Benchmark for Algorithms Segmenting the Left Atrium From 3D CT and MRI Datasets

Deep learning in 3D cardiac reconstruction: a systematic review of methodologies and dataset

This study presents an advanced methodology for 3D heart reconstruction using a combination of deep learning models and computational techniques, addressing critical challenges in cardiac modeling and segmentation. A multi-dataset approach was employed, including data from the UK Biobank, MICCAI Multi-Modality Whole Heart Segmentation (MM-WHS) challenge, and clinical datasets of congenital heart disease. Preprocessing steps involved segmentation, intensity normalization, and mesh generation, while the reconstruction was performed using a blend of statistical shape modeling (SSM), graph convolutional networks (GCNs), and progressive GANs. The statistical shape models were utilized to capture anatomical variations through principal component analysis (PCA), while GCNs refined the meshes derived from segmented slices. Synthetic data generated by progressive GANs enabled augmentation, particularly useful for congenital heart conditions. Evaluation of the reconstruction accuracy was performed using metrics such as Dice similarity coefficient (DSC), Chamfer distance, and Hausdorff distance, with the proposed framework demonstrating superior anatomical precision and functional relevance compared to traditional methods. This approach highlights the potential for automated, high-resolution 3D heart reconstruction applicable in both clinical and research settings. The results emphasize the critical role of deep learning in enhancing anatomical accuracy, particularly for rare and complex cardiac conditions. This paper is particularly important for researchers wanting to utilize deep learning in cardiac imaging and 3D heart reconstruction, bringing insights into the integration of modern computational methods.

The impact of atrial voltage and conduction velocity phenotypes on atrial fibrillation recurrence

Introduction

Low atrial voltage and slow conduction velocity (CV) have been associated with atrial fibrillation (AF); however, their interaction and relative importance as early disease markers remain incompletely understood. We aimed to elucidate the relationship between atrial voltage and CV using high-density electroanatomic (HDE) maps of patients with AF.

Methods

HDE maps obtained during sinus rhythm in 52 patients with AF and five healthy controls were analysed. Atrial voltage and CV maps were generated, and their correlations were assessed. Subgroup analyses were performed based on clinically relevant factors such as AF type, CV, and voltage levels. Finally, cluster analysis was conducted to identify distinct phenotypes within the population, reflecting different patterns of conduction and voltage.

Results

A moderate positive correlation was found between the mean atrial voltage and CV (r = 0.570). Subgroup analysis revealed differences in voltage (p = 0.0044) but not in global CV (p = 0.42), with no significant differences between AF types. Three distinct phenotypes emerged: normal voltage/normal CV, normal voltage/low CV, and low voltage/low CV, with distinct recurrence rates, suggesting different disease progression paths. Slower atrial CV was identified as a significant predictor of arrhythmia recurrence at 12 and 24 months after AF ablation, surpassing the predictive potential of atrial voltage.

Conclusion

Atrial voltage and CV analyses revealed distinct phenotypes. Lower atrial CV emerged as a significant predictor of AF recurrence, exceeding the predictive significance of atrial voltage. These findings emphasise the importance of considering CV and voltage in managing AF and offer potential insights for personalised strategies.

Learning Lossless Compression for High Bit-Depth Volumetric Medical Image

Recent advances in learning-based methods have markedly enhanced the capabilities of image compression. However, these methods struggle with high bit-depth volumetric medical images, facing issues such as degraded performance, increased memory demand, and reduced processing speed. To address these challenges, this paper presents the Bit-Division based Lossless Volumetric Image Compression (BD-LVIC) framework, which is tailored for high bit-depth medical volume compression. The BD-LVIC framework skillfully divides the high bit-depth volume into two lower bit-depth segments: the Most Significant Bit-Volume (MSBV) and the Least Significant Bit-Volume (LSBV). The MSBV concentrates on the most significant bits of the volumetric medical image, capturing vital structural details in a compact manner. This reduction in complexity greatly improves compression efficiency using traditional codecs. Conversely, the LSBV deals with the least significant bits, which encapsulate intricate texture details. To compress this detailed information effectively, we introduce an effective learning-based compression model equipped with a Transformer-Based Feature Alignment Module, which exploits both intra-slice and inter-slice redundancies to accurately align features. Subsequently, a Parallel Autoregressive Coding Module merges these features to precisely estimate the probability distribution of the least significant bit-planes. Our extensive testing demonstrates that the BD-LVIC framework not only sets new performance benchmarks across various datasets but also maintains a competitive coding speed, highlighting its significant potential and practical utility in the realm of volumetric medical image compression.

