Evaluation of state-of-the-art segmentation algorithms for left ventricle infarct from late Gadolinium enhancement MR images

Optimizing ventricular tachycardia ablation through imaging-based assessment of arrhythmic substrate: A comprehensive review and roadmap for the future

Ventricular tachycardia (VT) is a life-threatening heart rhythm and has long posed a complex challenge in the field of cardiology. Recent developments in advanced imaging modalities have aimed to improve comprehension of underlying arrhythmic substrate for VT. To this extent, high-resolution cardiac magnetic resonance (CMR) and cardiac computed tomography (CCT) have emerged as tools for accurately visualizing and characterizing scar tissue, fibrosis, and other critical structural abnormalities within the heart, providing novel insights into VT triggers and substrate. However, clinical implementation of knowledge derived from these advanced imaging techniques in improving VT treatment and guiding invasive therapeutic strategies continues to pose significant challenges. A pivotal concern lies in the absence of standardized imaging protocols and analysis methodologies, resulting in a large variance in data quality and consistency. Furthermore, the clinical significance and outcomes associated with VT substrate characterization through CMR and CCT remain dynamic and subject to ongoing evolution. This highlights the need for refinement of these techniques before their reliable integration into routine patient care can be realized. The primary objectives of this study are twofold: firstly, to provide a comprehensive overview of the studies conducted over the last 15 years, summarizing the current available literature on imaging-based assessment of VT substrate. Secondly, to critically analyze and evaluate the selected studies, with the aim of providing valuable insights that can inform current clinical practice and future research.

Toward Enabling Cardiac Digital Twins of Myocardial Infarction Using Deep Computational Models for Inverse Inference

Cardiac digital twins (CDTs) have the potential to offer individualized evaluation of cardiac function in a non-invasive manner, making them a promising approach for personalized diagnosis and treatment planning of myocardial infarction (MI). The inference of accurate myocardial tissue properties is crucial in creating a reliable CDT of MI. In this work, we investigate the feasibility of inferring myocardial tissue properties from the electrocardiogram (ECG) within a CDT platform. The platform integrates multi-modal data, such as cardiac MRI and ECG, to enhance the accuracy and reliability of the inferred tissue properties. We perform a sensitivity analysis based on computer simulations, systematically exploring the effects of infarct location, size, degree of transmurality, and electrical activity alteration on the simulated QRS complex of ECG, to establish the limits of the approach. We subsequently present a novel deep computational model, comprising a dual-branch variational autoencoder and an inference model, to infer infarct location and distribution from the simulated QRS. The proposed model achieves mean Dice scores of 0.457 ±0.317 and 0.302 ±0.273 for the inference of left ventricle scars and border zone, respectively. The sensitivity analysis enhances our understanding of the complex relationship between infarct characteristics and electrophysiological features. The in silico experimental results show that the model can effectively capture the relationship for the inverse inference, with promising potential for clinical application in the future. The code is available at https://github.com/lileitech/MI_inverse_inference.

Computational modeling of cardiac electrophysiology and arrhythmogenesis: toward clinical translation

The complexity of cardiac electrophysiology, involving dynamic changes in numerous components across multiple spatial (from ion channel to organ) and temporal (from milliseconds to days) scales, makes an intuitive or empirical analysis of cardiac arrhythmogenesis challenging. Multiscale mechanistic computational models of cardiac electrophysiology provide precise control over individual parameters, and their reproducibility enables a thorough assessment of arrhythmia mechanisms. This review provides a comprehensive analysis of models of cardiac electrophysiology and arrhythmias, from the single cell to the organ level, and how they can be leveraged to better understand rhythm disorders in cardiac disease and to improve heart patient care. Key issues related to model development based on experimental data are discussed, and major families of human cardiomyocyte models and their applications are highlighted. An overview of organ-level computational modeling of cardiac electrophysiology and its clinical applications in personalized arrhythmia risk assessment and patient-specific therapy of atrial and ventricular arrhythmias is provided. The advancements presented here highlight how patient-specific computational models of the heart reconstructed from patient data have achieved success in predicting risk of sudden cardiac death and guiding optimal treatments of heart rhythm disorders. Finally, an outlook toward potential future advances, including the combination of mechanistic modeling and machine learning/artificial intelligence, is provided. As the field of cardiology is embarking on a journey toward precision medicine, personalized modeling of the heart is expected to become a key technology to guide pharmaceutical therapy, deployment of devices, and surgical interventions.

Automatic detection of myocardial ischaemia using generalisable spatio-temporal hierarchical Bayesian modelling of DCE-MRI

Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) can be used as a non-invasive method for the assessment of myocardial perfusion. The acquired images can be utilised to analyse the spatial extent and severity of myocardial ischaemia (regions with impaired microvascular blood flow). In the present paper, we propose a novel generalisable spatio-temporal hierarchical Bayesian model (GST-HBM) to automate the detection of ischaemic lesions and improve the in silico prediction accuracy by systematically integrating spatio-temporal context information. We present a computational inference procedure with an adequate trade-off between accuracy and computational efficiency, whereby model parameters are sampled from the posterior distribution with Gibbs sampling, while lower-level hyperparameters are selected using model selection strategies based on the Watanabe Akaike information criterion (WAIC). We have assessed our method on both synthetic (in silico) data with known gold-standard and 12 sets of clinical first-pass myocardial perfusion DCE-MRI datasets. We have also carried out a comparative performance evaluation with four established alternative methods: Gaussian mixture model (GMM), opening and closing operations based on Gaussian mixture model (GMMC&Omax), Markov random field constrained Gaussian mixture model (GMM-MRF) and model-based hierarchical Bayesian model (M-HBM). Our results show that the proposed GST-HBM method achieves much higher in silico prediction accuracy than the established alternative methods. Furthermore, this method appears to provide a more robust delineation of ischaemic lesions in datasets affected by spatially variant noise.

Impact of late gadolinium enhancement image acquisition resolution on neural network based automatic scar segmentation

BACKGROUND

Automatic myocardial scar segmentation from late gadolinium enhancement (LGE) images using neural networks promises an alternative to time-consuming and observer-dependent semi-automatic approaches. However, alterations in data acquisition, reconstruction as well as post-processing may compromise network performance. The objective of the present work was to systematically assess network performance degradation due to a mismatch of point-spread function between training and testing data.

METHODS

Thirty-six high-resolution (0.7×0.7×2.0 mm3) LGE k-space datasets were acquired post-mortem in porcine models of myocardial infarction. The in-plane point-spread function and hence in-plane resolution Δx was retrospectively degraded using k-space lowpass filtering, while field-of-view and matrix size were kept constant. Manual segmentation of the left ventricle (LV) and healthy remote myocardium was performed to quantify location and area (% of myocardium) of scar by thresholding (≥ SD5 above remote). Three standard U-Nets were trained on training resolutions Δxtrain = 0.7, 1.2 and 1.7 mm to predict endo- and epicardial borders of LV myocardium and scar. The scar prediction of the three networks for varying test resolutions (Δxtest = 0.7 to 1.7 mm) was compared against the reference SD5 thresholding at 0.7 mm. Finally, a fourth network trained on a combination of resolutions (Δxtrain = 0.7 to 1.7 mm) was tested.

RESULTS

The prediction of relative scar areas showed the highest precision when the resolution of the test data was identical to or close to the resolution used during training. The median fractional scar errors and precisions (IQR) from networks trained and tested on the same resolution were 0.0 percentage points (p.p.) (1.24 - 1.45), and - 0.5 - 0.0 p.p. (2.00 - 3.25) for networks trained and tested on the most differing resolutions, respectively. Deploying the network trained on multiple resolutions resulted in reduced resolution dependency with median scar errors and IQRs of 0.0 p.p. (1.24 - 1.69) for all investigated test resolutions.

CONCLUSION

A mismatch of the imaging point-spread function between training and test data can lead to degradation of scar segmentation when using current U-Net architectures as demonstrated on LGE porcine myocardial infarction data. Training networks on multi-resolution data can alleviate the resolution dependency.

Segment anything model for medical images?

