Cardiac MRI assessment of right ventricular function in acquired heart disease: factors of variability

Data-driven computational models of ventricular-arterial hemodynamics in pediatric pulmonary arterial hypertension

Pulmonary arterial hypertension (PAH) is a complex disease involving increased resistance in the pulmonary arteries and subsequent right ventricular (RV) remodeling. Ventricular-arterial interactions are fundamental to PAH pathophysiology but are rarely captured in computational models. It is important to identify metrics that capture and quantify these interactions to inform our understanding of this disease as well as potentially facilitate patient stratification. Towards this end, we developed and calibrated two multi-scale high-resolution closed-loop computational models using open-source software: a high-resolution arterial model implemented using CRIMSON, and a high-resolution ventricular model implemented using FEniCS. Models were constructed with clinical data including non-invasive imaging and invasive hemodynamic measurements from a cohort of pediatric PAH patients. A contribution of this work is the discussion of inconsistencies in anatomical and hemodynamic data routinely acquired in PAH patients. We proposed and implemented strategies to mitigate these inconsistencies, and subsequently use this data to inform and calibrate computational models of the ventricles and large arteries. Computational models based on adjusted clinical data were calibrated until the simulated results for the high-resolution arterial models matched within 10% of adjusted data consisting of pressure and flow, whereas the high-resolution ventricular models were calibrated until simulation results matched adjusted data of volume and pressure waveforms within 10%. A statistical analysis was performed to correlate numerous data-derived and model-derived metrics with clinically assessed disease severity. Several model-derived metrics were strongly correlated with clinically assessed disease severity, suggesting that computational models may aid in assessing PAH severity.

A U-snake based deep learning network for right ventricle segmentation

PURPOSE

Ventricular segmentation is of great importance for the heart condition monitoring. However, manual segmentation is time-consuming, cumbersome and subjective. Many segmentation methods perform poorly due to the complex structure and uncertain shape of the right ventricle, so we combine deep learning to achieve automatic segmentation.

METHOD

This paper proposed a method named U-Snake network which is based on the improvement of deep snake⁵ together with level set⁸ to segment the right ventricular in the MR images. U-snake aggregates the information of each receptive field which is learned by circular convolution of multiple different dilation rates. At the same time, we also added dice loss functions and transferred the result of U-Snake to the level set so as to further enhance the effect of small object segmentation. our method is tested on the test1 and test2 datasets in the Right Ventricular Segmentation Challenge, which shows the effectiveness.

RESULTS

The experiment showed that we have obtained good result in the right ventricle segmentation challenge(RVSC). The highest segmentation accuracy on the right ventricular test set 2 reached a dice coefficient of 0.911, and the segmentation speed reached 5fps.

CONCLUSIONS

Our method, a new deep learning network named U-snake, has surpassed the previous excellent ventricular segmentation method based on mathematical theory and other classical deep learning methods, such as Residual U-net²⁷ , Inception cnn³³ , Dilated cnn²⁹ , etc. However, it can only be used as an auxiliary tool instead of replacing the work of human beings. This article is protected by copyright. All rights reserved.

TSU-net: Two-stage multi-scale cascade and multi-field fusion U-net for right ventricular segmentation

Accurate segmentation of the right ventricle from cardiac magnetic resonance images (MRI) is a critical step in cardiac function analysis and disease diagnosis. It is still an open problem due to some difficulties, such as a large variety of object sizes and ill-defined borders. In this paper, we present a TSU-net network that grips deeper features and captures targets of different sizes with multi-scale cascade and multi-field fusion in the right ventricle. TSU-net mainly contains two major components: Dilated-Convolution Block (DB) and Multi-Layer-Pool Block (MB). DB extracts and aggregates multi-scale features for the right ventricle. MB mainly relies on multiple effective field-of-views to detect objects at different sizes and fill boundary features. Different from previous networks, we used DB and MB to replace the convolution layer in the encoding layer, thus, we can gather multi-scale information of right ventricle, detect different size targets and fill boundary information in each encoding layer. In addition, in the decoding layer, we used DB to replace the convolution layer, so that we can aggregate the multi-scale features of the right ventricle in each decoding layer. Furthermore, the two-stage U-net structure is used to further improve the utilization of DB and MB through a two-layer encoding/decoding layer. Our method is validated on the RVSC, a public right ventricular data set. The results demonstrated that TSU-net achieved an average Dice coefficient of 0.86 on endocardium and 0.90 on the epicardium, thereby outperforming other models. It effectively assists doctors to diagnose the disease and promotes the development of medical images. In addition, we also provide an intuitive explanation of our network, which fully explain MB and TSU-net's ability to detect targets of different sizes and fill in boundary features.

