Segmental longitudinal strain (SLS) of the left ventricle (LV) is an important prognostic indicator for evaluating regional LV dysfunction, in particular for diagnosing and managing myocardial ischemia. Current techniques for strain estimation require significant manual intervention and expertise, limiting their efficiency and making them too resource-intensive for monitoring purposes. This study introduces the first automated pipeline, autoStrain, for SLS estimation in transesophageal echocardiography (TEE) using deep learning (DL) methods for motion estimation. We present a comparative analysis of two DL approaches: TeeFlow, based on the RAFT optical flow model for dense frame-to-frame predictions, and TeeTracker, based on the CoTracker point trajectory model for sparse long-sequence predictions. As ground truth motion data from real echocardiographic sequences are hardly accessible, we took advantage of a unique simulation pipeline (SIMUS) to generate a highly realistic synthetic TEE (synTEE) dataset of 80 patients with ground truth myocardial motion to train and evaluate both models. Our evaluation shows that TeeTracker outperforms TeeFlow in accuracy, achieving a mean distance error in motion estimation of 0.65 $\pm$ 0.20 mm on a synTEE test dataset. Clinical validation on 16 patients further demonstrated that SLS estimation with our autoStrain pipeline aligned with clinical references, achieving a mean difference (95% limits of agreement) of 1.09% (-8.90% to 11.09%). Incorporation of simulated ischemia in the synTEE data improved the accuracy of the models in quantifying abnormal deformation. Our findings indicate that integrating AI-driven motion estimation with TEE can significantly enhance the precision and efficiency of cardiac function assessment in clinical settings.

Medical imaging studies often rely on a single sample per subject, assuming it is representative of their physiological traits. However, variations in how input descriptors are defined or computed (e.g. due to a lack of consensus in the scientific field) may have a crucial impact on the analysis, and are hardly considered in practice. In this paper, we propose an original strategy based on representation learning to estimate a parametric map reflecting the impact of such definitional differences on a given physiological descriptor, previously extracted from medical images. We consider the different definitions or computations of such physiological descriptors as different high-dimensional data, potentially of heterogeneous types. We specifically focus on myocardial deformation (strain), for which there is limited agreement on its definition. We first use manifold alignment to match the latent representations associated with the different definitions of this descriptor. Then, we formulate plausible distributions in the latent space to represent definitional divergence across descriptors, from which we reconstruct a high-dimensional parametric map to visualize such definitional divergence. Due to the lack of proper ground truth for this specific clinical application, we first demonstrate this methodology on toy experiments and then expand the evaluation on right ventricular strain data from subjects obtained from 3D echocardiographic image sequences, for which different types of strain are available at each point of the right ventricle endocardial surface mesh. Beyond this illustrative application, our methodology has the potential to be generalised to many other population analyses considering heterogeneous high-dimensional descriptors.

Domain adaptation methods aim to bridge the gap between datasets by enabling knowledge transfer across domains, reducing the need for additional expert annotations. However, many approaches struggle with reliability in the target domain, an issue particularly critical in medical image segmentation, where accuracy and anatomical validity are essential. This challenge is further exacerbated in spatio-temporal data, where the lack of temporal consistency can significantly degrade segmentation quality, and particularly in echocardiography, where the presence of artifacts and noise can further hinder segmentation performance. To address these issues, we present RL4Seg3D, an unsupervised domain adaptation framework for 2D + time echocardiography segmentation. RL4Seg3D integrates novel reward functions and a fusion scheme to enhance key landmark precision in its segmentations while processing full-sized input videos. By leveraging reinforcement learning for image segmentation, our approach improves accuracy, anatomical validity, and temporal consistency while also providing, as a beneficial side effect, a robust uncertainty estimator, which can be used at test time to further enhance segmentation performance. We demonstrate the effectiveness of our framework on over 30,000 echocardiographic videos, showing that it outperforms standard domain adaptation techniques without the need for any labels on the target domain. Code is available at https://github.com/arnaudjudge/RL4Seg3D.

Cardiac imaging protocols usually result in several types of acquisitions and descriptors extracted from the images. The statistical analysis of such data across a population may be challenging, and can be addressed by fusion techniques within a dimensionality reduction framework. However, directly combining different data types may lead to unfair comparisons (for heterogeneous descriptors) or over-exploitation of information (for strongly correlated modalities). In contrast, physicians progressively consider each type of data based on hierarchies derived from their experience or evidence-based recommendations, an inspiring approach for data fusion strategies. In this paper, we propose a novel methodology for hierarchical data fusion and unsupervised representation learning. It mimics the physicians' approach by progressively integrating different high-dimensional data descriptors according to a known hierarchy. We model this hierarchy with a Hierarchical Gaussian Process Latent Variable Model (GP-LVM), which links the estimated low-dimensional latent representation and high-dimensional observations at each level in the hierarchy, with additional links between consecutive levels of the hierarchy. We demonstrate the relevance of this approach on a dataset of 1726 magnetic resonance image slices from 123 patients revascularized after acute myocardial infarction (MI) (first level in the hierarchy), some of them undergoing reperfusion injury (microvascular obstruction (MVO), second level in the hierarchy). Our experiments demonstrate that our hierarchical model provides consistent data organization across levels of the hierarchy and according to physiological characteristics of the lesions. This allows more relevant statistical analysis of myocardial lesion patterns, and in particular subtle lesions such as MVO.

