Segmentation of cardiac infarction in delayed-enhancement MRI using probability map and transformers-based neural networks.

BACKGROUND AND OBJECTIVE: Automatic segmentation of myocardial infarction is of great clinical interest for the quantitative evaluation of myocardial infarction (MI). Late Gadolinium Enhancement cardiac MRI (LGE-MRI) is commonly used in clinical practice to quantify MI, which is crucial for clinical diagnosis and treatment of cardiac diseases. However, the segmentation of infarcted tissue in LGE-MRI is highly challenging due to its high anisotropy and inhomogeneities. METHODS: The innovative aspect of our work lies in the utilization of a probability map of the healthy myocardium to guide the localization of infarction, as well as the combination of 2D U-Net and U-Net transformers to achieve the final segmentation. Instead of employing a binary segmentation map, we propose using a probability map of the normal myocardium, obtained through a dedicated 2D U-Net. To leverage spatial information, we employ a U-Net transformers network where we incorporate the probability map into the original image as an additional input. Then, To address the limitations of U-Net in segmenting accurately the contours, we introduce an adapted loss function. RESULTS: Our method has been evaluated on the 2020 MICCAI EMIDEC challenge dataset, yielding competitive results. Specifically, we achieved a Dice score of 92.94% for the myocardium and 92.36% for the infarction. These outcomes highlight the competitiveness of our approach. CONCLUSION: In the case of the infarction class, our proposed method outperforms state-of-the-art techniques across all metrics evaluated in the challenge, establishing its superior performance in infarction segmentation. This study further reinforces the importance of integrating a contour loss into the segmentation process.

Characterization of motion patterns by a spatio-temporal saliency descriptor in cardiac cine MRI.

BACKGROUND AND OBJECTIVE: Abnormalities of the heart motion reveal the presence of a disease. However, a quantitative interpretation of the motion is still a challenge due to the complex dynamics of the heart. This work proposes a quantitative characterization of regional cardiac motion patterns in cine magnetic resonance imaging (MRI) by a novel spatio-temporal saliency descriptor. METHOD: The strategy starts by dividing the cardiac sequence into a progression of scales which are in due turn mapped to a feature space of regional orientation changes, mimicking the multi-resolution decomposition of oriented primitive changes of visual systems. These changes are estimated as the difference between a particular time and the rest of the sequence. This decomposition is then temporarily and regionally integrated for a particular orientation and then for the set of different orientations. A final spatio-temporal 4D saliency map is obtained as the summation of the previously integrated information for the available scales. The saliency dispersion of this map was computed in standard cardiac locations as a measure of the regional motion pattern and was applied to discriminate control and hypertrophic cardiomyopathy (HCM) subjects during the diastolic phase. RESULTS: Salient motion patterns were estimated from an experimental set, which consisted of 3D sequences acquired by MRI from 108 subjects (33 control, 35 HCM, 20 dilated cardiomyopathy (DCM), and 20 myocardial infarction (MINF) from heterogeneous datasets). HCM and control subjects were classified by an SVM that learned the salient motion patterns estimated from the presented strategy, by achieving a 94% AUC. In addition, statistical differences (test t-student, p<0.05) were found among groups of disease in the septal and anterior ventricular segments at both the ED and ES, with salient motion characteristics aligned with existing knowledge on the diseases. CONCLUSIONS: Regional wall motion abnormality in the apical, anterior, basal, and inferior segments was associated with the saliency dispersion in HCM, DCM, and MINF compared to healthy controls during the systolic and diastolic phases. This saliency analysis may be used to detect subtle changes in heart function.

Desynchronization Strain Patterns and Contractility in Left Bundle Branch Block through Computer Model Simulation.