A cascaded FC-DenseNet and level set method (FCDL) for fully automatic segmentation of the right ventricle in cardiac MRI

Accurate segmentation of the right ventricle (RV) from cardiac magnetic resonance imaging (MRI) images is an essential step in estimating clinical indices such as stroke volume and ejection fraction. Recently, image segmentation methods based on fully convolutional neural networks (FCN) have drawn much attention and shown promising results. In this paper, a new fully automatic RV segmentation method combining the FC-DenseNet and the level set method (FCDL) is proposed. The FC-DenseNet is efficiently trained end-to-end, using RV images and ground truth masks to make a per-pixel semantic inference. As a result, probability images are produced, followed by the level set method responsible for smoothing and converging contours to improve accuracy. It is noted that the iteration times of the level set method is only 4 times, which is due to the semantic segmentation of the FC-DenseNet for RV. Finally, multi-object detection algorithm is applied to locate the RV. Experimental results (including 45 cases, 15 cases for training, 30 cases for testing) show that the FCDL method outperforms the U-net + level set (UL) and the level set methods that use the same dataset and the cardiac functional parameters are computed robustly by the FCDL method. The results validate the FCDL method as an efficient and satisfactory approach to RV segmentation.

Analysis of right ventricular mass from magnetic resonance imaging data: a simple post-processing algorithm for correction of partial-volume effects

Magnetic resonance imaging (MRI) of the right ventricle (RV) offers important diagnostic information, but the accuracy of this information is hampered by the complex geometry of the RV. Here, we propose a novel postprocessing algorithm that corrects for partial-volume effects in the analysis of standard MRI cine images of RV mass (RVm) and evaluate the method in clinical and preclinical data. Self-corrected RVm measurement was compared with conventionally measured RVm in 16 patients who showed different clinical indications for cardiac MRI and in 17 Wistar rats with different degrees of pulmonary congestion. The rats were studied under isoflurane anaesthesia. To evaluate the reliability of the proposed method, the measured end-systolic and end-diastolic RVm were compared. Accuracy was evaluated by comparing preclinical RVm to ex vivo RV weight (RVw). We found that use of the self-correcting algorithm improved reliability compared with conventional segmentation. For clinical data, the limits of agreement (LOAs) were -1.8 ± 8.6g (self-correcting) vs. 5.8 ± 7.8g (conventional), and coefficients of variation (CoVs) were 7.0% (self-correcting) vs. 14.3% (conventional). For preclinical data, LOAs were 21 ± 46 mg (self-correcting) vs. 64 ± 89 mg (conventional), and CoVs were 9.0% (self-correcting) and 17.4% (conventional). Self-corrected RVm also showed better correspondence with the ex vivo RVw: LOAs were -5 ± 80 mg (self-correcting) vs. 94 ± 116 mg (conventional) in end-diastole and -26 ± 74 mg (self-correcting) vs. 31 ± 98 mg (conventional) in end-systole. The new self-correcting algorithm improves the reliability and accuracy of RVm measurements in both clinical and preclinical MRI. It is simple and easy to implement and does not require any additional MRI data.NEW & NOTEWORTHY Magnetic resonance imaging (MRI) of the right ventricle (RV) offers important diagnostic information, but the accuracy of this information is hampered by the complex geometry of the RV. In particular, the crescent shape of the RV renders it particularly vulnerable to partial-volume effects. We present a new, simple, self-correcting algorithm that can be applied to correct partial-volume effects in MRI-based RV mass estimation. The self-correcting algorithm offers improved reliability and accuracy compared with the conventional approach.