The Segment Anything Model (SAM) is the first foundation model for general image segmentation. It has achieved impressive results on various natural image segmentation tasks. However, medical image segmentation (MIS) is more challenging because of the complex modalities, fine anatomical structures, uncertain and complex object boundaries, and wide-range object scales. To fully validate SAM's performance on medical data, we collected and sorted 53 open-source datasets and built a large medical segmentation dataset with 18 modalities, 84 objects, 125 object-modality paired targets, 1050K 2D images, and 6033K masks. We comprehensively analyzed different models and strategies on the so-called COSMOS 1050K dataset. Our findings mainly include the following: (1) SAM showed remarkable performance in some specific objects but was unstable, imperfect, or even totally failed in other situations. (2) SAM with the large ViT-H showed better overall performance than that with the small ViT-B. (3) SAM performed better with manual hints, especially box, than the Everything mode. (4) SAM could help human annotation with high labeling quality and less time. (5) SAM was sensitive to the randomness in the center point and tight box prompts, and may suffer from a serious performance drop. (6) SAM performed better than interactive methods with one or a few points, but will be outpaced as the number of points increases. (7) SAM's performance correlated to different factors, including boundary complexity, intensity differences, etc. (8) Finetuning the SAM on specific medical tasks could improve its average DICE performance by 4.39% and 6.68% for ViT-B and ViT-H, respectively. Codes and models are available at: https://github.com/yuhoo0302/Segment-Anything-Model-for-Medical-Images. We hope that this comprehensive report can help researchers explore the potential of SAM applications in MIS, and guide how to appropriately use and develop SAM.

Artificial Intelligence in Heart Failure: Friend or Foe?

In recent times, there have been notable changes in cardiovascular medicine, propelled by the swift advancements in artificial intelligence (AI). The present work provides an overview of the current applications and challenges of AI in the field of heart failure. It emphasizes the "garbage in, garbage out" issue, where AI systems can produce inaccurate results with skewed data. The discussion covers issues in heart failure diagnostic algorithms, particularly discrepancies between existing models. Concerns about the reliance on the left ventricular ejection fraction (LVEF) for classification and treatment are highlighted, showcasing differences in current scientific perceptions. This review also delves into challenges in implementing AI, including variable considerations and biases in training data. It underscores the limitations of current AI models in real-world scenarios and the difficulty in interpreting their predictions, contributing to limited physician trust in AI-based models. The overarching suggestion is that AI can be a valuable tool in clinicians' hands for treating heart failure patients, as far as existing medical inaccuracies have been addressed before integrating AI into these frameworks.

Deep learning for classification of late gadolinium enhancement lesions based on the 16-segment left ventricular model

PURPOSE

This study aimed to develop and validate a deep learning-based method that allows for segmental analysis of myocardial late gadolinium enhancement (LGE) lesions.

METHODS

Cardiac LGE data from 170 patients with coronary artery disease and non-ischemic heart disease were used for training, validation, and testing. Short-axis images were transformed to polar space after identification of the left ventricular (LV) center point and anterior right ventricular (RV) insertion point. Images were obtained after dividing the polar transformed images into segments based on the 16-segment LV model. Five different deep convolutional neural network (CNN) models were developed and validated using the labeled data, where the image after the division corresponded to a segment, and the lesion labeling was based on the 16-segment LV model. Unseen testing data were used to evaluate the performance of the lesion classification.

RESULTS

Without manual lesion segmentation and annotation, the proposed method showed an area under the curve (AUC) of 0.875, and a precision, recall, and F1-score of 0.723, 0.783, and 0.752, respectively for the lesion class when the pretrained ResNet50 model was tested for all slice images. The two pretrained models of ResNet50 and EfficientNet-B0 outperformed the three non-pretrained CNN models in terms of AUCs (0.873-0.875 vs. 0.834-0.841).

CONCLUSION

The proposed method is based on learning a deep CNN model from polar transformed images to predict LGE lesions with good accuracy and does not require time-consuming annotation procedures such as lesion segmentation.

FK-means: automatic atrial fibrosis segmentation using fractal-guided K-means clustering with Voronoi-clipping feature extraction of anatomical structures

Assessment of left atrial (LA) fibrosis from late gadolinium enhancement (LGE) magnetic resonance imaging (MRI) adds to the management of patients with atrial fibrillation. However, accurate assessment of fibrosis in the LA wall remains challenging. Excluding anatomical structures in the LA proximity using clipping techniques can reduce misclassification of LA fibrosis. A novel FK-means approach for combined automatic clipping and automatic fibrosis segmentation was developed. This approach combines a feature-based Voronoi diagram with a hierarchical 3D K-means fractal-based method. The proposed automatic Voronoi clipping method was applied on LGE-MRI data and achieved a Dice score of 0.75, similar to the score obtained by a deep learning method (3D UNet) for clipping (0.74). The automatic fibrosis segmentation method, which uses the Voronoi clipping method, achieved a Dice score of 0.76. This outperformed a 3D UNet method for clipping and fibrosis classification, which had a Dice score of 0.69. Moreover, the proposed automatic fibrosis segmentation method achieved a Dice score of 0.90, using manual clipping of anatomical structures. The findings suggest that the automatic FK-means analysis approach enables reliable LA fibrosis segmentation and that clipping of anatomical structures in the atrial proximity can add to the assessment of atrial fibrosis.

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.

Quality control-driven deep ensemble for accountable automated segmentation of cardiac magnetic resonance LGE and VNE images

Background

Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) imaging is the gold standard for non-invasive myocardial tissue characterisation. However, accurate segmentation of the left ventricular (LV) myocardium remains a challenge due to limited training data and lack of quality control. This study addresses these issues by leveraging generative adversarial networks (GAN)-generated virtual native enhancement (VNE) images to expand the training set and incorporating an automated quality control-driven (QCD) framework to improve segmentation reliability.

Methods

A dataset comprising 4,716 LGE images (from 1,363 patients with hypertrophic cardiomyopathy and myocardial infarction) was used for development. To generate additional clinically validated data, LGE data were augmented with a GAN-based generator to produce VNE images. LV was contoured on these images manually by clinical observers. To create diverse candidate segmentations, the QCD framework involved multiple U-Nets, which were combined using statistical rank filters. The framework predicted the Dice Similarity Coefficient (DSC) for each candidate segmentation, with the highest predicted DSC indicating the most accurate and reliable result. The performance of the QCD ensemble framework was evaluated on both LGE and VNE test datasets (309 LGE/VNE images from 103 patients), assessing segmentation accuracy (DSC) and quality prediction (mean absolute error (MAE) and binary classification accuracy).

Results

The QCD framework effectively and rapidly segmented the LV myocardium (<1 s per image) on both LGE and VNE images, demonstrating robust performance on both test datasets with similar mean DSC (LGE: 0.845 ± 0.075 ; VNE: 0.845 ± 0.071 ; p = n s ). Incorporating GAN-generated VNE data into the training process consistently led to enhanced performance for both individual models and the overall framework. The quality control mechanism yielded a high performance (MAE = 0.043 , accuracy = 0.951 ) emphasising the accuracy of the quality control-driven strategy in predicting segmentation quality in clinical settings. Overall, no statistical difference (p = n s ) was found when comparing the LGE and VNE test sets across all experiments.

Conclusions

The QCD ensemble framework, leveraging GAN-generated VNE data and an automated quality control mechanism, significantly improved the accuracy and reliability of LGE segmentation, paving the way for enhanced and accountable diagnostic imaging in routine clinical use.

MyoPS: A benchmark of myocardial pathology segmentation combining three-sequence cardiac magnetic resonance images

Assessment of myocardial viability is essential in diagnosis and treatment management of patients suffering from myocardial infarction, and classification of pathology on the myocardium is the key to this assessment. This work defines a new task of medical image analysis, i.e., to perform myocardial pathology segmentation (MyoPS) combining three-sequence cardiac magnetic resonance (CMR) images, which was first proposed in the MyoPS challenge, in conjunction with MICCAI 2020. Note that MyoPS refers to both myocardial pathology segmentation and the challenge in this paper. The challenge provided 45 paired and pre-aligned CMR images, allowing algorithms to combine the complementary information from the three CMR sequences for pathology segmentation. In this article, we provide details of the challenge, survey the works from fifteen participants and interpret their methods according to five aspects, i.e., preprocessing, data augmentation, learning strategy, model architecture and post-processing. In addition, we analyze the results with respect to different factors, in order to examine the key obstacles and explore the potential of solutions, as well as to provide a benchmark for future research. The average Dice scores of submitted algorithms were 0.614±0.231 and 0.644±0.153 for myocardial scars and edema, respectively. We conclude that while promising results have been reported, the research is still in the early stage, and more in-depth exploration is needed before a successful application to the clinics. MyoPS data and evaluation tool continue to be publicly available upon registration via its homepage (www.sdspeople.fudan.edu.cn/zhuangxiahai/0/myops20/).