Left bundle branch block (LBBB) is associated with specific septal-to-lateral wall activation patterns which are strongly influenced by the intrinsic left ventricular (LV) contractility and myocardial scar localization. The objective of this study was to propose a computational-model-based interpretation of the different patterns of LV contraction observed in the case of LBBB and preserved contractility or myocardial scarring. Two-dimensional transthoracic echocardiography was used to obtain LV volumes and deformation patterns in three patients with LBBB: (1) a patient with non-ischemic dilated cardiomyopathy, (2) a patient with antero-septal myocardial scar, and (3) a patient with lateral myocardial scar. Scar was confirmed by the distribution of late gadolinium enhancement with cardiac magnetic resonance imaging (cMRI). Model parameters were evaluated manually to reproduce patient-derived data such as strain curves obtained from echocardiographic apical views. The model was able to reproduce the specific strain patterns observed in patients. A typical septal flash with pre-ejection shortening, rebound stretch, and delayed lateral wall activation was observed in the case of non-ischemic cardiomyopathy. In the case of lateral scar, the contractility of the lateral wall was significantly impaired and septal flash was absent. In the case of septal scar, septal flash and rebound stretch were also present as previously described in the literature. Interestingly, the model was also able to simulate the specific contractile properties of the myocardium, providing an excellent localization of LV scar in ischemic patients. The model was able to simulate the electromechanical delay and specific contractility patterns observed in patients with LBBB of ischemic and non-ischemic etiology. With further improvement and validation, this technique might be a useful tool for the diagnosis and treatment planning of heart failure patients needing CRT.

Left Ventricle Quantification Challenge: A Comprehensive Comparison and Evaluation of Segmentation and Regression for Mid-Ventricular Short-Axis Cardiac MR Data.

Automatic quantification of the left ventricle (LV) from cardiac magnetic resonance (CMR) images plays an important role in making the diagnosis procedure efficient, reliable, and alleviating the laborious reading work for physicians. Considerable efforts have been devoted to LV quantification using different strategies that include segmentation-based (SG) methods and the recent direct regression (DR) methods. Although both SG and DR methods have obtained great success for the task, a systematic platform to benchmark them remains absent because of differences in label information during model learning. In this paper, we conducted an unbiased evaluation and comparison of cardiac LV quantification methods that were submitted to the Left Ventricle Quantification (LVQuan) challenge, which was held in conjunction with the Statistical Atlases and Computational Modeling of the Heart (STACOM) workshop at the MICCAI 2018. The challenge was targeted at the quantification of 1) areas of LV cavity and myocardium, 2) dimensions of the LV cavity, 3) regional wall thicknesses (RWT), and 4) the cardiac phase, from mid-ventricle short-axis CMR images. First, we constructed a public quantification dataset Cardiac-DIG with ground truth labels for both the myocardium mask and these quantification targets across the entire cardiac cycle. Then, the key techniques employed by each submission were described. Next, quantitative validation of these submissions were conducted with the constructed dataset. The evaluation results revealed that both SG and DR methods can offer good LV quantification performance, even though DR methods do not require densely labeled masks for supervision. Among the 12 submissions, the DR method LDAMT offered the best performance, with a mean estimation error of 301 mm 2 for the two areas, 2.15 mm for the cavity dimensions, 2.03 mm for RWTs, and a 9.5% error rate for the cardiac phase classification. Three of the SG methods also delivered comparable performances. Finally, we discussed the advantages and disadvantages of SG and DR methods, as well as the unsolved problems in automatic cardiac quantification for clinical practice applications.

Fusion of 3D real-time echocardiography and cine MRI using a saliency analysis.