Mini Review: Deep Learning for Atrial Segmentation From Late Gadolinium-Enhanced MRIs

Segmentation and 3D reconstruction of the human atria is of crucial importance for precise diagnosis and treatment of atrial fibrillation, the most common cardiac arrhythmia. However, the current manual segmentation of the atria from medical images is a time-consuming, labor-intensive, and error-prone process. The recent emergence of artificial intelligence, particularly deep learning, provides an alternative solution to the traditional methods that fail to accurately segment atrial structures from clinical images. This has been illustrated during the recent 2018 Atrial Segmentation Challenge for which most of the challengers developed deep learning approaches for atrial segmentation, reaching high accuracy (>90% Dice score). However, as significant discrepancies exist between the approaches developed, many important questions remain unanswered, such as which deep learning architectures and methods to ensure reliability while achieving the best performance. In this paper, we conduct an in-depth review of the current state-of-the-art of deep learning approaches for atrial segmentation from late gadolinium-enhanced MRIs, and provide critical insights for overcoming the main hindrances faced in this task.

Right Ventricular Segmentation from MRI Using Deep Convolutional Neural Networks

The assessment of right ventricular (RV) function is essential in the diagnosis of many cardiac diseases. Magnetic resonance imaging (MRI) offers an excellent solution to image right ventricle non-invasively with high contrast and temporal resolution. Manual assessment of the RV function from MRI sequences is tedious and time-consuming and automating the process is of great interest. This study proposes a convolutional neural network-based machine learning approach to automate the delineation of the RV from a sequence of MRI. The architecture of the neural network differs from that of a widely-known U-Net approach. Additionally, the proposed approach used image concatenation to create and utilize 3D spatial information in the segmentation process. Quantitative evaluations were performed over 256 images acquired from 16 patients in publicly available data in comparison to manual delineations. Comparisons with the results by U-Net demonstrated that the proposed method outperforms the prior state-of-the-art method.

Evaluation of a Semi-automatic Right Ventricle Segmentation Method on Short-Axis MR Images

The purpose of this study was to evaluate a semi-automatic right ventricle segmentation method on short-axis cardiac cine MR images which segment all right ventricle contours in a cardiac phase using one seed contour. Twenty-eight consecutive short-axis, four-chamber, and tricuspid valve view cardiac cine MRI examinations of healthy volunteers were used. Two independent observers performed the manual and automatic segmentations of the right ventricles. Analyses were based on the ventricular volume and ejection fraction of the right heart chamber. Reproducibility of the manual and semi-automatic segmentations was assessed using intra- and inter-observer variability. Validity of the semi-automatic segmentations was analyzed with reference to the manual segmentations. The inter- and intra-observer variability of manual segmentations were between 0.8 and 3.2%. The semi-automatic segmentations were highly correlated with the manual segmentations (R² 0.79-0.98), with median difference of 0.9-4.8% and of 3.3% for volume and ejection fraction parameters, respectively. In comparison to the manual segmentation, the semi-automatic segmentation produced contours with median dice metrics of 0.95 and 0.87 and median Hausdorff distance of 5.05 and 7.35 mm for contours at end-diastolic and end-systolic phases, respectively. The inter- and intra-observer variability of the semi-automatic segmentations were lower than observed in the manual segmentations. Both manual and semi-automatic segmentations performed better at the end-diastolic phase than at the end-systolic phase. The investigated semi-automatic segmentation method managed to produce a valid and reproducible alternative to manual right ventricle segmentation.

Dilated-Inception Net: Multi-Scale Feature Aggregation for Cardiac Right Ventricle Segmentation

Segmentation of cardiac ventricle from magnetic resonance images is significant for cardiac disease diagnosis, progression assessment, and monitoring cardiac conditions. Manual segmentation is so time consuming, tedious, and subjective that automated segmentation methods are highly desired in practice. However, conventional segmentation methods performed poorly in cardiac ventricle, especially in the right ventricle. Compared with the left ventricle, whose shape is a simple thick-walled circle, the structure of the right ventricle is more complex due to ambiguous boundary, irregular cavity, and variable crescent shape. Hence, effective feature extractors and segmentation models are preferred. In this paper, we propose a dilated-inception net (DIN) to extract and aggregate multi-scale features for right ventricle segmentation. The DIN outperforms many state-of-the-art models on the benchmark database of right ventricle segmentation challenge. In addition, the experimental results indicate that the proposed model has potential to reach expert-level performance in right ventricular epicardium segmentation. More importantly, DIN behaves similarly to clinical expert with high correlation coefficients in four clinical cardiac indices. Therefore, the proposed DIN is promising for automated cardiac right ventricle segmentation in clinical applications.

