Avan Suinesiaputra

The Cardiac Atlas Project—an imaging database for computational modeling and statistical atlases of the heart

Motivation: Integrative mathematical and statistical models of cardiac anatomy and physiology can play a vital role in understanding cardiac disease phenotype and planning therapeutic strategies. However, the accuracy and predictive power of such models is dependent upon the breadth and depth of noninvasive imaging datasets. The Cardiac Atlas Project (CAP) has established a large-scale database of cardiac imaging examinations and associated clinical data in order to develop a shareable, web-accessible, structural and functional atlas of the normal and pathological heart for clinical, research and educational purposes. A goal of CAP is to facilitate collaborative statistical analysis of regional heart shape and wall motion and characterize cardiac function among and within population groups. Results: Three main open-source software components were developed: (i) a database with web-interface; (ii) a modeling client for 3D + time visualization and parametric description of shape and motion; and (iii) open data formats for semantic characterization of models and annotations. The database was implemented using a three-tier architecture utilizing MySQL, JBoss and Dcm4chee, in compliance with the DICOM standard to provide compatibility with existing clinical networks and devices. Parts of Dcm4chee were extended to access image specific attributes as search parameters. To date, approximately 3000 de-identified cardiac imaging examinations are available in the database. All software components developed by the CAP are open source and are freely available under the Mozilla Public License Version 1.1 (http://www.mozilla.org/MPL/MPL-1.1.txt). Availability: http://www.cardiacatlas.org Contact: a.young@auckland.ac.nz Supplementary information: Supplementary data are available at Bioinformatics online.

A collaborative resource to build consensus for automated left ventricular segmentation of cardiac MR images.

A collaborative framework was initiated to establish a community resource of ground truth segmentations from cardiac MRI. Multi-site, multi-vendor cardiac MRI datasets comprising 95 patients (73 men, 22 women; mean age 62.73±11.24years) with coronary artery disease and prior myocardial infarction, were randomly selected from data made available by the Cardiac Atlas Project (Fonseca et al., 2011). Three semi- and two fully-automated raters segmented the left ventricular myocardium from short-axis cardiac MR images as part of a challenge introduced at the STACOM 2011 MICCAI workshop (Suinesiaputra et al., 2012). Consensus myocardium images were generated based on the Expectation-Maximization principle implemented by the STAPLE algorithm (Warfield et al., 2004). The mean sensitivity, specificity, positive predictive and negative predictive values ranged between 0.63 and 0.85, 0.60 and 0.98, 0.56 and 0.94, and 0.83 and 0.92, respectively, against the STAPLE consensus. Spatial and temporal agreement varied in different amounts for each rater. STAPLE produced high quality consensus images if the region of interest was limited to the area of discrepancy between raters. To maintain the quality of the consensus, an objective measure based on the candidate automated rater performance distribution is proposed. The consensus segmentation based on a combination of manual and automated raters were more consistent than any particular rater, even those with manual input. The consensus is expected to improve with the addition of new automated contributions. This resource is open for future contributions, and is available as a test bed for the evaluation of new segmentation algorithms, through the Cardiac Atlas Project (www.cardiacatlas.org).

Atlas-Based Quantification of Cardiac Remodeling Due to Myocardial Infarction

Myocardial infarction leads to changes in the geometry (remodeling) of the left ventricle (LV) of the heart. The degree and type of remodeling provides important diagnostic information for the therapeutic management of ischemic heart disease. In this paper, we present a novel analysis framework for characterizing remodeling after myocardial infarction, using LV shape descriptors derived from atlas-based shape models. Cardiac magnetic resonance images from 300 patients with myocardial infarction and 1991 asymptomatic volunteers were obtained from the Cardiac Atlas Project. Finite element models were customized to the spatio-temporal shape and function of each case using guide-point modeling. Principal component analysis was applied to the shape models to derive modes of shape variation across all cases. A logistic regression analysis was performed to determine the modes of shape variation most associated with myocardial infarction. Goodness of fit results obtained from end-diastolic and end-systolic shapes were compared against the traditional clinical indices of remodeling: end-diastolic volume, end-systolic volume and LV mass. The combination of end-diastolic and end-systolic shape parameter analysis achieved the lowest deviance, Akaike information criterion and Bayesian information criterion, and the highest area under the receiver operating characteristic curve. Therefore, our framework quantitatively characterized remodeling features associated with myocardial infarction, better than current measures. These features enable quantification of the amount of remodeling, the progression of disease over time, and the effect of treatments designed to reverse remodeling effects.