An accurate and time-efficient deep learning-based system for automated segmentation and reporting of cardiac magnetic resonance-detected ischemic scar

BACKGROUND AND OBJECTIVES

Myocardial infarction scar (MIS) assessment by cardiac magnetic resonance provides prognostic information and guides patients' clinical management. However, MIS segmentation is time-consuming and not performed routinely. This study presents a deep-learning-based computational workflow for the segmentation of left ventricular (LV) MIS, for the first time performed on state-of-the-art dark-blood late gadolinium enhancement (DB-LGE) images, and the computation of MIS transmurality and extent.

METHODS

DB-LGE short-axis images of consecutive patients with myocardial infarction were acquired at 1.5T in two centres between Jan 1, 2019, and June 1, 2021. Two convolutional neural network (CNN) models based on the U-Net architecture were trained to sequentially segment the LV and MIS, by processing an incoming series of DB-LGE images. A 5-fold cross-validation was performed to assess the performance of the models. Model outputs were compared respectively with manual (LV endo- and epicardial border) and semi-automated (MIS, 4-Standard Deviation technique) ground truth to assess the accuracy of the segmentation. An automated post-processing and reporting tool was developed, computing MIS extent (expressed as relative infarcted mass) and transmurality.

RESULTS

The dataset included 1355 DB-LGE short-axis images from 144 patients (MIS in 942 images). High performance (> 0.85) as measured by the Intersection over Union metric was obtained for both the LV and MIS segmentations on the training sets. The performance for both LV and MIS segmentations was 0.83 on the test sets. Compared to the 4-Standard Deviation segmentation technique, our system was five times quicker (<1 min versus 7 ± 3 min), and required minimal user interaction.

CONCLUSIONS

Our solution successfully addresses different issues related to automatic MIS segmentation, including accuracy, time-effectiveness, and the automatic generation of a clinical report.

Automatic development of 3D anatomical models of border zone and core scar regions in the left ventricle

Patients with myocardial infarction are at elevated risk of sudden cardiac death, and scar tissue arising from infarction is known to play a role. The accurate identification of scars therefore is crucial for risk assessment, quantification and guiding interventions. Typically, core scars and grey peripheral zones are identified by radiologists and clinicians based on cardiac late gadolinium enhancement magnetic resonance images (LGE-MRI). Scar regions from LGE-MRI vary in size, shape, heterogeneity, artifacts, and image resolution. Thus, manual segmentation is time consuming, and influenced by the observer's experience (bias effect). We propose a fully automatic framework that develops 3D anatomical models of the left ventricle with border zone and core scar regions that are free from bias effect. Our myocardium (SOCRATIS), border scar and core scar (BZ-SOCRATIS) segmentation pipelines were evaluated using internal and external validation datasets. The automatic myocardium segmentation framework performed a Dice score of 81.9% and 70.0% in the internal and external validation dataset. The automatic scar segmentation pipeline achieved a Dice score of 60.9% for the core scar segmentation and 43.7% for the border zone scar segmentation in the internal dataset and in the external dataset a Dice score of 44.2% for the core scar segmentation and 54.8% for the border scar segmentation respectively. To the best of our knowledge, this is the first study outlining a fully automatic framework to develop 3D anatomical models of the left ventricle with border zone and core scar regions. Our method exhibits high performance without the need for training or tuning in an unseen cohort (unsupervised).

DE-MRI myocardial fibrosis segmentation and classification model based on multi-scale self-supervision and transformer

OBJECTIVE

The segmentation and categorization of fibrotic tissue in time-lapse enhanced MRI scanning are quite challenging, and it is mainly done manually for myocardial DE-MRI images. On the other hand, DE-MRI instructions for segmenting and classifying cardiac hypertrophy are complex and prone to inaccuracy. Developing cardiac DE-MRI classification and prediction methods is crucial.

METHODS

This paper introduces a self-supervised myocardial histology segmentation algorithm with multi-scale portrayal consistency to address the degree of sophistication of cardiology DE-MRI. The model retrieves multi-scale representations from multiple expanded viewpoints using a Siamese system and uses resemblance learning instruction to achieve unlabeled representations. The DE-MRI data train the network weights to generate a superior segmentation effect by accurately reflecting the exact scale information. The paper provides an end-to-end method for detecting myocardial fibrosis tissue using a Transformer as a result of the poor classification outcomes of myocardial fibrosis substance in DE-MRI. A deep learning model is created using the Pre-LN Transformer decoded simultaneously with the Multi-Scale Transformer backbone structure developed in this paper. In addition, the joint regression cost, which incorporates the CIoU Loss and the L1 Loss, is used to determine the distance between forecast blocks and labels.

RESULTS

Increasing the independent evaluation and annotations position compared enhances performance compared to the segmentation method without canvas matching by 1.76%, 1.27%, 0.93%, and -1.17 mm on Dice, PPV, SEN, and HD, respectively. Based on the strongest of the three single-scale representation methodologies, the segmentation model in this study is enhanced by 0.71%, 0.79%, and 1.47%, as well as -1.49 mm on Dice, PPV, SEN, and HD, respectively. The effectiveness and reliability of the segmentation model are confirmed. Additionally, testing results show that this study's recognition system's mAP is 84.97%, which is greater than the benchmark techniques used in most other studies. The framework converges round is compressed by 18.1% compared to the DETR detection approach, and the identification rate is improved by 3.5%, proving the strategy's value.

CONCLUSION

The self-supervised cardiac fibrosis segmentation method with multi-scale portrayal consistency and end-to-end myocardial histology categorization is introduced in this study. To solve the challenges of segmentation and myocardial fibrosis identification in cardiology DE-MRI, a Transformer-based detection approach is put forth. It may address the issue of the myocardial scarring material's low accuracy in segmentation and classification in DE-MRI, as well as provide clinicians with a fibrosis diagnosis that is supplementary to the conventional therapy of heart ailments.

Classification of myocardial fibrosis in DE-MRI based on semi-supervised semantic segmentation and dual attention mechanism

OBJECTIVE

It is essential to utilize cardiac delayed-enhanced magnetic resonance imaging (DE-MRI) to diagnose cardiovascular disease. By segmenting myocardium DE-MRI images, it provides critical information for the evaluation and treatment of myocardial infarction. As a consequence, it is vital to investigate the segmentation and classification technique of myocardial DE-MRI.

METHODS

Firstly, an end-to-end minimally supervised and semi-supervised semantic DE-MRI myocardial fibrosis segmentation framework is proposed, which combines image classification and semantic segmentation branches based on the self-attention mechanism. Following that, a residual hole network fused with the dual attention mechanism was built, and a double attention metabolic pathway classification method for cardiac fibrosis in DE-MRI images was developed.

RESULTS

By adding pixel-level labels to an extra 40 training images, the segmentation model may enhance semantic segmentation performance by 2.6 percent (from 61.2 percent to 63.8 percent). When the number of pixel-level labels is increased to 80, semi-supervised feature extraction increases by 4.7 percent when compared to weakly guided semantic segmentation. Adding an attention mechanism to the critical network DRN (Deep Residual Network) can increase the classifier's performance by a small amount. Experiments revealed that the models worked effectively.

CONCLUSION

This paper investigates the segmentation and classification of cardiac fibrosis in DE-MRI data using a semi-supervised semantic segmentation and dual attention mechanism, dealing with the issue that existing segmentation algorithms have difficulty segmenting myocardial fibrosis tissue. In the future, we can consider optimizing the design of the attention module to reduce the module computation.

CMRSegTools: An open-source software enabling reproducible research in segmentation of acute myocardial infarct in CMR images

In the last decade, a large number of clinical trials have been deployed using Cardiac Magnetic Resonance (CMR) to evaluate cardioprotective strategies aiming at reducing the irreversible myocardial damage at the time of reperfusion. In these studies, segmentation and quantification of myocardial infarct lesion are often performed with a commercial software or an in-house closed-source code development thus creating a barrier for reproducible research. This paper introduces CMRSegTools: an open-source application software designed for the segmentation and quantification of myocardial infarct lesion enabling full access to state-of-the-art segmentation methods and parameters, easy integration of new algorithms and standardised results sharing. This post-processing tool has been implemented as a plug-in for the OsiriX/Horos DICOM viewer leveraging its database management functionalities and user interaction features to provide a bespoke tool for the analysis of cardiac MR images on large clinical cohorts. CMRSegTools includes, among others, user-assisted segmentation of the left-ventricle, semi- and automatic lesion segmentation methods, advanced statistical analysis and visualisation based on the American Heart Association 17-segment model. New segmentation methods can be integrated into the plug-in by developing components based on image processing and visualisation libraries such as ITK and VTK in C++ programming language. CMRSegTools allows the creation of training and testing data sets (labeled features such as lesion, microvascular obstruction and remote ROI) for supervised Machine Learning methods, and enables the comparative assessment of lesion segmentation methods via a single and integrated platform. The plug-in has been successfully used by several CMR imaging studies.