PURPOSE: This paper presents a novel 3D multimodal registration strategy to fuse 3D real-time echocardiography images with cardiac cine MRI images. This alignment is performed in a saliency space, which is designed to maximize similarity between the two imaging modalities. This fusion improves the quality of the available information. METHODS: The method performs in two steps: temporal and spatial registrations. A temporal alignment is firstly achieved by nonlinearly matching pairs of correspondences between the two modalities using a dynamic time warping. A temporal registration is then carried out by applying nonrigid transformations in a common saliency space where normalized cross correlation between temporal pairs of salient volumes is maximized. RESULTS: The alignment performance was evaluated with a set of 18 subjects, 3 with cardiomyopathies and 15 healthy, by computing the Dice score and Hausdorff distance with respect to manual delineations of the left ventricle cavity in both modalities. A Dice score and Hausdorff distance of [Formula: see text] and [Formula: see text], respectively, were obtained. In addition, the deformation field was estimated by quantifying its foldings, obtaining a 98% of regularity in the deformation field. CONCLUSIONS: The 3D multimodal registration strategy presented is performed in a saliency space. Unlike state-of-the-art methods, the presented one takes advantage of the temporal information of the heart to construct this common space, ending up with two well-aligned modalities and regular deformation fields. This preliminary study was evaluated on heterogeneous data composed of two different datasets, healthy and pathological cases, showing similar performances in both cases. Future work will focus on testing the presented strategy in a larger dataset with a balanced number of classes.

Similarity measures and attribute selection for case-based reasoning in transcatheter aortic valve implantation.

In a clinical decision support system, the purpose of case-based reasoning is to help clinicians make convenient decisions for diagnoses or interventional gestures. Past experience, which is represented by a case-base of previous patients, is exploited to solve similar current problems using four steps-retrieve, reuse, revise, and retain. The proposed case-based reasoning has been focused on transcatheter aortic valve implantation to respond to clinical issues pertaining vascular access and prosthesis choices. The computation of a relevant similarity measure is an essential processing step employed to obtain a set of retrieved cases from a case-base. A hierarchical similarity measure that is based on a clinical decision tree is proposed to better integrate the clinical knowledge, especially in terms of case representation, case selection and attributes weighting. A case-base of 138 patients is used to evaluate the case-based reasoning performance, and retrieve- and reuse-based criteria have been considered. The sensitivity for the vascular access and the prosthesis choice is found to 0.88 and 0.94, respectively, with the use of the hierarchical similarity measure as opposed to 0.53 and 0.79 for the standard similarity measure. Ninety percent of the suggested solutions are correctly classified for the proposed metric when four cases are retrieved. Using a dedicated similarity measure, with relevant and weighted attributes selected through a clinical decision tree, the set of retrieved cases, and consequently, the decision suggested by the case-based reasoning are substantially improved over state-of-the-art similarity measures.

Reproducibility of CT-based bone texture parameters of cancellous calf bone samples: influence of slice thickness.

Bone microarchitecture is an important determinant of the fracture risk, independently of bone mineral density. At present, bone biopsy is required for microarchitecture assessment, and accessible non-invasive techniques are needed. In this study, we tested the short-term reproducibility and parameter changes of a non-invasive method for microarchitecture assessment with a medical computed tomography. Texture parameters (run lengths and co-occurrence) were extracted from bone sample images. Reproducibility and the influence of slice thickness (1, 3, 5 and 8mm) were also studied. After five repositionings, short-term reproducibility was found to be good. All run length parameters but one fell significantly with increasing slice thickness. Co-occurrence parameters showed different patterns of change. Short-term coefficients of variation of texture parameters used to assess bone microarchitecture were similar to those obtained elsewhere with other techniques. The results were influenced by slice thicknesses, emphasizing the importance of the conditions of acquisition.

Markov random field modeling for three-dimensional reconstruction of the left ventricle in cardiac angiography.

This paper reports on a method for left ventricle three-dimensional (3-D) reconstruction from two orthogonal ventriculograms. The proposed algorithm is voxel-based and takes into account the conical projection geometry associated with the biplane image acquisition equipment. The reconstruction process starts with an initial ellipsoidal approximation derived from the input ventriculograms. This model is subsequently deformed in such a way as to match the input projections. To this end, the object is modeled as a 3-D Markov-Gibbs random field, and an energy function is defined so that it includes one term that models the projections compatibility and another one that includes the space-time regularity constraints. The performance of this reconstruction method is evaluated by considering the reconstruction of mathematically synthesized phantoms and two 3-D binary databases from two orthogonal synthesized projections. The method is also tested using real biplane ventriculograms. In this case, the performance of the reconstruction is expressed in terms of the projection error, which attains values between 9.50% and 11.78 % for two biplane sequences including a total of 55 images.

