Segmentation Of Cardiac Chambers Research Articles

Abstract 16809: Deep-Learning-Based Assessment of Left Atrial Volume in Coronary Calcium CT: Association With NT-proBNP and Clinical Outcomes

Introduction: Artificial Intelligence (AI) promises to extract imaging biomarkers for cardio-cerebrovascular (CV) risk prediction, though validation is still lacking. Recognizing the importance of left atrial pathology on CV events, we developed an AI algorithm to measure left atrial volume (LAvol) from CAC-CT. Its correlation with MRI, biomarkers, and CV incidents was then tested in the Dallas Heart Study 2 (DHS). Hypothesis: LAvol will be associated with elevations in baseline NT-proBNP levels and incident CV events. Methods: We trained a 3D U-Net model for cardiac chamber segmentation on CAC-CT, assessing congruity with manual segmentation via the Dice score. Normalized left atrial volume (LAvol/BSA) from 3D segmentations was compared to MRI using ICC, and its associations with baseline NT-proBNP and incident Afib, TIA, and stroke were evaluated with logistic regression and Cox models. Results: Cardiac segmentation showed an excellent Dice score of 0.93 ± 0.01. Moderate correlation was observed between CAC-CT estimates of LAvol/BSA and MRI (ICC 0.57, 95% CI 0.52-0.62). Our study, excluding those with prior events, comprised 1852 participants (mean age 50, 41% male, 46% black, 15% Hispanic). Upon multivariable adjustment (detailed in the table), LAvol/BSA was linked with elevated NT-proBNP (≥ 100 pg/mL) (OR 1.9, 95% CI 1.6-2.2 per 1SD LAvol/BSA). Over a 12.5-year follow-up, CAC-CT-derived LAvol/BSA showed association with a composite of Afib, TIA, and stroke (55 events) (HR 1.4, 95% CI 1.1 - 1.8), and Afib (HR 1.7, 95% CI 1.2 - 2.3), post-adjustment. No significant association was found with TIA (HR 1.2, 95% CI 0.6 - 2.1) or stroke (HR 1.3, 95% CI 0.9 - 1.8). Among 749 subjects, MRI-derived LAvol/BSA correlated with NT-proBNP (OR 1.7, 95% CI 1.44 - 2.10) but not with other clinical outcomes. Conclusions: Given associations with NT-proBNP levels and incident CV events, AI-measured LAvol from CAC-CT promises to improve risk prediction.

Heart Chamber Segmentation in Cardiomegaly Conditions Using the CNN Method with U-Net Architecture

Cardiomegaly is a disease in which sufferers show no symptoms and have symptoms such as shortness of breath, abnormal heartbeat and edema. Cardiomegaly will cause the sufferer's heart to pump harder than usual. Early diagnosis of cardiomegaly can help make decisions about whether the heart is abnormal or normal. In addition, due to the problem that manual examination takes time and requires human interpretation and experience, tools are needed to automatically develop and identify normal and abnormal hearts. Therefore, this study proposes cardiac chamber segmentation using 2D (two-dimensional) ultrasound convolutional neural networks for rapid cardiomegaly screening in clinical applications based on heart ultrasound examination. The proposed approach uses a CNN with a U-Net architecture model with abnormal and normal heart data. The research results obtained used the pixel matrix evaluation Avg_accuracy of 99.50%, Val_accuracy of 97.98% and Mean_IoU of 90.01%.

Abstract 16178: A Deep Learning Algorithm to Estimate Left Ventricular Mass From Coronary Calcium CT Scans: Association With Incident Heart Failure Events From the Dallas Heart Study