Convolutional neural networks (CNN) have demonstrated their ability to segment 2D cardiac ultrasound images. However, despite recent successes according to which the intra-observer variability on end-diastole and end-systole images has been reached, CNNs still struggle to leverage temporal information to provide accurate and temporally consistent segmentation maps across the whole cycle. Such consistency is required to accurately describe the cardiac function, a necessary step in diagnosing many cardiovascular diseases. In this paper, we propose a framework to learn the 2D+time apical long-axis cardiac shape such that the segmented sequences can benefit from temporal and anatomical consistency constraints. Our method is a post-processing that takes as input segmented echocardiographic sequences produced by any state-of-the-art method and processes it in two steps to (i) identify spatio-temporal inconsistencies according to the overall dynamics of the cardiac sequence and (ii) correct the inconsistencies. The identification and correction of cardiac inconsistencies relies on a constrained autoencoder trained to learn a physiologically interpretable embedding of cardiac shapes, where we can both detect and fix anomalies. We tested our framework on 98 full-cycle sequences from the CAMUS dataset, which are available alongside this paper. Our temporal regularization method not only improves the accuracy of the segmentation across the whole sequences, but also enforces temporal and anatomical consistency.

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 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 website1.

Collecting large databases of annotated medical images is crucial for the validation and testing of feature extraction, statistical analysis, and machine learning algorithms. Recent advances in cardiac electromechanical modeling and image synthesis provided a framework to generate synthetic images based on realistic mesh simulations. Nonetheless, their potential to augment an existing database with large amounts of synthetic cases requires further investigation. We build upon these works and propose a revised scheme for synthesizing pathological cardiac sequences from real healthy sequences. Our new pipeline notably involves a much easier registration problem to reduce potential artifacts, and takes advantage of mesh correspondences to generate new data from a given case without additional registration. The output sequences are thoroughly examined in terms of quality and usability on a given application: the assessment of myocardial viability, via the generation of 465 synthetic cine MR sequences (15 healthy and 450 with pathological tissue viability [random location, extent, and grade, up to myocardial infarct]). We demonstrate that: 1) our methodology improves the state-of-the-art algorithms in terms of realism and accuracy of the simulated images and 2) our methodology is well-suited for the generation of large databases at small computational cost.

We propose a method based on deep learning to perform cardiac segmentation on short axis Magnetic resonance imaging stacks iteratively from the top slice (around the base) to the bottom slice (around the apex). At each iteration, a novel variant of the U-net is applied to propagate the segmentation of a slice to the adjacent slice below it. In other words, the prediction of a segmentation of a slice is dependent upon the already existing segmentation of an adjacent slice. The 3-D consistency is hence explicitly enforced. The method is trained on a large database of 3078 cases from the U.K. Biobank. It is then tested on the 756 different cases from the U.K. Biobank and three other state-of-the-art cohorts (ACDC with 100 cases, Sunnybrook with 30 cases, and RVSC with 16 cases). Results comparable or even better than the state of the art in terms of distance measures are achieved. They also emphasize the assets of our method, namely, enhanced spatial consistency (currently neither considered nor achieved by the state of the art), and the generalization ability to unseen cases even from other databases.

Diagnosing and localizing myocardial infarct is crucial for early patient management and therapy planning. We propose a new method for predicting the location of myocardial infarct from local wall deformation, which has value for risk stratification from routine examinations such as (3D) echocardiography. The pipeline combines non-linear dimensionality reduction of deformation patterns and two multi-scale kernel regressions. Confidence in the diagnosis is assessed by a map of local uncertainties, which integrates plausible infarct locations generated from the space of reduced dimensionality. These concepts were tested on 500 synthetic cases generated from a realistic cardiac electromechanical model, and 108 pairs of 3D echocardiographic sequences and delayed-enhancement magnetic resonance images from real cases. Infarct prediction is made at a spatial resolution around 4 mm, more than 10 times smaller than the current diagnosis, made regionally. Our method is accurate, and significantly outperforms the clinically-used thresholding of the deformation patterns (on real data: sensitivity/specificity of 0.828/0.804, area under the curve: 0.909 versus 0.742 for the most predictive strain component). Uncertainty adds value to refine the diagnosis and eventually re-examine suspicious cases.

We propose a technique for myocardial motion estimation based on image registration using both B-mode echocardiographic images and tissue Doppler sequences acquired interleaved. The velocity field is modeled continuously using B-splines and the spatiotemporal transform is constrained to be diffeomorphic. Images before scan conversion are used to improve the accuracy of the estimation. The similarity measure includes a model of the speckle pattern distribution of B-mode images. It also penalizes the disagreement between tissue Doppler velocities and the estimated velocity field. Registration accuracy is evaluated and compared to other alternatives using a realistic synthetic dataset, obtaining mean displacement errors of about 1 mm. Finally, the method is demonstrated on data acquired from six volunteers, both at rest and during exercise. Robustness is tested against low image quality and fast heart rates during exercise. Results show that our method provides a robust motion estimate in these situations.

Atlases and statistical models play important roles in the personalization and simulation of cardiac physiology. For the study of the heart, however, the construction of comprehensive atlases and spatio-temporal models is faced with a number of challenges, in particular the need to handle large and highly variable image datasets, the multi-region nature of the heart, and the presence of complex as well as small cardiovascular structures. In this paper, we present a detailed atlas and spatio-temporal statistical model of the human heart based on a large population of 3D+time multi-slice computed tomography sequences, and the framework for its construction. It uses spatial normalization based on nonrigid image registration to synthesize a population mean image and establish the spatial relationships between the mean and the subjects in the population. Temporal image registration is then applied to resolve each subject-specific cardiac motion and the resulting transformations are used to warp a surface mesh representation of the atlas to fit the images of the remaining cardiac phases in each subject. Subsequently, we demonstrate the construction of a spatio-temporal statistical model of shape such that the inter-subject and dynamic sources of variation are suitably separated. The framework is applied to a 3D+time data set of 138 subjects. The data is drawn from a variety of pathologies, which benefits its generalization to new subjects and physiological studies. The obtained level of detail and the extendability of the atlas present an advantage over most cardiac models published previously.