Cardiac segmentation on late gadolinium enhancement MRI: A benchmark study from multi-sequence cardiac MR segmentation challenge

Accurate computing, analysis and modeling of the ventricles and myocardium from medical images are important, especially in the diagnosis and treatment management for patients suffering from myocardial infarction (MI). Late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) provides an important protocol to visualize MI. However, compared with the other sequences LGE CMR images with gold standard labels are particularly limited. This paper presents the selective results from the Multi-Sequence Cardiac MR (MS-CMR) Segmentation challenge, in conjunction with MICCAI 2019. The challenge offered a data set of paired MS-CMR images, including auxiliary CMR sequences as well as LGE CMR, from 45 patients who underwent cardiomyopathy. It was aimed to develop new algorithms, as well as benchmark existing ones for LGE CMR segmentation focusing on myocardial wall of the left ventricle and blood cavity of the two ventricles. In addition, the paired MS-CMR images could enable algorithms to combine the complementary information from the other sequences for the ventricle segmentation of LGE CMR. Nine representative works were selected for evaluation and comparisons, among which three methods are unsupervised domain adaptation (UDA) methods and the other six are supervised. The results showed that the average performance of the nine methods was comparable to the inter-observer variations. Particularly, the top-ranking algorithms from both the supervised and UDA methods could generate reliable and robust segmentation results. The success of these methods was mainly attributed to the inclusion of the auxiliary sequences from the MS-CMR images, which provide important label information for the training of deep neural networks. The challenge continues as an ongoing resource, and the gold standard segmentation as well as the MS-CMR images of both the training and test data are available upon registration via its homepage (www.sdspeople.fudan.edu.cn/zhuangxiahai/0/mscmrseg/).

Artificial intelligence study on left ventricular function among normal individuals, hypertrophic cardiomyopathy and dilated cardiomyopathy patients using 1.5T cardiac cine MR images obtained by SSFP sequence

OBJECTIVES

To evaluate the performance of a deep learning-based method to automatically quantify left ventricular (LV) function from MR images in different cardiomyopathy.

METHODS

This retrospective study included MRI data sets from 2013 to 2020. Data on left ventricular function from patients with hypertrophic cardiomyopathy (HCM), patients with dilated cardiomyopathy (DCM), and healthy participants were analyzed. MRI data from a total of 388 patients were measured manually and automatically.The performance of Convolutional Neural Networks (CNNs) was evaluated based on the manual notes of two experienced observers: (a) LV segmentation accuracy, and (b) LV functional parameter accuracy. Bland-Altman analysis, Receiver operating Characteristic (ROC) curve analysis and Pearson correlation analysis were used to evaluate the consistency between fully automatic and manual diagnosis of HCM and DCM.

RESULTS

The deep-learning CNN performed best in HCM in evaluating LV function and worst in DCM. Compared with manual analysis, four parameters of LV function in the HCM group showed high correlation (r at least >0.901), and the correlation of DCM in all parameters was weaker than that of HCM, especially EF (r2 = 0.776) and SV (r2 = 0.645). ROC curve analysis indicated that at the optimal cut-off value, EF from automatic segmentation identified DCM and HCM patients with sensitivity of 92.31 and 78.05%, specificity of 82.96 and 54.07%, respectively.

CONCLUSION

Among different heart diseases, the analysis of cardiac function based on deep-learning CNN may have different performances, with DCM performing the worst and HCM the best and thus, special attention should be paid to DCM patients when assessing LV function through artificial intelligence method. LV function parameter obtained by artificial intelligence method may play an important role in the future AI diagnosis of HCM and DCM.

ADVANCES IN KNOWLEDGE

These data for the first time objectively evaluate the performance of a commercially available deep learning-based method in cardiac function evaluation of different cardiomyopathy and point out its advantages and disadvantages in different cardiomyopathy. This work did not attempt to design the algorithm itself, but rather applied an already existing method to a test dataset of clinical data and evaluated the results for a limited number of cardiomyopathy.

An Explainable Machine Learning Approach Reveals Prognostic Significance of Right Ventricular Dysfunction in Nonischemic Cardiomyopathy

OBJECTIVES

The authors implemented an explainable machine learning (ML) model to gain insight into the association between cardiac magnetic resonance markers and adverse outcomes of cardiovascular hospitalization and all-cause death (composite endpoint) in patients with nonischemic dilated cardiomyopathy (NICM).

BACKGROUND

Risk stratification of patients with NICM remains challenging. An explainable ML model has the potential to provide insight into the contributions of different risk markers in the prediction model.

METHODS

An explainable ML model based on extreme gradient boosting (XGBoost) machines was developed using cardiac magnetic resonance and clinical parameters. The study cohorts consist of patients with NICM from 2 academic medical centers: Beth Israel Deaconess Medical Center (BIDMC) and Brigham and Women's Hospital (BWH), with 328 and 214 patients, respectively. XGBoost was trained on 70% of patients from the BIDMC cohort and evaluated based on the other 30% as internal validation. The model was externally validated using the BWH cohort. To investigate the contribution of different features in our risk prediction model, we used Shapley additive explanations (SHAP) analysis.

RESULTS

During a mean follow-up duration of 40 months, 34 patients from BIDMC and 33 patients from BWH experienced the composite endpoint. The area under the curve for predicting the composite endpoint was 0.71 for the internal BIDMC validation and 0.69 for the BWH cohort. SHAP analysis identified parameters associated with right ventricular (RV) dysfunction and remodeling as primary markers of adverse outcomes. High risk thresholds were identified by SHAP analysis and thus provided thresholds for top predictive continuous clinical variables.

CONCLUSIONS

An explainable ML-based risk prediction model has the potential to identify patients with NICM at risk for cardiovascular hospitalization and all-cause death. RV ejection fraction, end-systolic and end-diastolic volumes (as indicators of RV dysfunction and remodeling) were determined to be major risk markers.

The role of artificial intelligence in paediatric cardiovascular magnetic resonance imaging

Artificial intelligence (AI) offers the potential to change many aspects of paediatric cardiac imaging. At present, there are only a few clinically validated examples of AI applications in this field. This review focuses on the use of AI in paediatric cardiovascular MRI, using examples from paediatric cardiovascular MRI, adult cardiovascular MRI and other radiologic experience.

The Role of AI in Characterizing the DCM Phenotype

Dilated Cardiomyopathy is conventionally defined by left ventricular dilatation and dysfunction in the absence of coronary disease. Emerging evidence suggests many patients remain vulnerable to major adverse outcomes despite clear therapeutic success of modern evidence-based heart failure therapy. In this era of personalized medical care, the conventional assessment of left ventricular ejection fraction falls short in fully predicting evolution and risk of outcomes in this heterogenous group of heart muscle disease, as such, a more refined means of phenotyping this disease appears essential. Cardiac MRI (CMR) is well-placed in this respect, not only for its diagnostic utility, but the wealth of information captured in global and regional function assessment with the addition of unique tissue characterization across different disease states and patient cohorts. Advanced tools are needed to leverage these sensitive metrics and integrate with clinical, genetic and biochemical information for personalized, and more clinically useful characterization of the dilated cardiomyopathy phenotype. Recent advances in artificial intelligence offers the unique opportunity to impact clinical decision making through enhanced precision image-analysis tasks, multi-source extraction of relevant features and seamless integration to enhance understanding, improve diagnosis, and subsequently clinical outcomes. Focusing particularly on deep learning, a subfield of artificial intelligence, that has garnered significant interest in the imaging community, this paper reviews the main developments that could offer more robust disease characterization and risk stratification in the Dilated Cardiomyopathy phenotype. Given its promising utility in the non-invasive assessment of cardiac diseases, we firstly highlight the key applications in CMR, set to enable comprehensive quantitative measures of function beyond the standard of care assessment. Concurrently, we revisit the added value of tissue characterization techniques for risk stratification, showcasing the deep learning platforms that overcome limitations in current clinical workflows and discuss how they could be utilized to better differentiate at-risk subgroups of this phenotype. The final section of this paper is dedicated to the allied clinical applications to imaging, that incorporate artificial intelligence and have harnessed the comprehensive abundance of data from genetics and relevant clinical variables to facilitate better classification and enable enhanced risk prediction for relevant outcomes.