Introduction: Coronary artery calcium CT (CAC-CT) scans used for ASCVD risk prediction may contain additional clinical information that is currently underexploited. We developed a novel deep-learning algorithm to estimate LV mass from CAC-CT (CT-LVM) and analyzed associations with cardiac biomarkers and incident CV outcomes in the Dallas Heart Study (DHS). Hypothesis: Higher CT-LVM will be associated with baseline cardiac biomarker elevations (hsTnT and NT-proBNP) and a higher risk of incident CVD (heart failure, MI, stroke, and all-cause death). Methods: We included 1841 DHS phase 2 participants (mean age 50, 59% female, 46% Black race) who were free of baseline CVD and underwent CAC-CT and cardiac MRI. Our novel deep-learning algorithm features automatic cardiac chamber segmentation via a 3D U-Net architecture on CAC-CT, followed by a 3D DenseNet model to predict LV mass, using concurrent cardiac MRI as gold standard. LV mass was indexed to BSA. Statistical analysis included multivariable logistic regression models for baseline biomarker elevation assessment (hs-cTnT > 5ng/L or NT-proBNP > 100pg/mL) and Cox proportional hazards models for clinical outcomes. Models were adjusted for age, sex, ethnicity, BMI, hypertension, blood pressure, diabetes, smoking, creatinine, and CAC score. Results: CT-LVM correlated strongly with MRI defined LV mass (R = 0.87). CT-LVM/BSA was associated with elevated hsTnT (adjusted OR 1.3, 95% CI 1.1-1.6) and NT-proBNP (adjusted OR 1.8, 95% CI 1.5-2.2). There were 172 CVD events over a median follow up of 12.5 years. Higher CT-LVM was associated with higher CVD risk (adjusted HR 1.4, 95% CI 1.2-1.6) and higher risk of incident heart failure (adjusted HR 2.1, 95% CI 1.5-2.9). Conclusions: A novel deep-learning method for estimating LV mass from CAC-CT provided independent risk information from traditional risk factors and CAC score. This represents a novel low-cost approach to improving the risk assessment opportunities from a CAC-CT.

Optimizing Deep Learning for Cardiac MRI Segmentation: The Impact of Automated Slice Range Classification

Cardiac magnetic resonance imaging is crucial for diagnosing cardiovascular diseases, but lengthy postprocessing and manual segmentation can lead to observer bias. Deep learning (DL) has been proposed for automated cardiac segmentation; however, its effectiveness is limited by the slice range selection from base to apex. In this study, we integrated an automated slice range classification step to identify basal to apical short-axis slices before DL-based segmentation. We employed publicly available Multi-Disease, Multi-View & Multi-Center Right Ventricular Segmentation in Cardiac MRI data set with short-axis cine data from 160 training, 40 validation, and 160 testing cases. Three classification and seven segmentation DL models were studied. The top-performing segmentation model was assessed with and without the classification model. Model validation to compare automated and manual segmentation was performed using Dice score and Hausdorff distance and clinical indices (correlation score and Bland-Altman plots). The combined classification (CBAM-integrated 2D-CNN) and segmentation model (2D-UNet with dilated convolution block) demonstrated superior performance, achieving Dice scores of 0.952 for left ventricle (LV), 0.933 for right ventricle (RV), and 0.875 for myocardium, compared to the stand-alone segmentation model (0.949 for LV, 0.925 for RV, and 0.867 for myocardium). Combined classification and segmentation model showed high correlation (0.92-0.99) with manual segmentation for biventricular volumes, ejection fraction, and myocardial mass. The mean absolute difference (2.8-8.3mL) for clinical parameters between automated and manual segmentation was within the interobserver variability range, indicating comparable performance to manual annotation. Integrating an initial automated slice range classification step into the segmentation process improves the performance of DL-based cardiac chamber segmentation.

Deep learning architecture for 3D image super-resolution of late gadolinium enhanced cardiac MRI.