Synchronization and Registration of Cine Magnetic Resonance and Dynamic Computed Tomography Images of the Heart.

The synchronization and registration of dynamic computed tomography (CT) and magnetic resonance images (MRI) of the heart is required to perform a combined analysis of their complementary information. We propose a novel method that synchronizes and registers intrapatient dynamic CT and cine-MRI short axis view (SAX). For the synchronization step, a normalized cross-correlation curve is computed from each image sequence to describe the global cardiac dynamics. The time axes of these curves are then warped using an adapted dynamic time warping (DTW) procedure. The adaptation constrains the time deformation to obtain a coherent warping function. The registration step then computes the rigid transformation that maximizes the multiimage normalized mutual information of DTW-synchronized images. The DTW synchronization and the multiimage registration were evaluated using dynamic CT and cine-SAX acquisitions from nine patients undergoing cardiac resynchronization therapy. The distance between the end-systolic phases after DTW was used to evaluate the synchronization. Mean errors, expressed as a percentage of the RR-intervals, were 3.9% and 3.7% after adapted DTW synchronization against 10.8% and 11.3% after linear synchronization, for dynamic CT and cine-SAX, respectively. This suggests that the adapted DTW synchronization leads to a coherent warping of cardiac dynamics. The multiimage registration was evaluated using fiducial points. Compared to a monoimage and a two-image registration, the multiimage registration of DTW-synchronized images obtained the lowest mean fiducial error showing that the use of dynamic voxel intensity information improves the registration.

Registration of dynamic multiview 2D ultrasound and late gadolinium enhanced images of the heart: Application to hypertrophic cardiomyopathy characterization.

Describing and analyzing heart multiphysics requires the acquisition and fusion of multisensor cardiac images. Multisensor image fusion enables a combined analysis of these heterogeneous modalities. We propose to register intra-patient multiview 2D+t ultrasound (US) images with multiview late gadolinium-enhanced (LGE) images acquired during cardiac magnetic resonance imaging (MRI), in order to fuse mechanical and tissue state information. The proposed procedure registers both US and LGE to cine MRI. The correction of slice misalignment and the rigid registration of multiview LGE and cine MRI are studied, to select the most appropriate similarity measure. It showed that mutual information performs the best for LGE slice misalignment correction and for LGE and cine registration. Concerning US registration, dynamic endocardial contours resulting from speckle tracking echocardiography were exploited in a geometry-based dynamic registration. We propose the use of an adapted dynamic time warping procedure to synchronize cardiac dynamics in multiview US and cine MRI. The registration of US and LGE MRI was evaluated on a dataset of patients with hypertrophic cardiomyopathy. A visual assessment of 330 left ventricular regions from US images of 28 patients resulted in 92.7% of regions successfully aligned with cardiac structures in LGE. Successfully-aligned regions were then used to evaluate the abilities of strain indicators to predict the presence of fibrosis. Longitudinal peak-strain and peak-delay of aligned left ventricular regions were computed from corresponding regional strain curves from US. The Mann-Withney test proved that the expected values of these indicators change between the populations of regions with and without fibrosis (p < 0.01). ROC curves otherwise proved that the presence of fibrosis is one factor amongst others which modifies longitudinal peak-strain and peak-delay.

MSCT labelling for pre-operative planning in cardiac resynchronization therapy.

The objective of this paper is twofold: (i) to show how multislice computed tomography (MSCT) data sets bring the information required for cardiac resynchronisation therapy (CRT) planning; (ii) to demonstrate the feasibility of 3D navigation into the veins where left ventricular leads have to be placed. The former has been achieved by exploring and labelling the cardiac structures of concern, the latter has been performed by using the concept of virtual navigation with high resolution surface detection and estimation algorithms.

Multi-modal data fusion for Cardiac Resynchronization Therapy planning and assistance.