Detection of Cardiac Tissues using K-means Analysis Methods in Nuclear Medicine Images

BACKGROUND: Nuclear cardiology uses to diagnose the cardiac disorders such as ischemic and inflammation disorders. In cardiac scintigraphy, unraveling closely adjacent tissues in the image are challenging issue. AIM: The aim of the study is to detect of cardiac tissues using K-means analysis methods in nuclear medicine images. This study also aimed to reduce the existence of fleck noise that disturbs the contrast and make its analysis more difficult. METHODS: Thus, digital image processing uses to increase the detection rate of myocardium easily using its color-based algorithms. In this study, color-based K-means was used. The scintographs were converted into color space presentation. Then, each pixel in the image was segmented using color analysis algorithms. RESULTS: The segmented scintograph was displayed in distinct fresh image. The proposed technique defines the myocardial tissues and borders precisely. Both exactness rate and recall reckoning were calculated. The results were 97.3 + 8.46 (p > 0.05). CONCLUSION: The proposed technique offered recognition of the heart tissue with high exactness amount.

Automatic deep learning-based myocardial infarction segmentation from delayed enhancement MRI

Delayed Enhancement cardiac MRI (DE-MRI) has become indispensable for the diagnosis of myocardial diseases. However, to quantify the disease severity, doctors need time to manually annotate the scar and myocardium. To address this issue, in this paper we propose an automatic myocardial infarction segmentation approach on the left ventricle from short-axis DE-MRI based on Convolutional Neural Networks (CNN). The objective is to segment myocardial infarction on short-axis DE-MRI images of the left ventricle acquired 10 min after the injection of a gadolinium-based contrast agent. The segmentation of the infarction area is realized in two stages: a first CNN model finds the contour of myocardium and a second CNN model segments the infarction. Compared to the manual intra-observer and inter-observer variations for the segmentation of myocardial infarction, and to the automatic segmentation with Gaussian Mixture Model, our proposal achieves satisfying segmentation results on our dataset of 904 DE-MRI slices.

MA-SOCRATIS: An automatic pipeline for robust segmentation of the left ventricle and scar

Multi-atlas segmentation of cardiac regions and total infarct scar (MA-SOCRATIS) is an unsupervised automatic pipeline to segment left ventricular myocardium and scar from late gadolinium enhanced MR images (LGE-MRI) of the heart. We implement two different pipelines for myocardial and scar segmentation from short axis LGE-MRI. Myocardial segmentation has two steps; initial segmentation and re-estimation. The initial segmentation step makes a first estimate of myocardium boundaries by using multi-atlas segmentation techniques. The re-estimation step refines the myocardial segmentation by a combination of k-means clustering and a geometric median shape variation technique. An active contour technique determines the unhealthy and healthy myocardial wall. The scar segmentation pipeline is a combination of a Rician-Gaussian mixture model and full width at half maximum (FWHM) thresholding, to determine the intensity pixels in scar regions. Following this step a watershed method with an automatic seed-points framework segments the final scar region. MA-SOCRATIS was evaluated using two different datasets. In both datasets ground truths were based on manual segmentation of short axis images from LGE-MRI scans. The first dataset included 40 patients from the MS-CMRSeg 2019 challenge dataset (STACOM at MICCAI 2019). The second is a collection of 20 patients with scar regions that are challenging to segment. MA-SOCRATIS achieved robust and accurate performance in automatic segmentation of myocardium and scar regions without the need of training or tuning in both cohorts, compared with state-of-the-art techniques (intra-observer and inter observer myocardium segmentation: 81.9% and 70% average Dice value, and scar (intra-observer and inter observer segmentation: 70.5% and 70.5% average Dice value).

The DAPA Trial in the Context of Previous Prophylactic ICD Landmark Trials

In patients with ischaemic cardiomyopathy and severely reduced left ventricular ejection fraction (LVEF), an arrhythmogenic milieu is created by a complex interplay between myocardial scarring (assessed by cardiac MRI) and multiple other factors (ventricular ectopy, ischaemia and autonomic imbalance), favouring the occurrence of arrhythmic sudden cardiac death (SCD). Currently, a dynamic and robust model of dichotomised SCD risk assessment after primary percutaneous coronary intervention (PCI) is lacking, underlining the urgent need for further refinement of the widely accepted and guidelines-based criteria (ischaemic cardiomyopathy, LVEF ≤35%) for primary prevention. This review addresses the potential additional value of the recently published Defibrillator After Primary Angioplasty (DAPA) trial results. The DAPA trial conveys important messages and provides novel perspectives regarding left ventricular function post-primary PCI as an (early) risk marker for SCD and the impact of prophylactic ICD implantation on survival in this cohort. In the context of other previous primary prevention trials, DAPA was the first trial including only ST-elevation MI patients all treated with acute PCI.

Recent Advances in Fibrosis and Scar Segmentation From Cardiac MRI: A State-of-the-Art Review and Future Perspectives

Segmentation of cardiac fibrosis and scars is essential for clinical diagnosis and can provide invaluable guidance for the treatment of cardiac diseases. Late Gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) has been successful in guiding the clinical diagnosis and treatment reliably. For LGE CMR, many methods have demonstrated success in accurately segmenting scarring regions. Co-registration with other non-contrast-agent (non-CA) modalities [e.g., balanced steady-state free precession (bSSFP) cine magnetic resonance imaging (MRI)] can further enhance the efficacy of automated segmentation of cardiac anatomies. Many conventional methods have been proposed to provide automated or semi-automated segmentation of scars. With the development of deep learning in recent years, we can also see more advanced methods that are more efficient in providing more accurate segmentations. This paper conducts a state-of-the-art review of conventional and current state-of-the-art approaches utilizing different modalities for accurate cardiac fibrosis and scar segmentation.

Improved Quantification of Myocardium Scar in Late Gadolinium Enhancement Images: Deep Learning Based Image Fusion Approach

BACKGROUND

Quantification of myocardium scarring in late gadolinium enhanced (LGE) cardiac magnetic resonance imaging can be challenging due to low scar-to-background contrast and low image quality. To resolve ambiguous LGE regions, experienced readers often use conventional cine sequences to accurately identify the myocardium borders.

PURPOSE

To develop a deep learning model for combining LGE and cine images to improve the robustness and accuracy of LGE scar quantification.

STUDY TYPE

Retrospective.

POPULATION

A total of 191 hypertrophic cardiomyopathy patients: 1) 162 patients from two sites randomly split into training (50%; 81 patients), validation (25%, 40 patients), and testing (25%; 41 patients); and 2) an external testing dataset (29 patients) from a third site.

FIELD STRENGTH/SEQUENCE

1.5T, inversion-recovery segmented gradient-echo LGE and balanced steady-state free-precession cine sequences ASSESSMENT: Two convolutional neural networks (CNN) were trained for myocardium and scar segmentation, one with and one without LGE-Cine fusion. For CNN with fusion, the input was two aligned LGE and cine images at matched cardiac phase and anatomical location. For CNN without fusion, only LGE images were used as input. Manual segmentation of the datasets was used as reference standard.

STATISTICAL TESTS

Manual and CNN-based quantifications of LGE scar burden and of myocardial volume were assessed using Pearson linear correlation coefficients (r) and Bland-Altman analysis.

RESULTS

Both CNN models showed strong agreement with manual quantification of LGE scar burden and myocardium volume. CNN with LGE-Cine fusion was more robust than CNN without LGE-Cine fusion, allowing for successful segmentation of significantly more slices (603 [95%] vs. 562 (89%) of 635 slices; P < 0.001). Also, CNN with LGE-Cine fusion showed better agreement with manual quantification of LGE scar burden than CNN without LGE-Cine fusion (%ScarLGE-cine = 0.82 × %Scarmanual , r = 0.84 vs. %ScarLGE  = 0.47 × %Scarmanual , r = 0.81) and myocardium volume (VolumeLGE-cine = 1.03 × Volumemanual , r = 0.96 vs. VolumeLGE  = 0.91 × Volumemanual , r = 0.91).

DATA CONCLUSION

CNN based LGE-Cine fusion can improve the robustness and accuracy of automated scar quantification.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: 1.