High-resolution late gadolinium enhanced (LGE) cardiac magnetic resonance imaging (MRI) volumes are difficult to acquire due to the limitations of the maximal breath-hold time achievable by the patient. This results in anisotropic 3D volumes of the heart with high in-plane resolution, but low-through-plane resolution. Thus, we propose a 3D convolutional neural network (CNN) approach to improve the through-plane resolution of the cardiac LGE-MRI volumes. We present a 3D CNN-based framework with two branches: a super-resolution branch to learn the mapping between low-resolution and high-resolution LGE-MRI volumes, and a gradient branch that learns the mapping between the gradient map of low-resolution LGE-MRI volumes and the gradient map of high-resolution LGE-MRI volumes. The gradient branch provides structural guidance to the CNN-based super-resolution framework. To assess the performance of the proposed CNN-based framework, we train two CNN models with and without gradient guidance, namely, dense deep back-projection network (DBPN) and enhanced deep super-resolution network. We train and evaluate our method on the 2018 atrial segmentation challenge dataset. Additionally, we also evaluate these trained models on the left atrial and scar quantification and segmentation challenge 2022 dataset to assess their generalization ability. Finally, we investigate the effect of the proposed CNN-based super-resolution framework on the 3D segmentation of the left atrium (LA) from these cardiac LGE-MRI image volumes. Experimental results demonstrate that our proposed CNN method with gradient guidance consistently outperforms bicubic interpolation and the CNN models without gradient guidance. Furthermore, the segmentation results, evaluated using Dice score, obtained using the super-resolved images generated by our proposed method are superior to the segmentation results obtained using the images generated by bicubic interpolation () and the CNN models without gradient guidance (). The presented CNN-based super-resolution method with gradient guidance improves the through-plane resolution of the LGE-MRI volumes and the structure guidance provided by the gradient branch can be useful to aid the 3D segmentation of cardiac chambers, such as LA, from the 3D LGE-MRI images.

Abstract 14453: Deep Learning Enabled Measurement of Cardiac Sphericity Predicts Disease Risk Among Subjects With Normal Left Ventricular Size and Function

Introduction: Cardiac imaging primarily focuses on measurements of chamber size and function. Quantitative assessment of morphology may provide additional prognostic information. Left ventricle (LV) sphericity has been proposed as one metric but has not been studied in large populations and is poorly characterized in patients with normal LVs. We aimed to assess the clinical relevance and genetic drivers of sphericity among individuals with normal LV size and systolic function using deep learning enabled phenotyping of cardiac MRIs. Methods: The study cohort consisted of UK Biobank participants with normal LV ejection fraction, LV end-systolic, and end-diastolic volumes by MRI. A fully convolutional neural network was used to perform image segmentation of cardiac chambers. The main exposure was the LV sphericity index, defined as the ratio of short-axis to long-axis lengths of the smallest rectangle encompassing the LV in a 4-chamber view. Risk for incident outcomes were measured using Cox analysis. Genome-wide association study (GWAS) was performed using HAIL software with standard quality-control. Genetic correlations were measured using linkage-disequilibrium score regression. Causal inference was performed using two-sample Mendelian randomization (MR). Results: We identified 38,897 subjects with normal LV volumes and ejection fraction. Mean age was 55 ± 8, and 54% were female. After adjusting for age, sex, BMI, hypertension, and pulse-rate, a 1 standard deviation increase in sphericity index was associated with increased risk of atrial fibrillation (HR 1.31, 95% CI 1.23-1.38, P <2x10 -16 ) and cardiomyopathy (HR 1.62, 95% CI 1.29-2.02, P 2.4x10 -5 ). GWAS of sphericity index identified 4 loci at genome-wide significance, with 3 candidate genes previously implicated in cardiac development and conduction. Sphericity index showed a genetic correlation with cardiomyopathy ( rg 0.42, P 0.01) but not with atrial fibrillation ( rg 0.04, P 0.5). Bidirectional MR analyses supported cardiomyopathy as being causal for sphericity but not the reverse. Conclusion LV sphericity in subjects with normal size and function is genetically mediated and may reflect subclinical manifestations of cardiomyopathic risk prior to LV remodeling by standard measures.

MAEF-Net: Multi-attention efficient feature fusion network for left ventricular segmentation and quantitative analysis in two-dimensional echocardiography