Improved Workflow for Quantification of Right Ventricular Volumes Using Free-Breathing Motion Corrected Cine Imaging

Cardiac MR traditionally requires breath-holding for cine imaging. Younger or less stable patients benefit from free-breathing during cardiac MR but current free-breathing cine images can be spatially blurred. Motion corrected re-binning (MOC) is a novel approach that acquires and then reformats real-time images over multiple cardiac cycles with high spatial resolution. The technique was previously limited by reconstruction time but distributed computing has reduced these times. Using this technique, left ventricular volumetry has compared favorably to breath-held balanced steady-state free precession cine imaging (BH), the current gold-standard, however, right ventricular volumetry validation remains incomplete, limiting the applicability of MOC in clinical practice. Fifty subjects underwent cardiac MR for evaluation of right ventricular size and function by end-diastolic (EDV) and end-systolic (ESV) volumetry. Measurements using MOC were compared to those using BH. Pearson correlation coefficients and Bland-Altman plots tested agreement across techniques. Total scan plus reconstruction times were tested for significant differences using paired t-test. Volumes obtained by MOC compared favorably to BH (R = 0.9911 for EDV, 0.9690 for ESV). Combined acquisition and reconstruction time (previously reported) were reduced 37% for MOC, requiring a mean of 5.2 min compared to 8.2 min for BH (p < 0.0001). Right ventricular volumetry compares favorably to BH using MOC image reconstruction, but is obtained in a fraction of the time. Combined with previous validation of its use for the left ventricle, this novel method now offers an alternative imaging approach in appropriate clinical settings.

On the right side of the heart: Medical and mechanical support of the failing right ventricle

Acute right ventricular failure remains an immensely challenging clinical condition associated with a high mortality rate. In this narrative review, we highlight the pathophysiological mechanisms underlying right ventricular failure and suggest an initial diagnostic approach to this critically ill group of patients. Based on the best available evidence and our cumulative clinical experience as a national cardiothoracic centre, we summarize the basic principles of medical management and mechanical salvage therapy, finalizing with a series of recommendations for the management of right ventricular failure in specific clinical scenarios.

Automatic segmentation of right ventricle in cardiac cine MR images using a saliency analysis

PURPOSE

Accurate measurement of the right ventricle (RV) volume is important for the assessment of the ventricular function and a biomarker of the progression of any cardiovascular disease. However, the high RV variability makes difficult a proper delineation of the myocardium wall. This paper introduces a new automatic method for segmenting the RV volume from short axis cardiac magnetic resonance (MR) images by a salient analysis of temporal and spatial observations.

METHODS

The RV volume estimation starts by localizing the heart as the region with the most coherent motion during the cardiac cycle. Afterward, the ventricular chambers are identified at the basal level using the isodata algorithm, the right ventricle extracted, and its centroid computed. A series of radial intensity profiles, traced from this centroid, is used to search a salient intensity pattern that models the inner-outer myocardium boundary. This process is iteratively applied toward the apex, using the segmentation of the previous slice as a regularizer. The consecutive 2D segmentations are added together to obtain the final RV endocardium volume that serves to estimate also the epicardium.

RESULTS

Experiments performed with a public dataset, provided by the RV segmentation challenge in cardiac MRI, demonstrated that this method is highly competitive with respect to the state of the art, obtaining a Dice score of 0.87, and a Hausdorff distance of 7.26 mm while a whole volume was segmented in about 3 s.

CONCLUSIONS

The proposed method provides an useful delineation of the RV shape using only the spatial and temporal information of the cine MR images. This methodology may be used by the expert to achieve cardiac indicators of the right ventricle function.

Influence of the short-axis cine acquisition protocol on the cardiac function evaluation: A reproducibility study

•

Evaluate several scan protocols for cardiac MRI analysis.

•

Assess inter and intra observer variability for the selection of the best protocol.

•

Consistency in the scan protocol is essential for study comparison.

•

For biventricular studies, short-axis slices should be perpendicular to septum.

Purpose

To define the optimal cardiac short-axis cine acquisition protocol for the assessment of the left and rightventricular functions.