Whole-Heart High-Resolution Late Gadolinium Enhancement: Techniques and Clinical Applications

In cardiovascular magnetic resonance, late gadolinium enhancement (LGE) has become the cornerstone of myocardial tissue characterization. It is widely used in clinical routine to diagnose and characterize the myocardial tissue in a wide range of ischemic and nonischemic cardiomyopathies. The recent growing interest in imaging left atrial fibrosis has led to the development of novel whole‐heart high‐resolution late gadolinium enhancement (HR‐LGE) techniques. Indeed, conventional LGE is acquired in multiple breath‐holds with limited spatial resolution: ~1.4–1.8 mm in plane and 6–8 mm slice thickness, according to the Society for Cardiovascular Magnetic Resonance standardized guidelines. Such large voxel size prevents its use in thin structures such as the atrial or right ventricular walls. Whole‐heart 3D HR‐LGE images are acquired in free breathing to increase the spatial resolution (up to 1.3 × 1.3 × 1.3 mm³) and offer a better detection and depiction of focal atrial fibrosis. The downside of this increased resolution is the extended scan time of around 10 min, which hampers the spread of HR‐LGE in clinical practice. Initially introduced for atrial fibrosis imaging, HR‐LGE interest has evolved to be a tool to detect small scars in the ventricles and guide ablation procedures. Indeed, the detection of scars, nonvisible with conventional LGE, can be crucial in the diagnosis of myocardial infarction with nonobstructed coronary arteries, in the detection of the arrhythmogenic substrate triggering ventricular arrhythmia, and improve the confidence of clinicians in the challenging diagnoses such as the arrhythmogenic right ventricular cardiomyopathy. HR‐LGE also offers a precise visualization of left ventricular scar morphology that is particularly useful in planning ablation procedures and guiding them through the fusion of HR‐LGE images with electroanatomical mapping systems. In this narrative review, we attempt to summarize the technical particularities of whole‐heart HR‐LGE acquisition and provide an overview of its clinical applications with a particular focus on the ventricles.

Automatic left ventricular cavity segmentation via deep spatial sequential network in 4D computed tomography

Automated segmentation of left ventricular cavity (LVC) in temporal cardiac image sequences (consisting of multiple time-points) is a fundamental requirement for quantitative analysis of cardiac structural and functional changes. Deep learning methods for segmentation are the state-of-the-art in performance; however, these methods are generally formulated to work on a single time-point, and thus disregard the complementary information available from the temporal image sequences that can aid in segmentation accuracy and consistency across the time-points. In particular, single time-point segmentation methods perform poorly in segmenting the end-systole (ES) phase image in the cardiac sequence, where the left ventricle deforms to the smallest irregular shape, and the boundary between the blood chamber and the myocardium becomes inconspicuous and ambiguous. To overcome these limitations in automatically segmenting temporal LVCs, we present a spatial sequential network (SS-Net) to learn the deformation and motion characteristics of the LVCs in an unsupervised manner; these characteristics are then integrated with sequential context information derived from bi-directional learning (BL) where both chronological and reverse-chronological directions of the image sequence are used. Our experimental results on a cardiac computed tomography (CT) dataset demonstrate that our spatial-sequential network with bi-directional learning (SS-BL-Net) outperforms existing methods for spatiotemporal LVC segmentation.

Machine learning and network medicine: a novel approach for precision medicine and personalized therapy in cardiomyopathies

The early identification of pathogenic mechanisms is essential to predict the incidence and progression of cardiomyopathies and to plan appropriate preventive interventions. Noninvasive cardiac imaging such as cardiac computed tomography, cardiac magnetic resonance, and nuclear imaging plays an important role in diagnosis and management of cardiomyopathies and provides useful prognostic information. Most molecular factors exert their functions by interacting with other cellular components, thus many diseases reflect perturbations of intracellular networks. Indeed, complex diseases and traits such as cardiomyopathies are caused by perturbations of biological networks. The network medicine approach, by integrating systems biology, aims to identify pathological interacting genes and proteins, revolutionizing the way to know cardiomyopathies and shifting the understanding of their pathogenic phenomena from a reductionist to a holistic approach. In addition, artificial intelligence tools, applied to morphological and functional imaging, could allow imaging scans to be automatically analyzed to extract new parameters and features for cardiomyopathy evaluation. The aim of this review is to discuss the tools of network medicine in cardiomyopathies that could reveal new candidate genes and artificial intelligence imaging-based features with the aim to translate into clinical practice as diagnostic, prognostic, and predictive biomarkers and shed new light on the clinical setting of cardiomyopathies. The integration and elaboration of clinical habits, molecular big data, and imaging into machine learning models could provide better disease phenotyping, outcome prediction, and novel drug targets, thus opening a new scenario for the implementation of precision medicine for cardiomyopathies.

Fully Automatic Scar Segmentation for Late Gadolinium Enhancement MRI Images in Left Ventricle with Myocardial Infarction

Numerous methods have been published to segment the infarct tissue in the left ventricle, most of them either need manual work, post-processing, or suffer from poor reproducibility. We proposed an automatic segmentation method for segmenting the infarct tissue in left ventricle with myocardial infarction. Cardiac images of a total of 60 diseased hearts (55 human hearts and 5 porcine hearts) were used in this study. The epicardial and endocardial boundaries of the ventricles in every 2D slice of the cardiac magnetic resonance with late gadolinium enhancement images were manually segmented. The subsequent pipeline of infarct tissue segmentation is fully automatic. The segmentation results with the automatic algorithm proposed in this paper were compared to the consensus ground truth. The median of Dice overlap between our automatic method and the consensus ground truth is 0.79. We also compared the automatic method with the consensus ground truth using different image sources from different centers with different scan parameters and different scan machines. The results showed that the Dice overlap with the public dataset was 0.83, and the overall Dice overlap was 0.79. The results show that our method is robust with respect to different MRI image sources, which were scanned by different centers with different image collection parameters. The segmentation accuracy we obtained is comparable to or better than that of the conventional semi-automatic methods. Our segmentation method may be useful for processing large amount of dataset in clinic.

Cine and late gadolinium enhancement MRI registration and automated myocardial infarct heterogeneity quantification

PURPOSE

To develop an approach for automated quantification of myocardial infarct heterogeneity in late gadolinium enhancement (LGE) cardiac MRI.

METHODS

We acquired 2D short-axis cine and 3D LGE in 10 pigs with myocardial infarct. The 2D cine myocardium was segmented and registered to the LGE images. LGE image signal intensities within the warped cine myocardium masks were analyzed to determine the thresholds of infarct core (IC) and gray zone (GZ) for the standard-deviation (SD) and full-width-at-halfmaximum (FWHM) methods. The initial IC, GZ, and IC + GZ segmentations were postprocessed using a normalized cut approach. Cine segmentation and cine-LGE registration accuracies were evaluated using dice similarity coefficient and average symmetric surface distance. Automated IC, GZ, and IC + GZ volumes were compared with manual results using Pearson correlation coefficient (r), Bland-Altman analyses, and intraclass correlation coefficient.

RESULTS

For n = 87 slices containing scar, we achieved cine segmentation dice similarity coefficient = 0.87 ± 0.12, average symmetric surface distance = 0.94 ± 0.74 mm (epicardium), and 1.03 ± 0.82 mm (endocardium) in the scar region. For cine-LGE registration, dice similarity coefficient was 0.90 ± 0.06 and average symmetric surface distance was 0.72 ± 0.39 mm (epicardium) and 0.86 ± 0.53 mm (endocardium) in the scar region. For both SD and FWHM methods, automated IC, GZ, and IC + GZ volumes were strongly (r > 0.70) correlated with manual measurements, and the correlations were not significantly different from interobserver correlations (P > .05). The agreement between automated and manual scar volumes (intraclass correlation coefficient = 0.85-0.96) was similar to that between two observers (intraclass correlation coefficient = 0.81-0.99); automated scar segmentation errors were not significantly different from interobserver segmentation differences (P > .05).

CONCLUSIONS

Our approach provides fully automated cine-LGE MRI registration and LGE myocardial infarct heterogeneity quantification in preclinical studies.

Comparison of Segmentation Algorithms for Detecting Myocardial Infarction Using Late Gadolinium Enhancement Magnetic Resonance Imaging

Objective: The aim of this study was to validate the accuracy of a new automatic method for scar segmentation and compare its performance with that of two other frequently used segmentation algorithms.Methods: Twenty-six late gadolinium enhancement cardiovascular magnetic resonance images of diseased hearts were segmented by the full width at half maximum (FWHM) method, the n standard deviations (nSD) method, and our new automatic method. The results of the three methods were compared with the consensus ground truth obtained by manual segmentation of the ventricular boundaries.Results: Our automatic method yielded the highest Dice score and the lowest volume difference compared with the consensus ground truth segmentation. The nSD method produced large variations in the Dice score and the volume difference. The FWHM method yielded the lowest Dice score and the greatest volume difference compared with the automatic, 6SD, and 8SD methods, but resulted in less variation when different observers segmented the images.Conclusion: The automatic method introduced in this study is highly reproducible and objective. Because it requires no manual intervention, it may be useful for processing large datasets produced in clinical applications.