The segmentation of cardiac chambers and the quantification of clinical functional metrics in dynamic echocardiography are the keys to the clinical diagnosis of heart disease. Identifying the end-diastolic frames (EDFs) and end-systolic frames (ESFs) and manually segmenting the left ventricle in the echocardiographic cardiac cycle before obtaining the left ventricular ejection fraction (LVEF) is a time-consuming and tedious task for clinicians. In this work, we proposed a deep learning-based fully automated echocardiographic analysis method. We proposed a multi-attention efficient feature fusion network (MAEF-Net) to automatically segment the left ventricle. Then, EDFs and ESFs in all cardiac cycles were automatically detected to compute LVEF. The MAEF-Net method used a multi-attention mechanism to guide the network to capture heartbeat features effectively, while suppressing noise, and incorporated deep supervision mechanism and spatial pyramid feature fusion to enhance feature extraction capabilities. The proposed method was validated on the public EchoNet-Dynamic dataset (n = 1226). The Dice similarity coefficient (DSC) of the left ventricular segmentation reached (93.10 ± 2.22)%, and the mean absolute error (MAE) of cardiac phase detection was (2.36 ± 2.23) frames. The MAE for predicting LVEF was 6.29 %. The proposed method was also validated on a private clinical dataset (n = 22). The DSC of the left ventricular segmentation reached (92.81 ± 2.85)%, and the MAE of cardiac phase detection was (2.25 ± 2.27) frames. The MAE for predicting LVEF was 5.91 %, and the Pearson correlation coefficient r reached 0.96. The proposed method may be used as a new method for automatic left ventricular segmentation and quantitative analysis in two-dimensional echocardiography. Our code and trained models will be made available publicly at https://github.com/xiaojinmao-code/MAEF-Net.

A method for direct estimation of left ventricular global longitudinal strain rate from echocardiograms

We present a new method for measuring global longitudinal strain and global longitudinal strain rate from 2D echocardiograms using a logarithmic-transform correlation (LTC) method. Traditional echocardiography strain analysis depends on user inputs and chamber segmentation, which yield increased measurement variability. In contrast, our approach is automated and does not require cardiac chamber segmentation and regularization, thus eliminating these issues. The algorithm was benchmarked against two conventional strain analysis methods using synthetic left ventricle ultrasound images. Measurement error was assessed as a function of contrast-to-noise ratio (CNR) using mean absolute error and root-mean-square error. LTC showed better agreement to the ground truth strain {({varvec{R}}}^{2}=0.91) and ground truth strain rate {({varvec{R}}}^{2}=0.85) compared with agreement to ground truth for two block-matching speckle tracking algorithms (one based on sum of square difference and the other on Fourier transform correlation; strain {({varvec{R}}}^{2}=0.70), strain rate {({varvec{R}}}^{2}=0.70)). A 200% increase in strain measurement accuracy was observed compared to the conventional algorithms. Subsequently, we tested the method using a 53-subject clinical cohort (20 subjects diseased with cardiomyopathy, 33 healthy controls). Our method distinguished between normal and abnormal left ventricular function with an AUC = 0.89, a 5% improvement over the conventional GLS algorithms.

Fetal Heart and Descending Aorta Detection in Four-Chamber View of Fetal Echocardiography.

Automatic analysis of fetal heart and related components in fetal echocardiography can help cardiologists to reach a diagnosis for Congenital Heart Disease (CHD). Previous studies mainly focused on cardiac chamber segmentation, while few researches deal with the cardiac component detection. In this paper, we tackle the task of simultaneous detection of the fetal heart and descending aorta in four-chamber view of fetal echocardiography, which is useful to analyze some kinds of CHD, such as left/right atrial isomerism, dextroversion of heart, etc. Several CNN-based object detection methods with different backbones are thoroughly evaluated, and finally, the Hybrid Task Cascade method with HRNet is selected as the detection method. Experiments on a fetal echocardiography dataset show that the method can achieve superior performance according to common-used evaluation metrics.Clinical relevance-This can be used to help the cardiologists to estimate the position of the fetal heart and the descending aorta, which is also useful to estimate the direction of the cardiac axis and apex and analyze some kinds of CHD, such as left/right atrial isomerism, dextroversion of heart, etc.

Simultaneous Segmentation of Four Cardiac Chambers in Fetal Echocardiography.

Accurate segmentation of cardiac chambers is helpful for the diagnosis of Congenital Heart Disease (CHD) in fetal echocardiography. Previous studies mainly focused on single cardiac chamber segmentation, which cannot provide sufficient information for the cardiologists. In this paper, we present an instance segmentation approach capable of segmenting four cardiac chambers accurately and simultaneously. A novel object proposal recovery strategy is further deployed to retrieve possible missing objects. To alleviate the shortage of medical data and further improve the segmentation performance, we utilize a rotation and distortion method for data augmentation. Experiments on a fetal echocardiography dataset of 319 fetuses demonstrate that the proposed approach can achieve superior performance according to common-used evaluation metrics.Clinical relevance-This can be used to help the cardiologists to better analyze the structure and function of the fetal heart.