Materials and methods

20 volunteers were recruited and breath-hold CINE images were acquired on a Siemens Prisma 3T MRI. Four short-axis acquisition planes were defined from the 4-chamber view. AV Junctions: short-axis slices parallel to the plane that cuts through the external right and left atrioventricular junctions. Left AV Junctions: short-axis slices parallel to the plane that cuts through both left atrioventricular junctions. Septum: short-axis slices perpendicular to the septum with one cutting through the septum junction. LongAxis: short-axis slices perpendicular to the long axis with one cutting through the septum junction. Intra and inter reproducibility was assessed using Bland-Altman coefficient of variation (CV) and Lin’s concordance correlation coefficient (CCC). The influence of the protocol on the ejection fraction (EF) and stroke volume (SV) was quantified statistically using pair-wise CV and Pearson’s correlation coefficient R².

Results

All protocols led to high reproducibility for the LV EF (mean intra CV = 3.83%, mean inter CV = 4.81%, lowest CV = 4.20% (AV junctions) and highest CV = 5.24% (Left AV Junctions)). Reproducibility of the RV measurements was lower (mean intra CV = 7.84%, mean inter CV = 9.17%). Septum protocol led to significantly lower variability compared to the other 3 protocols for RV EF (CV = 7.62% (Septum), CV = 8.42% (Long Axis), CV = 9.54% (Left AV Junctions) and CV = 11.08% (AV Junctions) with Lin’s CCC varying from 0.4 (AV Junctions) to 0.69 (Septum) for inter-observer reproducibility). No differences in group average for clinical parameters was found for both LV and RV clinical measurements. However, patient-specific RV EF evaluation is dependent on the chosen protocol (CV = 9.95%, R² = 0.52).

Conclusion

Based on the results of the study cine mode short-axis acquisitions should be planned perpendicular to the septum in order to guarantee optimal RV and LV measurements.

An Algorithm for the Segmentation of Highly Abnormal Hearts Using a Generic Statistical Shape Model

Statistical shape models (SSMs) have been widely employed in cardiac image segmentation. However, in conditions that induce severe shape abnormality and remodeling, such as in the case of pulmonary hypertension (PH) or hypertrophic cardiomyopathy (HCM), a single SSM is rarely capable of capturing the anatomical variability in the extremes of the distribution. This work presents a new algorithm for the segmentation of severely abnormal hearts. The algorithm is highly flexible, as it does not require a priori knowledge of the involved pathology or any specific parameter tuning to be applied to the cardiac image under analysis. The fundamental idea is to approximate the gross effect of the abnormality with a virtual remodeling transformation between the patient-specific geometry and the average shape of the reference model (e.g., average normal morphology). To define this mapping, a set of landmark points are automatically identified during boundary point search, by estimating the reliability of the candidate points. With the obtained transformation, the feature points extracted from the patient image volume are then projected onto the space of the reference SSM, where the model is used to effectively constrain and guide the segmentation process. The extracted shape in the reference space is finally propagated back to the original image of the abnormal heart to obtain the final segmentation. Detailed validation with patients diagnosed with PH and HCM shows the robustness and flexibility of the technique for the segmentation of highly abnormal hearts of different pathologies.

Cardiac magnetic resonance findings predicting mortality in patients with pulmonary arterial hypertension: a systematic review and meta-analysis

Objectives

To provide a comprehensive overview of all reported cardiac magnetic resonance (CMR) findings that predict clinical deterioration in pulmonary arterial hypertension (PAH).

Methods

MEDLINE and EMBASE electronic databases were systematically searched for longitudinal studies published by April 2015 that reported associations between CMR findings and adverse clinical outcome in PAH. Studies were appraised using previously developed criteria for prognostic studies. Meta-analysis using random effect models was performed for CMR findings investigated by three or more studies.

Results

Eight papers (539 patients) investigating 21 different CMR findings were included. Meta-analysis showed that right ventricular (RV) ejection fraction was the strongest predictor of mortality in PAH (pooled HR 1.23 [95 % CI 1.07–1.41], p = 0.003) per 5 % decrease. In addition, RV end-diastolic volume index (pooled HR 1.06 [95 % CI 1.00–1.12], p = 0.049), RV end-systolic volume index (pooled HR 1.05 [95 % CI 1.01–1.09], p = 0.013) and left ventricular end-diastolic volume index (pooled HR 1.16 [95 % CI 1.00–1.34], p = 0.045) were of prognostic importance. RV and LV mass did not provide prognostic information (p = 0.852 and p = 0.983, respectively).