Cardiac Resynchronization Therapy (CRT) has been validated as an efficient treatment for selected patients suffering from heart failure with cardiac dyssynchrony. In case of bi-ventricular stimulation, the response to the therapy may be improved by an optimal choice of the left ventricle (LV) pacing sites. The characterization of LV properties to select the best candidate sites and to precise their access modes would be useful for the clinician in pre- and per-operative stages. For that purpose, we propose a new pre-operative analysis solution integrating previously developed multi-modal data registration methods and a new segmentation process of their coronary venous access. Moreover, a novel visualization interface is proposed to help the clinician to visualize the most relevant pacing sites and their access during the implantation in the operating room. This work is illustrated on real CRT data patients.

Classification of LV wall motion in cardiac MRI using kernel Dictionary Learning with a parametric approach.

In this paper, we propose a parametric approach for the assessment of wall motion in Left Ventricle (LV) function in cardiac cine-Magnetic Resonance Imaging (MRI). Time-signal intensity curves (TSICs) are identified in Spatio-temporal image profiles extracted from different anatomical segments in a cardiac MRI sequence. Different parameters are constructed from specific TSICs that present a decreasing then increasing shape reflecting dynamic information of the LV contraction. The parameters extracted from these curves are related to: 1) an average curve based on a clustering process, 2) curve skewness and 3) cross correlation values between each average clustered curve and a patient-specific reference. Several tests are performed in order to construct different vectors to train a sparse classifier based on kernel Dictionary Learning (DL). Results are compared with other classifiers like Support Vector Machine (SVM) and Discriminative Dictionary Learning. The best classification performance is obtained with information of skewness and the average curve with an accuracy about 94% using the mentioned sparse based kernel DL with a radial basis function kernel.

Improved Estimation of Cardiac Function Parameters Using a Combination of Independent Automated Segmentation Results in Cardiovascular Magnetic Resonance Imaging.

This work aimed at combining different segmentation approaches to produce a robust and accurate segmentation result. Three to five segmentation results of the left ventricle were combined using the STAPLE algorithm and the reliability of the resulting segmentation was evaluated in comparison with the result of each individual segmentation method. This comparison was performed using a supervised approach based on a reference method. Then, we used an unsupervised statistical evaluation, the extended Regression Without Truth (eRWT) that ranks different methods according to their accuracy in estimating a specific biomarker in a population. The segmentation accuracy was evaluated by estimating six cardiac function parameters resulting from the left ventricle contour delineation using a public cardiac cine MRI database. Eight different segmentation methods, including three expert delineations and five automated methods, were considered, and sixteen combinations of the automated methods using STAPLE were investigated. The supervised and unsupervised evaluations demonstrated that in most cases, STAPLE results provided better estimates than individual automated segmentation methods. Overall, combining different automated segmentation methods improved the reliability of the segmentation result compared to that obtained using an individual method and could achieve the accuracy of an expert.

Multimodal registration and data fusion for cardiac resynchronization therapy optimization.

Cardiac resynchronization therapy (CRT) has been shown to improve cardiovascular function in specific patients suffering from heart failure. This procedure still needs to be optimized to overcome the high rate of implanted patients that do not respond to this therapy. We propose in this work a better characterization of the electro-mechanical (EM) coupling of each region of the left ventricle (LV) that could be useful to precise the best implantation site. A new descriptor is proposed with the extraction of local electro-mechanical delays. Their measurement is based on the fusion of anatomical, functional and electrical data acquired using computed tomography (CT), speckle tracking echocardiography (STE), and electro-anatomical mappings (EAM). We propose a workflow to place multimodal data in the same geometrical referential system and to extract local electro-mechanical descriptors. It implies the fusion of electrical and mechanical data on a 3D+ t anatomical model of the LV. It mainly consists in four steps: 1) the modeling of the endocardium using a dynamic surface estimated from CT images; 2) the semi-interactive registration of EAM data and CT images; 3) the automatic registration of STE data on the dynamic model, using a metric based on Fourier descriptors and dynamic time warping; 4) the temporal alignment between EAM and STE and the estimation of local electro-mechanical delays. The proposed process has been applied to real data corresponding to five patients undergoing CRT. Results show that local electro-mechanical delays provide meaningful information on the local characterization of the LV and may be useful for the optimal pacing site selection in CRT.