Left ventricular shape variation in asymptomatic populations: the multi-ethnic study of atherosclerosis

BackgroundAlthough left ventricular cardiac geometric indices such as size and sphericity characterize adverse remodeling and have prognostic value in symptomatic patients, little is known of shape distributions in subclinical populations. We sought to quantify shape variation across a large number of asymptomatic volunteers, and examine differences among sub-cohorts.MethodsAn atlas was constructed comprising 1,991 cardiovascular magnetic resonance (CMR) cases contributed from the Multi-Ethnic Study of Atherosclerosis baseline examination. A mathematical model describing regional wall motion and shape was used to establish a coordinate map registered to the cardiac anatomy. The model was automatically customized to left ventricular contours and anatomical landmarks, corrected for breath-hold mis-registration between image slices. Mathematical techniques were used to characterize global shape distributions, after removal of translations, rotations, and scale due to height. Differences were quantified among ethnicity, sex, smoking, hypertension and diabetes sub-cohorts.ResultsThe atlas construction process yielded accurate representations of global shape (errors between manual and automatic surface points in 244 validation cases were less than the image pixel size). After correction for height, the dominant shape component was associated with heart size, explaining 32% of the total shape variance at end-diastole and 29% at end-systole. After size, the second dominant shape component was sphericity at end-diastole (13%), and concentricity at end-systole (10%). The resulting shape components distinguished differences due to ethnicity and risk factors with greater statistical power than traditional mass and volume indices.ConclusionsWe have quantified the dominant components of global shape variation in the adult asymptomatic population. The data and results are available at cardiacatlas.org. Shape distributions were principally explained by size, sphericity and concentricity, which are known correlates of adverse outcomes. Atlas-based global shape analysis provides a powerful method for quantifying left ventricular shape differences in asymptomatic populations.Trial registrationClinicalTrials.gov NCT00005487

Atlas-based analysis of cardiac shape and function: correction of regional shape bias due to imaging protocol for population studies

BackgroundCardiovascular imaging studies generate a wealth of data which is typically used only for individual study endpoints. By pooling data from multiple sources, quantitative comparisons can be made of regional wall motion abnormalities between different cohorts, enabling reuse of valuable data. Atlas-based analysis provides precise quantification of shape and motion differences between disease groups and normal subjects. However, subtle shape differences may arise due to differences in imaging protocol between studies.MethodsA mathematical model describing regional wall motion and shape was used to establish a coordinate system registered to the cardiac anatomy. The atlas was applied to data contributed to the Cardiac Atlas Project from two independent studies which used different imaging protocols: steady state free precession (SSFP) and gradient recalled echo (GRE) cardiovascular magnetic resonance (CMR). Shape bias due to imaging protocol was corrected using an atlas-based transformation which was generated from a set of 46 volunteers who were imaged with both protocols.ResultsShape bias between GRE and SSFP was regionally variable, and was effectively removed using the atlas-based transformation. Global mass and volume bias was also corrected by this method. Regional shape differences between cohorts were more statistically significant after removing regional artifacts due to imaging protocol bias.ConclusionsBias arising from imaging protocol can be both global and regional in nature, and is effectively corrected using an atlas-based transformation, enabling direct comparison of regional wall motion abnormalities between cohorts acquired in separate studies.

Big Heart Data: Advancing Health Informatics Through Data Sharing in Cardiovascular Imaging

The burden of heart disease is rapidly worsening due to the increasing prevalence of obesity and diabetes. Data sharing and open database resources for heart health informatics are important for advancing our understanding of cardiovascular function, disease progression and therapeutics. Data sharing enables valuable information, often obtained at considerable expense and effort, to be reused beyond the specific objectives of the original study. Many government funding agencies and journal publishers are requiring data reuse, and are providing mechanisms for data curation and archival. Tools and infrastructure are available to archive anonymous data from a wide range of studies, from descriptive epidemiological data to gigabytes of imaging data. Meta-analyses can be performed to combine raw data from disparate studies to obtain unique comparisons or to enhance statistical power. Open benchmark datasets are invaluable for validating data analysis algorithms and objectively comparing results. This review provides a rationale for increased data sharing and surveys recent progress in the cardiovascular domain. We also highlight the potential of recent large cardiovascular epidemiological studies enabling collaborative efforts to facilitate data sharing, algorithms benchmarking, disease modeling and statistical atlases.