Conclusion

This meta-analysis substantiates the clinical yield of specific CMR findings in the prognostication of PAH patients. Decreased RV ejection is the strongest and most well established predictor of mortality.

Key Points

• Cardiac magnetic resonance imaging is useful for prognostication in pulmonary arterial hypertension.

• Right ventricular ejection fraction is the strongest predictor of mortality.

• Serial CMR evaluation seems to be of additional prognostic importance.

• Accurate prognostication can aid in adequate and timely intensification of PAH-specific therapy.

Electronic supplementary material

The online version of this article (doi:10.1007/s00330-016-4217-6) contains supplementary material, which is available to authorized users.

A GPU accelerated moving mesh correspondence algorithm with applications to RV segmentation

This study proposes a parallel nonrigid registration algorithm to obtain point correspondence between a sequence of images. Several recent studies have shown that computation of point correspondence is an excellent way to delineate organs from a sequence of images, for example, delineation of cardiac right ventricle (RV) from a series of magnetic resonance (MR) images. However, nonrigid registration algorithms involve optimization of similarity functions, and are therefore, computationally expensive. We propose Graphics Processing Unit (GPU) computing to accelerate the algorithm. The proposed approach consists of two parallelization components: 1) parallel Compute Unified Device Architecture (CUDA) version of the non-rigid registration algorithm; and 2) application of an image concatenation approach to further parallelize the algorithm. The proposed approach was evaluated over a data set of 16 subjects and took an average of 4.36 seconds to segment a sequence of 19 MR images, a significant performance improvement over serial image registration approach.

RV mass measurement at end-systole: Improved accuracy, Reproducibility, and reduced segmentation time

PURPOSE

To evaluate the accuracy, reproducibility, and contouring time of RV mass in end-systole (ES) and end-diastole (ED). Magnetic resonance imaging (MRI) has been shown to be accurate and reproducible for the evaluation of right ventricular (RV) volume and function. RV mass, assessed in end-diastolic (ED) phase, is one of the least reproducible variables. The choice of end-systolic (ES) phase could offer an alternative to improve reproducibility, since the selection of the basal slice and the visualization of the usually thin RV wall are easier in this phase.

MATERIALS AND METHODS

To evaluate accuracy, 11 sheep were imaged in vivo and their RV free walls were weighed after removing epicardial fat. To evaluate reproducibility, 30 normal subjects and 30 subjects with pulmonary arterial hypertension (PAH) were imaged and interobserver and intraobserver variabilities were assessed in the ES and the ED. Segmentation time was recorded after visual selection of ES and ED phases.

RESULTS

ES RV mass measurement has less absolute variability (5.2% ± 3.2) compared to ED (10.6% ± 6.3) using weighed RV mass in sheep as the gold standard (P < 0.001). ES segmentation yielded higher intraobserver (intraclass correlation coefficients [ICC] = 0.94-0.99; coefficient of variability [CoV] = 6-7.3%) and interobserver (ICC = 0.85-0.98; CoV = 10.9-11.7%) reproducibility than ED segmentation. Segmentation time in humans was 25-28% faster in ES (P < 0.001).

CONCLUSION

The MRI assessment of RV mass is more accurate, reproducible, and faster in the ES phase.

Right ventricular segmentation in cardiac MRI with moving mesh correspondences

This study investigates automatic propagation of the right ventricle (RV) endocardial and epicardial boundaries in 4D (3D+time) magnetic resonance imaging (MRI) sequences. Based on a moving mesh (or grid generation) framework, the proposed algorithm detects the endocardium and epicardium within each cardiac phase via point-to-point correspondences. The proposed method has the following advantages over prior RV segmentation works: (1) it removes the need for a time-consuming, manually built training set; (2) it does not make prior assumptions as to the intensity distributions or shape; (3) it provides a sequence of corresponding points over time, a comprehensive input that can be very useful in cardiac applications other than segmentation, e.g., regional wall motion analysis; and (4) it is more flexible for congenital heart disease where the RV undergoes high variations in shape. Furthermore, the proposed method allows comprehensive RV volumetric analysis over the complete cardiac cycle as well as automatic detections of end-systolic and end-diastolic phases because it provides a segmentation for each time step. Evaluated quantitatively over the 48-subject data set of the MICCAI 2012 RV segmentation challenge, the proposed method yielded an average Dice score of 0.84±0.11 for the epicardium and 0.79±0.17 for the endocardium. Further, quantitative evaluations of the proposed approach in comparisons to manual contours over 23 infant hypoplastic left heart syndrome patients yielded a Dice score of 0.82±0.14, which demonstrates the robustness of the algorithm.