AI-Based Quantification of Planned Radiation Therapy Dose to Cardiac Structures and Coronary Arteries in Patients With Breast Cancer

The purpose of this work is to develop and evaluate an automatic deep learning method for segmentation of cardiac chambers and large arteries, and localization of the 3 main coronary arteries in radiation therapy planning on computed tomography (CT). In addition, a second purpose is to determine the planned radiation therapy dose to cardiac structures for breast cancer therapy. Eighteen contrast-enhanced cardiac scans acquired with a dual-layer-detector CT scanner were included for method development. Manual reference annotations of cardiac chambers, large arteries, and coronary artery locations were made in the contrast scans and transferred to virtual noncontrast images, mimicking noncontrast-enhanced CT. In addition, 31 noncontrast-enhanced radiation therapy treatment planning CTs with corresponding dose-distribution maps of breast cancer cases were included for evaluation. For reference, cardiac chambers and large vessels were manually annotated in two 2-dimensional (2D) slices per scan (26 scans, totaling 52 slices) and in 3-dimensional (3D) scan volumes in 5 scans. Coronary artery locations were annotated on 3D imaging. The method uses an ensemble of convolutional neural networks with 2 output branches that perform 2 distinct tasks: (1) segmentation of the cardiac chambers and large arteries and (2) localization of coronary arteries. Training was performed using reference annotations and virtual noncontrast cardiac scans. Automatic segmentation of the cardiac chambers and large vessels and the coronary artery locations was evaluated in radiation therapy planning CT with Dice score (DSC) and average symmetrical surface distance (ASSD). The correlation between dosimetric parameters derived from the automatic and reference segmentations was evaluated with R2. For cardiac chambers and large arteries, median DSC was 0.76 to 0.88, and the median ASSD was 0.17 to 0.27 cm in 2D slice evaluation. 3D evaluation found a DSC of 0.87 to 0.93 and an ASSD of 0.07 to 0.10 cm. Median DSC of the coronary artery locations ranged from 0.80 to 0.91. R2 values of dosimetric parameters were 0.77 to 1.00 for the cardiac chambers and large vessels, and 0.76 to 0.95 for the coronary arteries. The developed and evaluated method can automatically obtain accurate estimates of planned radiation dose and dosimetric parameters for the cardiac chambers, large arteries, and coronary arteries.

Spectral augmentation for heart chambers segmentation on conventional contrasted and unenhanced CT scans: an in-depth study.

Recently, machine learning has outperformed established tools for automated segmentation in medical imaging. However, segmentation of cardiac chambers still proves challenging due to the variety of contrast agent injection protocols used in clinical practice, inducing disparities of contrast between cavities. Hence, training a generalist network requires large training datasets representative of these protocols. Furthermore, segmentation on unenhanced CT scans is further hindered by the challenge of obtaining ground truths from these images. Newly available spectral CT scanners allow innovative image reconstructions such as virtual non-contrast (VNC) imaging, mimicking non-contrasted conventional CT studies from a contrasted scan. Recent publications have demonstrated that networks can be trained using VNC to segment contrasted and unenhanced conventional CT scans to reduce annotated data requirements and the need for annotations on unenhanced scans. We propose an extensive evaluation of this statement. We undertake multiple trainings of a 3D multi-label heart segmentation network with (HU-VNC) and without (HUonly) VNC as augmentation, using decreasing training dataset sizes (114, 76, 57, 38, 29, 19 patients). At each step, both networks are tested on a multi-vendor, multi-centric dataset of 122 patients, including different protocols: pulmonary embolism (PE), chest-abdomen-pelvis (CAP), heart CT angiography (CTA) and true non-contrast scans (TNC). An in-depth comparison of resulting Dice coefficients and distance metrics is performed for the networks trained on the largest dataset. HU-VNC-trained on 57 patients significantly outperforms HUonly trained on 114 regarding CAP and TNC scans (mean Dice coefficients of 0.881/0.835 and 0.882/0.416, respectively). When trained on the largest dataset, significant improvements in all labels are noted for TNC and CAP scans (mean Dice coefficient of 0.882/0.416 and 0.891/0.835, respectively). Adding VNC images as training augmentation allows the network to perform on unenhanced scans and improves segmentations on other imaging protocols, while using a reduced training dataset.