Emidec: A Database Usable for the Automatic Evaluation of Myocardial Infarction from Delayed-Enhancement Cardiac MRI

One crucial parameter to evaluate the state of the heart after myocardial infarction (MI) is the viability of the myocardial segment, i.e., if the segment recovers its functionality upon revascularization. MRI performed several minutes after the injection of a contrast agent (delayed enhancement-MRI or DE-MRI) is a method of choice to evaluate the extent of MI, and by extension, to assess viable tissues after an injury. The Emidec dataset is composed of a series of exams with DE-MR images in short axis orientation covering the left ventricle from normal cases or patients with myocardial infarction, with the contouring of the myocardium and diseased areas (if present) from experts in the domains. Moreover, classical available clinical parameters when the patient is managed by an emergency department are provided for each case. To the best of our knowledge, the Emidec dataset is the first one where annotated DE-MRI are combined with clinical characteristics of the patient, allowing the development of methodologies for exam classification as for exam quantification.

Artificial Intelligence and Myocardial Contrast Enhancement Pattern

PURPOSE OF REVIEW

Machine learning (ML) and deep learning (DL) are two important categories of AI algorithms. Nowadays, AI technology has been gradually applied to cardiac magnetic resonance imaging (CMRI), covering the fields of myocardial contrast enhancement (MCE) pattern and automatic ventricular segmentation. This paper mainly discusses the relationship between machine learning and deep learning based on AI and pattern of MCE in CMRI.

RECENT FINDINGS

It found that some histogram and GLCM parameters in ML algorithm had significant statistical differences in diagnosis of cardiomyopathy and differentiation of fibrosis and normal myocardial tissue. In the DL algorithm, there was no significant difference between CNN and observers in measuring myocardial fibrosis. The rapid development of texture parameter analysis methods would promote the medical imaging based on AI into a new era. Histogram and GLCM parameters are the research hotspot of unsupervised learning of MCE images. CNN has a great advantage in automatically identifying and quantifying myocardial fibrosis reflected by LGE images.

Machine Learning-Based Segmentation of Left Ventricular Myocardial Fibrosis from Magnetic Resonance Imaging

PURPOSE OF REVIEW

Myocardial fibrosis (MF) arises due to myocardial infarction and numerous cardiac diseases. MF may lead to several heart disorders, such as heart failure, arrhythmias, and ischemia. Cardiac magnetic resonance (CMR) imaging techniques, such as late gadolinium enhancement (LGE) CMR, enable non-invasive assessment of MF in the left ventricle (LV). Manual assessment of MF on CMR is a tedious and time-consuming task that is subject to high observer variability. Automated segmentation and quantification of MF is important for risk stratification and treatment planning in patients with heart disorders. This article aims to review the machine learning (ML)-based methodologies developed for MF quantification in the LV using CMR images.

RECENT FINDINGS

With the availability of relatively large labeled datasets supervised learning methods based on both conventional ML and state-of-the-art deep learning (DL) methods have been successfully applied for automated segmentation of MF. The incorporation of ML algorithms into imaging techniques such as 3D LGE CMR permits fast characterization of MF on CMR imaging and may enhance the diagnosis and prognosis of patients with heart disorders. Concurrently, the studies using cine CMR images have revealed that accurate segmentation of MF on non-contrast CMR imaging might be possible. The application of ML/DL tools in CMR image interpretation is likely to result in accurate and efficient quantification of MF.

Deep Learning for Cardiac Image Segmentation: A Review

Deep learning has become the most widely used approach for cardiac image segmentation in recent years. In this paper, we provide a review of over 100 cardiac image segmentation papers using deep learning, which covers common imaging modalities including magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound and major anatomical structures of interest (ventricles, atria, and vessels). In addition, a summary of publicly available cardiac image datasets and code repositories are included to provide a base for encouraging reproducible research. Finally, we discuss the challenges and limitations with current deep learning-based approaches (scarcity of labels, model generalizability across different domains, interpretability) and suggest potential directions for future research.

Biomechanics of infarcted left Ventricle-A review of experiments

Myocardial infarction (MI) is one of leading diseases to contribute to annual death rate of 5% in the world. In the past decades, significant work has been devoted to this subject. Biomechanics of infarcted left ventricle (LV) is associated with MI diagnosis, understanding of remodelling, MI micro-structure and biomechanical property characterizations as well as MI therapy design and optimization, but the subject has not been reviewed presently. In the article, biomechanics of infarcted LV was reviewed in terms of experiments achieved in the subject so far. The concerned content includes experimental remodelling, kinematics and kinetics of infarcted LVs. A few important issues were discussed and several essential topics that need to be investigated further were summarized. Microstructure of MI tissue should be observed even carefully and compared between different methods for producing MI scar in the same animal model, and eventually correlated to passive biomechanical property by establishing innovative constitutive laws. More uniaxial or biaxial tensile tests are desirable on MI, border and remote tissues, and viscoelastic property identification should be performed in various time scales. Active contraction experiments on LV wall with MI should be conducted to clarify impaired LV pumping function and supply necessary data to the function modelling. Pressure-volume curves of LV with MI during diastole and systole for the human are also desirable to propose and validate constitutive laws for LV walls with MI.

Three-dimensional Deep Convolutional Neural Networks for Automated Myocardial Scar Quantification in Hypertrophic Cardiomyopathy: A Multicenter Multivendor Study

Background Cardiac MRI late gadolinium enhancement (LGE) scar volume is an important marker for outcome prediction in patients with hypertrophic cardiomyopathy (HCM); however, its clinical application is hindered by a lack of measurement standardization. Purpose To develop and evaluate a three-dimensional (3D) convolutional neural network (CNN)-based method for automated LGE scar quantification in patients with HCM. Materials and Methods We retrospectively identified LGE MRI data in a multicenter (n = 7) and multivendor (n = 3) HCM study obtained between November 2001 and November 2011. A deep 3D CNN based on U-Net architecture was used for LGE scar quantification. Independent CNN training and testing data sets were maintained with a 4:1 ratio. Stacks of short-axis MRI slices were split into overlapping substacks that were segmented and then merged into one volume. The 3D CNN per-site and per-vendor performances were evaluated with respect to manual scar quantification performed in a core laboratory setting using Dice similarity coefficient (DSC), Pearson correlation, and Bland-Altman analyses. Furthermore, the performance of 3D CNN was compared with that of two-dimensional (2D) CNN. Results This study included 1073 patients with HCM (733 men; mean age, 49 years ± 17 [standard deviation]). The 3D CNN-based quantification was fast (0.15 second per image) and demonstrated excellent correlation with manual scar volume quantification (r = 0.88, P < .001) and ratio of scar volume to total left ventricle myocardial volume (%LGE) (r = 0.91, P < .001). The 3D CNN-based quantification strongly correlated with manual quantification of scar volume (r = 0.82-0.99, P < .001) and %LGE (r = 0.90-0.97, P < .001) for all sites and vendors. The 3D CNN identified patients with a large scar burden (>15%) with 98% accuracy (202 of 207) (95% confidence interval [CI]: 95%, 99%). When compared with 3D CNN, 2D CNN underestimated scar volume (r = 0.85, P < .001) and %LGE (r = 0.83, P < .001). The DSC of 3D CNN segmentation was comparable among different vendors (P = .07) and higher than that of 2D CNN (DSC, 0.54 ± 0.26 vs 0.48 ± 0.29; P = .02). Conclusion In the hypertrophic cardiomyopathy population, a three-dimensional convolutional neural network enables fast and accurate quantification of myocardial scar volume, outperforms a two-dimensional convolutional neural network, and demonstrates comparable performance across different vendors. © RSNA, 2019 Online supplemental material is available for this article.

Myocardial Radiomics in Cardiac MRI

OBJECTIVE. The purpose of this article is to review the nascent field of radiomics in cardiac MRI. CONCLUSION. Cardiac MRI produces a large number of images in a fairly inefficient manner with sometimes limited clinical application. In the era of precision medicine, there is increasing need for imaging to account for a broader array of diseases in an efficient and objective manner. Radiomics, the extraction and analysis of quantitative imaging features from medical imaging, may offer potential solutions to this need.

