume covered by the SA stack. Landmark detection isperformed using a hybrid random forest approach integrating both re-gression and structured classification models. The technique was appliedon 3000 cases from the UKBB and compared to visual assessment. Theobtained results (error rate = 2.3%, sens. = 73%, spec. = 90%) indicatethat the proposed technique is able to correctly detect the vast majorityof the cases with insufficient coverage, suggesting that it could be usedas a fully-automated quality control step for CMR SA image stacks.
Dawes T, Simoes monteiro de marvao A, Shi W, et al., 2017, Machine learning of three-dimensional right ventricular motion enables outcome prediction in pulmonary hypertension: a cardiac MR imaging study, Radiology, Vol: 283, Pages: 381-390, ISSN: 1527-1315
Purpose: To determine if patient survival and mechanisms of right ventricular (RV) failure in pulmonary hypertension (PH) could be predicted using supervised machine learning of three dimensional patterns of systolic cardiac motion. Materials and methods: The study was approved by a research ethics committee and participants gave written informed consent. 256 patients (143 females, mean age 63 ± 17) with newly diagnosed PH underwent cardiac MR imaging, right heart catheterization (RHC) and six minute walk testing (6MWT) with a median follow up of 4.0 years. Semi automated segmentation of short axis cine images was used to create a three dimensional model of right ventricular motion. Supervised principal components analysis identified patterns of systolic motion which were most strongly predictive of survival. Survival prediction was assessed by the difference in median survival time and the area under the curve (AUC) using time dependent receiver operator characteristic for one year survival. Results: At the end of follow up 33% (93/256) died and one underwent lung transplantation. Poor outcome was predicted by a loss of effective contraction in the septum and freewall coupled with reduced basal longitudinal motion. When added to conventional imaging, hemodynamic, functional and clinical markers, three dimensional cardiac motion improved survival prediction (area under the curve 0.73 vs 0.60, p<0.001) and provided greater differentiation by difference in median survival time between high and low risk groups (13.8 vs 10.7 years, p<0.001). Conclusion:Three dimensional motion modeling with machine learning approaches reveal the adaptations in function that occur early in right heart failure and independently predict outcomes in newly diagnosed PH patients.
Schafer S, de Marvao A, Adami E, et al., 2017, Titin-truncating variants affect heart function in disease cohorts and the general population, Nature Genetics, Vol: 49, Pages: 46-53, ISSN: 1546-1718
Titin-truncating variants (TTNtv) commonly cause dilated cardiomyopathy (DCM). TTNtv are also encountered in ~1% of the general population, where they may be silent, perhaps reflecting allelic factors. To better understand TTNtv, we integrated TTN allelic series, cardiac imaging and genomic data in humans and studied rat models with disparate TTNtv. In patients with DCM, TTNtv throughout titin were significantly associated with DCM. Ribosomal profiling in rat showed the translational footprint of premature stop codons in Ttn, TTNtv-position-independent nonsense-mediated degradation of the mutant allele and a signature of perturbed cardiac metabolism. Heart physiology in rats with TTNtv was unremarkable at baseline but became impaired during cardiac stress. In healthy humans, machine-learning-based analysis of high-resolution cardiac imaging showed TTNtv to be associated with eccentric cardiac remodeling. These data show that TTNtv have molecular and physiological effects on the heart across species, with a continuum of expressivity in health and disease.
Oktay O, Bai W, Lee M, et al., 2016, Multi-input cardiac image super-resolution using convolutional neural networks, International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Pages: 246-254, ISSN: 0302-9743
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.