CNN-Based Cardiac Motion Extraction to Generate Deformable Geometric Left Ventricle Myocardial Models from Cine MRI.

Patient-specific left ventricle (LV) myocardial models have the potential to be used in a variety of clinical scenarios for improved diagnosis and treatment plans. Cine cardiac magnetic resonance (MR) imaging provides high resolution images to reconstruct patient-specific geometric models of the LV myocardium. With the advent of deep learning, accurate segmentation of cardiac chambers from cine cardiac MR images and unsupervised learning for image registration for cardiac motion estimation on a large number of image datasets is attainable. Here, we propose a deep leaning-based framework for the development of patient-specific geometric models of LV myocardium from cine cardiac MR images, using the Automated Cardiac Diagnosis Challenge (ACDC) dataset. We use the deformation field estimated from the VoxelMorph-based convolutional neural network (CNN) to propagate the isosurface mesh and volume mesh of the end-diastole (ED) frame to the subsequent frames of the cardiac cycle. We assess the CNN-based propagated models against segmented models at each cardiac phase, as well as models propagated using another traditional nonrigid image registration technique. Additionally, we generate dynamic LV myocardial volume meshes at all phases of the cardiac cycle using the log barrier-based mesh warping (LBWARP) method and compare them with the CNN-propagated volume meshes.

Utility of deep learning networks for the generation of artificial cardiac magnetic resonance images in congenital heart disease

BackgroundDeep learning algorithms are increasingly used for automatic medical imaging analysis and cardiac chamber segmentation. Especially in congenital heart disease, obtaining a sufficient number of training images and data anonymity issues remain of concern.MethodsProgressive generative adversarial networks (PG-GAN) were trained on cardiac magnetic resonance imaging (MRI) frames from a nationwide prospective study to generate synthetic MRI frames. These synthetic frames were subsequently used to train segmentation networks (U-Net) and the quality of the synthetic training images, as well as the performance of the segmentation network was compared to U-Net-based solutions trained entirely on patient data.ResultsCardiac MRI data from 303 patients with Tetralogy of Fallot were used for PG-GAN training. Using this model, we generated 100,000 synthetic images with a resolution of 256 × 256 pixels in 4-chamber and 2-chamber views. All synthetic samples were classified as anatomically plausible by human observers. The segmentation performance of the U-Net trained on data from 42 separate patients was statistically significantly better compared to the PG-GAN based training in an external dataset of 50 patients, however, the actual difference in segmentation quality was negligible (< 1% in absolute terms for all models).ConclusionWe demonstrate the utility of PG-GANs for generating large amounts of realistically looking cardiac MRI images even in rare cardiac conditions. The generated images are not subject to data anonymity and privacy concerns and can be shared freely between institutions. Training supervised deep learning segmentation networks on this synthetic data yielded similar results compared to direct training on original patient data.

Cardiac Chamber Segmentation Using Deep Learning on Magnetic Resonance Images from Patients Before and After Atrial Septal Occlusion Surgery.

We propose a robust technique for segmenting magnetic resonance images of post-atrial septal occlusion intervention in the cardiac chamber. The technique can be used to determine the surgical outcomes of atrial septal defects before and after implantation of a septal occluder, which intends to provide volume restoration of the right and left atria. A variant of the U-Net architecture is used to perform atrial segmentation via a deep convolutional neural network. The method was evaluated on a dataset containing 550 two-dimensional image slices, outperforming conventional active contouring regarding the Dice similarity coefficient, Jaccard index, and Hausdorff distance, and achieving segmentation in the presence of ghost artifacts that occlude the atrium outline. Moreover, the proposed technique is closer to manual segmentation than the snakes active contour model. After segmentation, we computed the volume ratio of right to left atria, obtaining a smaller ratio that indicates better restoration. Hence, the proposed technique allows to evaluate the surgical success of atrial septal occlusion and may support diagnosis regarding the accurate evaluation of atrial septal defects before and after occlusion procedures.

Automated Segmentation of Cardiac Chambers from Cine Cardiac MRI Using an Adversarial Network Architecture.