Influence of the distribution of fibrosis within an area of myocardial infarction on wave propagation in ventricular tissue

The presence of fibrosis in heart tissue is strongly correlated with an incidence of arrhythmia, which is a leading cause of sudden cardiac death (SCD). However, it remains incompletely understood how different distributions, sizes and positions of fibrotic tissues contribute to arrhythmogenesis. In this study, we designed 4 different ventricular models mimicking wave propagation in cardiac tissues under normal, myocardial infarction (MI), MI with random fibrosis and MI with gradient fibrosis conditions. Simulation results of ideal square tissues indicate that vulnerable windows (VWs) of random and gradient fibrosis distributions are similar with low levels of fibrosis. However, with a high level of fibrosis, the VWs significantly increase in random fibrosis tissue but not in gradient fibrosis tissue. In addition, we systematically analyzed the effects of the size and position of fibrosis tissues on VWs. Simulation results show that it is more likely for a reentry wave to appear when the length of the infarcted area is greater than 25% of the perimeter of the ventricle, when the width is approximately half that of the ventricular wall, or when the infarcted area is attached to the inside or outside of the ventricular wall.

Artificial Intelligence Will Transform Cardiac Imaging-Opportunities and Challenges

Artificial intelligence (AI) using machine learning techniques will change healthcare as we know it. While healthcare AI applications are currently trailing behind popular AI applications, such as personalized web-based advertising, the pace of research and deployment is picking up and about to become disruptive. Overcoming challenges such as patient and public support, transparency over the legal basis for healthcare data use, privacy preservation, technical challenges related to accessing large-scale data from healthcare systems not designed for Big Data analysis, and deployment of AI in routine clinical practice will be crucial. Cardiac imaging and imaging of other body parts is likely to be at the frontier for the development of applications as pattern recognition and machine learning are a significant strength of AI with practical links to image processing. Many opportunities in cardiac imaging exist where AI will impact patients, medical staff, hospitals, commissioners and thus, the entire healthcare system. This perspective article will outline our vision for AI in cardiac imaging with examples of potential applications, challenges and some lessons learnt in recent years.

Computationally guided personalized targeted ablation of persistent atrial fibrillation

Atrial fibrillation (AF) — the most common arrhythmia — significantly increases the risk of stroke and heart failure. Although catheter ablation can restore normal heart rhythms, patients with persistent AF who develop atrial fibrosis often undergo multiple failed ablations and thus increased procedural risks. Here, we present personalized computational modelling for the reliable predetermination of ablation targets, which are then used to guide the ablation procedure in patients with persistent AF and atrial fibrosis. We first show that a computational model of the atria of patients identifies fibrotic tissue that if ablated will not sustain AF. We then integrated the target-ablation sites in a clinical-mapping system, and tested its feasibility in 10 patients with persistent AF. The computational prediction of ablation targets avoids lengthy electrical mapping and could improve the accuracy and efficacy of targeted AF ablation in patients whilst eliminating the need for repeat procedures.

Artificial intelligence in cardiovascular imaging: state of the art and implications for the imaging cardiologist

Healthcare, conceivably more than any other area of human endeavour, has the greatest potential to be affected by artificial intelligence (AI). This potential has been shown by several reports that demonstrate equal or superhuman performance in medical tasks that aim to improve efficiency, diagnosis and prognosis. This review focuses on the state of the art of AI applications in cardiovascular imaging. It provides an overview of the current applications and studies performed, including the potential value, implications, limitations and future directions of AI in cardiovascular imaging.It is envisioned that AI will dramatically change the way doctors practise medicine. In the short term, it will assist physicians with easy tasks, such as automating measurements, making predictions based on big data, and putting clinical findings into an evidence-based context. In the long term, AI will not only assist doctors, it has the potential to significantly improve access to health and well-being data for patients and their caretakers. This empowers patients. From a physician's perspective, reliable AI assistance will be available to support clinical decision-making. Although cardiovascular studies implementing AI are increasing in number, the applications have only just started to penetrate contemporary clinical care.

Benchmarking Wilms' tumor in multisequence MRI data: why does current clinical practice fail? Which popular segmentation algorithms perform well?

Wilms' tumor is one of the most frequent malignant solid tumors in childhood. Accurate segmentation of tumor tissue is a key step during therapy and treatment planning. Since it is difficult to obtain a comprehensive set of tumor data of children, there is no benchmark so far allowing evaluation of the quality of human or computer-based segmentations. The contributions in our paper are threefold: (i) we present the first heterogeneous Wilms' tumor benchmark data set. It contains multisequence MRI data sets before and after chemotherapy, along with ground truth annotation, approximated based on the consensus of five human experts. (ii) We analyze human expert annotations and interrater variability, finding that the current clinical practice of determining tumor volume is inaccurate and that manual annotations after chemotherapy may differ substantially. (iii) We evaluate six computer-based segmentation methods, ranging from classical approaches to recent deep-learning techniques. We show that the best ones offer a quality comparable to human expert annotations.

Creating three dimensional models of the right ventricular outflow tract: influence of contrast, sequence, operator, and threshold

The use of 3D printed models of the right ventricular outflow tract (RVOT) for surgical and interventional planning is growing and often requires image segmentation of cardiac magnetic resonance (CMR) images. Segmentation results may vary based on contrast, image sequence, signal threshold chosen by the operator, and manual post-processing. The purpose of this study was to determine potential biases and post-processing errors in image segmentation to enable informed decisions. Models of the RVOT and pulmonary arteries from twelve patients who had contrast enhanced CMR angiography with gadopentetate dimeglumine (GPD), gadofosveset trisodium (GFT), and a post-GFT inversion-recovery (IR) whole heart sequence were segmented, trimmed, and aligned by three operators. Geometric agreement and minimal RVOT diameters were compared between sequences and operators. To determine the contribution of threshold, interoperator variability was compared between models created by the same two operators using the same versus different thresholds. Geometric agreement by Dice between objects was high (intraoperator: 0.89-0.95; interoperator: 0.95-0.97), without differences between sequences. Minimal RVOT diameters differed on average by - 1.9 to - 1.3 mm (intraoperator) and by 0.4 to 1.4 mm (interoperator). The contribution of threshold to interoperator geometric agreement was not significant (same threshold: 0.96 ± 0.06, different threshold: 0.93 ± 0.05; p = 0.181), but minimal RVOT diameters were more variable with different versus constant thresholds (- 9.12% vs. 2.42%; p < 0.05). Thresholding does not significantly change interoperator variability for geometric agreement, but does for minimal RVOT diameter. Minimal RVOT diameters showed clinically relevant variation within and between operators.

Personalized Cardiac Computational Models: From Clinical Data to Simulation of Infarct-Related Ventricular Tachycardia

In the chronic stage of myocardial infarction, a significant number of patients develop life-threatening ventricular tachycardias (VT) due to the arrhythmogenic nature of the remodeled myocardium. Radiofrequency ablation (RFA) is a common procedure to isolate reentry pathways across the infarct scar that are responsible for VT. Unfortunately, this strategy show relatively low success rates; up to 50% of patients experience recurrent VT after the procedure. In the last decade, intensive research in the field of computational cardiac electrophysiology (EP) has demonstrated the ability of three-dimensional (3D) cardiac computational models to perform in-silico EP studies. However, the personalization and modeling of certain key components remain challenging, particularly in the case of the infarct border zone (BZ). In this study, we used a clinical dataset from a patient with a history of infarct-related VT to build an image-based 3D ventricular model aimed at computational simulation of cardiac EP, including detailed patient-specific cardiac anatomy and infarct scar geometry. We modeled the BZ in eight different ways by combining the presence or absence of electrical remodeling with four different levels of image-based patchy fibrosis (0, 10, 20, and 30%). A 3D torso model was also constructed to compute the ECG. Patient-specific sinus activation patterns were simulated and validated against the patient's ECG. Subsequently, the pacing protocol used to induce reentrant VTs in the EP laboratory was reproduced in-silico. The clinical VT was induced with different versions of the model and from different pacing points, thus identifying the slow conducting channel responsible for such VT. Finally, the real patient's ECG recorded during VT episodes was used to validate our simulation results and to assess different strategies to model the BZ. Our study showed that reduced conduction velocities and heterogeneity in action potential duration in the BZ are the main factors in promoting reentrant activity. Either electrical remodeling or fibrosis in a degree of at least 30% in the BZ were required to initiate VT. Moreover, this proof-of-concept study confirms the feasibility of developing 3D computational models for cardiac EP able to reproduce cardiac activation in sinus rhythm and during VT, using exclusively non-invasive clinical data.