Large scale left ventricular shape atlas using automated model fitting to contours

We demonstrate that large legacy databases of manually segmented cardiac MR images can be used to build a shape atlas based on 3D left-ventricular finite-element models. We make use of the Cardiac Atlas Project database to build an atlas of 2,045 asymptomatic cases from the MESA study. Manually placed anatomical landmarks on long-axis and short-axis magnetic resonance images were combined with manually drawn contours on the short axis images which were corrected for breath-hold mis-registration using an automated method. The contours were then fitted by the model using linear least squares optimisation. The fitting error was 0.5 ±0.4 mm at end-diastole and 0.5 ±0.6 mm at end-systole (mean ± std. dev.). Results were validated against 3D models created by experts in a sub-sample of 253 cases using manual breath-hold registration. The atlas surface error was 1.3 ±0.8 mm at end-diastole and 1.2 ±0.9 mm at end-systole. The end-diastolic volume error was 9.0 ±8.7 ml; the end-systolic volume error was 0.8 ±6.3 ml; and the mass error 5.9 ±12.9 g. These differences arose mainly at the base and apex because long-axis images were used in the validation models, but were only used in the automated models to define basal fiducial markers. All models were aligned and scaled, and finally analysed by principal component analysis. Significant differences were found in the first mode shape (sphericity) by gender, smoking, and hypertension.

An atlas for cardiac MRI regional wall motion and infarct scoring

Regional wall motion and infarct scoring of MR images are routine clinical tools to grade performance and scarring in the heart. The aim of this paper is to provide a framework for automatic scoring to alert the diagnostician to potential regions of abnormality. We investigated different shape and motion configurations of a finite-element cardiac atlas of the left ventricle. Two patient populations were used: 300 asymptomatic volunteers and 105 patients with myocardial infarction, both randomly selected from the Cardiac Atlas Project database. Support vector machines were employed to estimate the boundaries between the asymptomatic control and patient groups for each of 16 standard anatomical regions in the heart. Ground truth visual wall motion scores from standard cines and infarct scoring from late enhancement were provided by experienced observers. From all configurations, end-systolic shape best predicted wall motion abnormalities (global accuracy 78%, positive predictive value 85%, specificity 91%, sensitivity 60%) and infarct scoring (74%, 72%, 91%, 44%). In conclusion, computer assisted wall motion and infarct scoring has the potential to provide robust identification of those segments requiring further clinical attention; in particular, the high specificity and relatively low sensitivity could help avoid unnecessary late gadolinium rescanning of patients.

Statistical Shape Modeling of the Left Ventricle: Myocardial Infarct Classification Challenge

Statistical shape modeling is a powerful tool for visualizing and quantifying geometric and functional patterns of the heart. After myocardial infarction (MI), the left ventricle typically remodels in response to physiological challenges. Several methods have been proposed in the literature to describe statistical shape changes. Which method best characterizes the left ventricular remodeling after MI is an open research question. A better descriptor of remodeling is expected to provide a more accurate evaluation of disease status in MI patients. We therefore designed a challenge to test shape characterization in MI given a set of three-dimensional left ventricular surface points. The training set comprised 100 MI patients, and 100 asymptomatic volunteers (AV). The challenge was initiated in 2015 at the Statistical Atlases and Computational Models of the Heart workshop, in conjunction with the MICCAI conference. The training set with labels was provided to participants, who were asked to submit the likelihood of MI from a different (validation) set of 200 cases (100 AV and 100 MI). Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve were used as the outcome measures. The goals of this challenge were to 1) establish a common dataset for evaluating statistical shape modeling algorithms in MI, and 2) test whether statistical shape modeling provides additional information characterizing MI patients over standard clinical measures. Eleven groups with a wide variety of classification and feature extraction approaches participated in this challenge. All methods achieved excellent classification results with accuracy ranges from 0.83 to 0.98. The areas under the receiver operating characteristic curves were all above 0.90. Four methods showed significantly higher performance than standard clinical measures. The dataset and software for evaluation are available from the Cardiac Atlas Project website.11 http://www.cardiacatlas.org.

Cardiac image modelling: Breadth and depth in heart disease

With the advent of large-scale imaging studies and big health data, and the corresponding growth in analytics, machine learning and computational image analysis methods, there are now exciting opportunities for deepening our understanding of the mechanisms and characteristics of heart disease. Two emerging fields are computational analysis of cardiac remodelling (shape and motion changes due to disease) and computational analysis of physiology and mechanics to estimate biophysical properties from non-invasive imaging. Many large cohort studies now underway around the world have been specifically designed based on non-invasive imaging technologies in order to gain new information about the development of heart disease from asymptomatic to clinical manifestations. These give an unprecedented breadth to the quantification of population variation and disease development. Also, for the individual patient, it is now possible to determine biophysical properties of myocardial tissue in health and disease by interpreting detailed imaging data using computational modelling. For these population and patient-specific computational modelling methods to develop further, we need open benchmarks for algorithm comparison and validation, open sharing of data and algorithms, and demonstration of clinical efficacy in patient management and care. The combination of population and patient-specific modelling will give new insights into the mechanisms of cardiac disease, in particular the development of heart failure, congenital heart disease, myocardial infarction, contractile dysfunction and diastolic dysfunction.