Cine cardiac magnetic resonance imaging (CMRI), the current gold standard for cardiac function analysis, provides images with high spatio-temporal resolution. Computing clinical cardiac parameters like ventricular blood-pool volumes, ejection fraction and myocardial mass from these high resolution images is an important step in cardiac disease diagnosis, therapy planning and monitoring cardiac health. An accurate segmentation of left ventricle blood-pool, myocardium and right ventricle blood-pool is crucial for computing these clinical cardiac parameters. U-Net inspired models are the current state-of-the-art for medical image segmentation. SegAN, a novel adversarial network architecture with multi-scale loss function, has shown superior segmentation performance over U-Net models with single-scale loss function. In this paper, we compare the performance of stand-alone U-Net models and U-Net models in SegAN framework for segmentation of left ventricle blood-pool, myocardium and right ventricle blood-pool from the 2017 ACDC segmentation challenge dataset. The mean Dice scores achieved by training U-Net models was on the order of 89.03%, 89.32% and 88.71% for left ventricle blood-pool, myocardium and right ventricle blood-pool, respectively. The mean Dice scores achieved by training the U-Net models in SegAN framework are 91.31%, 88.68% and 90.93% for left ventricle blood-pool, myocardium and right ventricle blood-pool, respectively.

Fully Automated Echocardiogram Interpretation in Clinical Practice.

Automated cardiac image interpretation has the potential to transform clinical practice in multiple ways, including enabling serial assessment of cardiac function by nonexperts in primary care and rural settings. We hypothesized that advances in computer vision could enable building a fully automated, scalable analysis pipeline for echocardiogram interpretation, including (1) view identification, (2) image segmentation, (3) quantification of structure and function, and (4) disease detection. Using 14 035 echocardiograms spanning a 10-year period, we trained and evaluated convolutional neural network models for multiple tasks, including automated identification of 23 viewpoints and segmentation of cardiac chambers across 5 common views. The segmentation output was used to quantify chamber volumes and left ventricular mass, determine ejection fraction, and facilitate automated determination of longitudinal strain through speckle tracking. Results were evaluated through comparison to manual segmentation and measurements from 8666 echocardiograms obtained during the routine clinical workflow. Finally, we developed models to detect 3 diseases: hypertrophic cardiomyopathy, cardiac amyloid, and pulmonary arterial hypertension. Convolutional neural networks accurately identified views (eg, 96% for parasternal long axis), including flagging partially obscured cardiac chambers, and enabled the segmentation of individual cardiac chambers. The resulting cardiac structure measurements agreed with study report values (eg, median absolute deviations of 15% to 17% of observed values for left ventricular mass, left ventricular diastolic volume, and left atrial volume). In terms of function, we computed automated ejection fraction and longitudinal strain measurements (within 2 cohorts), which agreed with commercial software-derived values (for ejection fraction, median absolute deviation=9.7% of observed, N=6407 studies; for strain, median absolute deviation=7.5%, n=419, and 9.0%, n=110) and demonstrated applicability to serial monitoring of patients with breast cancer for trastuzumab cardiotoxicity. Overall, we found automated measurements to be comparable or superior to manual measurements across 11 internal consistency metrics (eg, the correlation of left atrial and ventricular volumes). Finally, we trained convolutional neural networks to detect hypertrophic cardiomyopathy, cardiac amyloidosis, and pulmonary arterial hypertension with C statistics of 0.93, 0.87, and 0.85, respectively. Our pipeline lays the groundwork for using automated interpretation to support serial patient tracking and scalable analysis of millions of echocardiograms archived within healthcare systems.

Heart Chamber Segmentation from CT Using Convolutional Neural Networks.

CT is routinely used for radiotherapy planning with organs and regions of interest being segmented for diagnostic evaluation and parameter optimization. For cardiac segmentation, many methods have been proposed for left ventricular segmentation, but few for simultaneous segmentation of the entire heart. In this work, we present a convolutional neural networks (CNN)-based cardiac chamber segmentation method for 3D CT with 5 classes: left ventricle, right ventricle, left atrium, right atrium, and background. We achieved an overall accuracy of 87.2% ± 3.3% and an overall chamber accuracy of 85.6 ± 6.1%. The deep learning based segmentation method may provide an automatic tool for cardiac segmentation on CT images.