Left ventricle: fully automated segmentation based on spatiotemporal continuity and myocardium information in cine cardiac magnetic resonance imaging (LV-FAST)

CMR quantification of LV chamber volumes typically and manually defines the basal-most LV, which adds processing time and user-dependence. This study developed an LV segmentation method that is fully automated based on the spatiotemporal continuity of the LV (LV-FAST). An iteratively decreasing threshold region growing approach was used first from the midventricle to the apex, until the LV area and shape discontinued, and then from midventricle to the base, until less than 50% of the myocardium circumference was observable. Region growth was constrained by LV spatiotemporal continuity to improve robustness of apical and basal segmentations. The LV-FAST method was compared with manual tracing on cardiac cine MRI data of 45 consecutive patients. Of the 45 patients, LV-FAST and manual selection identified the same apical slices at both ED and ES and the same basal slices at both ED and ES in 38, 38, 38, and 41 cases, respectively, and their measurements agreed within −1.6 ± 8.7 mL, −1.4 ± 7.8 mL, and 1.0 ± 5.8% for EDV, ESV, and EF, respectively. LV-FAST allowed LV volume-time course quantitatively measured within 3 seconds on a standard desktop computer, which is fast and accurate for processing the cine volumetric cardiac MRI data, and enables LV filling course quantification over the cardiac cycle.

Data-driven shape parameterization for segmentation of the right ventricle from 3D+t echocardiography

Model-based segmentation facilitates the accurate measurement of geometric properties of anatomy from ultrasound images. Regularization of the model surface is typically necessary due to the presence of noisy and incomplete boundaries. When simple regularizers are insufficient, linear basis shape models have been shown to be effective. However, for problems such as right ventricle (RV) segmentation from 3D+t echocardiography, where dense consistent landmarks and complete boundaries are absent, acquiring accurate training surfaces in dense correspondence is difficult. As a solution, this paper presents a framework which performs joint segmentation of multiple 3D+t sequences while simultaneously optimizing an underlying linear basis shape model. In particular, the RV is represented as an explicit continuous surface, and segmentation of all frames is formulated as a single continuous energy minimization problem. Shape information is automatically shared between frames, missing boundaries are implicitly handled, and only coarse surface initializations are necessary. The framework is demonstrated to successfully segment both multiple-view and multiple-subject collections of 3D+t echocardiography sequences, and the results confirm that the linear basis shape model is an effective model constraint. Furthermore, the framework is shown to achieve smaller segmentation errors than a state-of-art commercial semi-automatic RV segmentation package.

Evaluation of knowledge-based reconstruction for magnetic resonance volumetry of the right ventricle in tetralogy of Fallot

BACKGROUND

Cardiac magnetic resonance using the Simpson method is the gold standard for right ventricular volumetry. However, this method is time-consuming and not without sources of error. Knowledge-based reconstruction is a novel post-processing approach that reconstructs the right ventricular endocardial shape based on anatomical landmarks and a database of various right ventricular configurations.

OBJECTIVE

To assess the feasibility, accuracy and labor intensity of knowledge-based reconstruction in repaired tetralogy of Fallot (TOF).

MATERIALS AND METHODS

The short-axis cine cardiac MR datasets of 35 children and young adults (mean age 14.4 ± 2.5 years) after TOF repair were studied using both knowledge-based reconstruction and the Simpson method. Intraobserver, interobserver and inter-method variability were assessed using Bland-Altman analyses.

RESULTS

Knowledge-based reconstruction was feasible and highly accurate as compared to the Simpson method. Intra- and inter-method variability for knowledge-based reconstruction measurements showed good agreement. Volumetric assessment using knowledge-based reconstruction was faster when compared with the Simpson method (10.9 ± 2.0 vs. 7.1 ± 2.4 min, P < 0.001).

CONCLUSION

In patients with repaired tetralogy of Fallot, knowledge-based reconstruction is a feasible, accurate and reproducible method for measuring right ventricular volumes and ejection fraction. The post-processing time of right ventricular volumetry using knowledge-based reconstruction was significantly shorter when compared with the routine Simpson method.