Coronary vein extraction in MSCT volumes using minimum cost path and geometrical moments.

This work deals with the extraction of patient-specific coronary venous anatomy in preoperative multislice computed tomography (MSCT) volumes. A hybrid approach has been specifically designed for low-contrast vascular structure detection. It makes use of a minimum cost path technique with a Fast-Marching front propagation to extract the vessel centerline. A second procedure was applied to refine the position of the path and estimate the local radius along the vessel. This was achieved with an iterative multiscale algorithm based on geometrical moments. Parameter tuning was performed using a dedicated numerical phantom, and then the algorithm was applied to extract the coronary venous system. Results are provided on three MSCT volume sequences acquired for patients selected for a cardiac resynchronization therapy procedure. A visibility study was carried out by a medical expert who labeled venous segments on a set of 18 volumes. A comparison with two other Fast-Marching techniques and a geometrical moment based tracking method is also reported.

Nonsupervised ranking of different segmentation approaches: application to the estimation of the left ventricular ejection fraction from cardiac cine MRI sequences.

A statistical methodology is proposed to rank several estimation methods of a relevant clinical parameter when no gold standard is available. Based on a regression without truth method, the proposed approach was applied to rank eight methods without using any a priori information regarding the reliability of each method and its degree of automation. It was only based on a prior concerning the statistical distribution of the parameter of interest in the database. The ranking of the methods relies on figures of merit derived from the regression and computed using a bootstrap process. The methodology was applied to the estimation of the left ventricular ejection fraction derived from cardiac magnetic resonance images segmented using eight approaches with different degrees of automation: three segmentations were entirely manually performed and the others were variously automated. The ranking of methods was consistent with the expected performance of the estimation methods: the most accurate estimates of the ejection fraction were obtained using manual segmentations. The robustness of the ranking was demonstrated when at least three methods were compared. These results suggest that the proposed statistical approach might be helpful to assess the performance of estimation methods on clinical data for which no gold standard is available.

Similarity enhancement for automatic segmentation of cardiac structures in computed tomography volumes.

The aim of this research is proposing a 3-D similarity enhancement technique useful for improving the segmentation of cardiac structures in Multi-Slice Computerized Tomography (MSCT) volumes. The similarity enhancement is obtained by subtracting the intensity of the current voxel and the gray levels of their adjacent voxels in two volumes resulting after preprocessing. Such volumes are: a. - a volume obtained after applying a Gaussian distribution and a morphological top-hat filter to the input and b. - a smoothed volume generated by processing the input with an average filter. Then, the similarity volume is used as input to a region growing algorithm. This algorithm is applied to extract the shape of cardiac structures, such as left and right ventricles, in MSCT volumes. Qualitative and quantitative results show the good performance of the proposed approach for discrimination of cardiac cavities.

Comparison of different segmentation approaches without using gold standard. Application to the estimation of the left ventricle ejection fraction from cardiac cine MRI sequences.

A statistical method is proposed to compare several estimates of a relevant clinical parameter when no gold standard is available. The method is illustrated by considering the left ventricle ejection fraction derived from cardiac magnetic resonance images and computed using seven approaches with different degrees of automation. The proposed method did not use any a priori regarding with the reliability of each method and its degree of automation. The results showed that the most accurate estimates of the ejection fraction were obtained using manual segmentations, followed by the semiautomatic methods, while the methods with the least user input yielded the least accurate ejection fraction estimates. These results were consistent with the expected performance of the estimation methods, suggesting that the proposed statistical approach might be helpful to assess the performance of estimation methods on clinical data for which no gold standard is available.