Validation of a novel CARTOSEG™ segmentation module software for contrast-enhanced computed tomography-guided radiofrequency ablation in patients with atrial fibrillation.

Visualization of left atrial (LA) anatomy using image integration modules has been associated with decreased radiation exposure and improved procedural outcome when used for guidance of pulmonary vein isolation (PVI) in atrial fibrillation (AF) ablation. We evaluated the CARTOSEG™ CT Segmentation Module (Biosense Webster, Inc.) that offers a new CT-specific semiautomatic reconstruction of the atrial endocardium. The CARTOSEG™ CT Segmentation Module software was assessed prospectively in 80 patients undergoing AF ablation. Using preprocedural contrast-enhanced computed tomography (CE-CT), cardiac chambers, coronary sinus (CS), and esophagus were semiautomatically segmented. Segmentation quality was assessed from 1 (poor) to 4 (excellent). The reconstructed structures were registered with the electroanatomic map (EAM). PVI was performed using the registered 3D images. Semiautomatic reconstruction of the heart chambers was successfully performed in all 80 patients with AF. CE-CT DICOM file import, semiautomatic segmentation of cardiac chambers, esophagus, and CS was performed in 185 ± 105, 18 ± 5, 119 ± 47, and 69 ± 19 seconds, respectively. Average segmentation quality was 3.9 ± 0.2, 3.8 ± 0.3, and 3.8 ± 0.2 for LA, esophagus, and CS, respectively. Registration accuracy between the EAM and CE-CT-derived segmentation was 4.2 ± 0.9 mm. Complications consisted of one perforation (1%) which required pericardiocentesis, one increased pericardial effusion treated conservatively (1%), and one early termination of ablation due to thrombus formation on the ablation sheath without TIA/stroke (1%). All targeted PVs (n=309) were successfully isolated. The novel CT- CARTOSEG™ CT Segmentation Module enables a rapid and reliable semiautomatic 3D reconstruction of cardiac chambers and adjacent anatomy, which facilitates successful and safe PVI.

Evaluation of a multi‐atlas based method for segmentation of cardiac CTA data: a large‐scale, multicenter, and multivendor study

Computed tomography angiography (CTA) is increasingly used for the diagnosis of coronary artery disease (CAD). However, CTA is not commonly used for the assessment of ventricular and atrial function, although functional information extracted from CTA data is expected to improve the diagnostic value of the examination. In clinical practice, the extraction of ventricular and atrial functional information, such as stroke volume and ejection fraction, requires accurate delineation of cardiac chambers. In this paper, we investigated the accuracy and robustness of cardiac chamber delineation using a multiatlas based segmentation method on multicenter and multivendor CTA data. A fully automatic multiatlas based method for segmenting the whole heart (i.e., the outer surface of the pericardium) and cardiac chambers from CTA data is presented and evaluated. In the segmentation approach, eight atlas images are registered to a new patient's CTA scan. The eight corresponding manually labeled images are then propagated and combined using a per voxel majority voting procedure, to obtain a cardiac segmentation. The method was evaluated on a multicenter/multivendor database, consisting of (1) a set of 1380 Siemens scans from 795 patients and (2) a set of 60 multivendor scans (Siemens, Philips, and GE) from different patients, acquired in six different institutions worldwide. A leave-one-out 3D quantitative validation was carried out on the eight atlas images; we obtained a mean surface-to-surface error of 0.94 +/- 1.12 mm and an average Dice coefficient of 0.93 was achieved. A 2D quantitative evaluation was performed on the 60 multivendor data sets. Here, we observed a mean surface-to-surface error of 1.26 +/- 1.25 mm and an average Dice coefficient of 0.91 was achieved. In addition to this quantitative evaluation, a large-scale 2D and 3D qualitative evaluation was performed on 1380 and 140 images, respectively. Experts evaluated that 49% of the 1380 images were very accurately segmented (below 1 mm error) and that 29% were accurately segmented (error between 1 and 3 mm), which demonstrates the robustness of the presented method. A fully automatic method for whole heart and cardiac chamber segmentation was presented and evaluated using multicenter/multivendor CTA data. The accuracy and robustness of the method were demonstrated by successfully applying the method to 1420 multicenter/ multivendor data sets.