Assessment of LV ejection fraction using real-time 3D echocardiography in daily practice: direct comparison of the volumetric and speckle tracking methodologies to CMR

AIMS

This study is the first to directly compare two widely used real-time 3D echocardiography (RT3DE) methods of cardiac magnetic resonance imaging (CMR) and assess their reproducibility in experienced and less experienced observers.

METHODS

Consecutive patients planned for CMR underwent RT3DE within 8 h of CMR with Philips (volumetric method) and Toshiba Artida (speckle tracking method). Left ventricular ejection fraction (LVEF), left ventricular end-diastolic volume (LVEDV) and end-systolic volume (LVESV) were measured using RT3DE, by four trained observers, and compared with CMR values.

RESULTS

Thirty-five patients were included (49.7 ± 15.7 years; 55 % male), 30 (85.7 %) volumetric and 27 (77.1 %) speckle tracking datasets could be analysed. CMR derived LVEDV, LVESV and LVEF were 198 ± 58 ml, 106 ± 53 ml and 49 ± 15 %, respectively. LVEF derived from speckle tracking was accurate and reproducible in all observers (all intra-class correlation coefficients (ICC) > 0.86). LVEF derived from the volumetric method correlated well to CMR in experienced observers (ICC 0.85 and 0.86) but only moderately in less experienced observers (ICC 0.58 and 0.77) and was less reproducible in these observers (ICC = 0.55). Volumes were significantly underestimated compared with CMR (p < 0.001).

CONCLUSION

This study demonstrates that both RT3DE methodologies are sufficiently accurate and reproducible for use in daily practice. However, experience importantly influences the accuracy and reproducibility of the volumetric method, which should be considered when introducing this technique into clinical practice.

Fast, accurate, and fully automatic segmentation of the right ventricle in short-axis cardiac MRI

This paper presents a fully automatic method to segment the right ventricle (RV) from short-axis cardiac MRI. A combination of a novel window-constrained accumulator thresholding technique, binary difference of Gaussian (DoG) filters, optimal thresholding, and morphology are utilized to drive the segmentation. A priori segmentation window constraints are incorporated to guide and refine the process, as well as to ensure appropriate area confinement of the segmentation. Training and testing were performed using a combined 48 patient datasets supplied by the organizers of the MICCAI 2012 right ventricle segmentation challenge, allowing for unbiased evaluations and benchmark comparisons. Marked improvements in speed and accuracy over the top existing methods are demonstrated.

A probabilistic patch-based label fusion model for multi-atlas segmentation with registration refinement: application to cardiac MR images

The evaluation of ventricular function is important for the diagnosis of cardiovascular diseases. It typically involves measurement of the left ventricular (LV) mass and LV cavity volume. Manual delineation of the myocardial contours is time-consuming and dependent on the subjective experience of the expert observer. In this paper, a multi-atlas method is proposed for cardiac magnetic resonance (MR) image segmentation. The proposed method is novel in two aspects. First, it formulates a patch-based label fusion model in a Bayesian framework. Second, it improves image registration accuracy by utilizing label information, which leads to improvement of segmentation accuracy. The proposed method was evaluated on a cardiac MR image set of 28 subjects. The average Dice overlap metric of our segmentation is 0.92 for the LV cavity, 0.89 for the right ventricular cavity and 0.82 for the myocardium. The results show that the proposed method is able to provide accurate information for clinical diagnosis.

Cardiac right ventricular segmentation via point correspondence

This study presents an approach to the segmentation of the right ventricle (RV) from a sequence of cardiac magnetic resonance (MR) images. Automatic delineation of the RV is difficult because of its complex morphology, thin and ill-defined borders, and the photometric similarities between the connected cardiac regions such as papillary muscles and heart wall. Further, geometric/photometric models are hard to build from a finite training set because of the significant differences in size, shape, and intensity between subjects. In this study, we propose to use a non-rigid registration method developed recently to obtain the point correspondence in a sequence of cine MR images. Given the segmentation on the first frame, the proposed method segments both endocardial and epicardial borders of the RV using the obtained point correspondence, and relaxes the need of a training set. The proposed method is evaluated quantitatively on common data set by comparison with manual segmentation, which demonstrated competitive results in comparison with recent methods.