Novel Domain Knowledge-Encoding Algorithm Enables Label-Efficient Deep Learning for Cardiac CT Segmentation to Guide Atrial Fibrillation Treatment in a Pilot Dataset

Background: Segmenting computed tomography (CT) is crucial in various clinical applications, such as tailoring personalized cardiac ablation for managing cardiac arrhythmias. Automating segmentation through machine learning (ML) is hindered by the necessity for large, labeled training data, which can be challenging to obtain. This article proposes a novel approach for automated, robust labeling using domain knowledge to achieve high-performance segmentation by ML from a small training set. The approach, the domain knowledge-encoding (DOKEN) algorithm, reduces the reliance on large training datasets by encoding cardiac geometry while automatically labeling the training set. The method was validated in a hold-out dataset of CT results from an atrial fibrillation (AF) ablation study. Methods: The DOKEN algorithm parses left atrial (LA) structures, extracts "anatomical knowledge" by leveraging digital LA models (available publicly), and then applies this knowledge to achieve high ML segmentation performance with a small number of training samples. The DOKEN-labeled training set was used to train a nnU-Net deep neural network (DNN) model for segmenting cardiac CT in N = 20 patients. Subsequently, the method was tested in a hold-out set with N = 100 patients (five times larger than training set) who underwent AF ablation. Results: The DOKEN algorithm integrated with the nn-Unet model achieved high segmentation performance with few training samples, with a training to test ratio of 1:5. The Dice score of the DOKEN-enhanced model was 96.7% (IQR: 95.3% to 97.7%), with a median error in surface distance of boundaries of 1.51 mm (IQR: 0.72 to 3.12) and a mean centroid-boundary distance of 1.16 mm (95% CI: -4.57 to 6.89), similar to expert results (r = 0.99; p < 0.001). In digital hearts, the novel DOKEN approach segmented the LA structures with a mean difference for the centroid-boundary distances of -0.27 mm (95% CI: -3.87 to 3.33; r = 0.99; p < 0.0001). Conclusions: The proposed novel domain knowledge-encoding algorithm was able to perform the segmentation of six substructures of the LA, reducing the need for large training data sets. The combination of domain knowledge encoding and a machine learning approach could reduce the dependence of ML on large training datasets and could potentially be applied to AF ablation procedures and extended in the future to other imaging, 3D printing, and data science applications.

LinFlo-Net: A Two-Stage Deep Learning Method to Generate Simulation Ready Meshes of the Heart

We present a deep learning model to automatically generate computer models of the human heart from patient imaging data with an emphasis on its capability to generate thin-walled cardiac structures. Our method works by deforming a template mesh to fit the cardiac structures to the given image. Compared with prior deep learning methods that adopted this approach, our framework is designed to minimize mesh self-penetration, which typically arises when deforming surface meshes separated by small distances. We achieve this by using a two-stage diffeomorphic deformation process along with a novel loss function derived from the kinematics of motion that penalizes surface contact and interpenetration. Our model demonstrates comparable accuracy with state-of-the-art methods while additionally producing meshes free of self-intersections. The resultant meshes are readily usable in physics based simulation, minimizing the need for postprocessing and cleanup.

Multi-network approach for image segmentation in non-contrast enhanced cardiac 3D MRI of arrhythmic patients

Left atrial appendage (LAA) is the source of thrombi formation in more than 90% of strokes in patients with nonvalvular atrial fibrillation. Catheter-based LAA occlusion is being increasingly applied as a treatment strategy to prevent stroke. Anatomical complexity of LAA makes percutaneous occlusion commonly performed under transesophageal echocardiography (TEE) and X-ray (XR) guidance especially challenging. Image fusion techniques integrating 3D anatomical models derived from pre-procedural imaging into the live XR fluoroscopy can be applied to guide each step of the LAA closure. Cardiac magnetic resonance (CMR) imaging gains in importance for radiation-free evaluation of cardiac morphology as alternative to gold-standard TEE or computed tomography angiography (CTA). Manual delineation of cardiac structures from non-contrast enhanced CMR is, however, labor-intensive, tedious, and challenging due to the rather low contrast. Additionally, arrhythmia often impairs the image quality in ECG synchronized acquisitions causing blurring and motion artifacts. Thus, for cardiac segmentation in arrhythmic patients, there is a strong need for an automated image segmentation method. Deep learning-based methods have shown great promise in medical image analysis achieving superior performance in various imaging modalities and different clinical applications. Fully-convolutional neural networks (CNNs), especially U-Net, have become the method of choice for cardiac segmentation. In this paper, we propose an approach for automatic segmentation of cardiac structures from non-contrast enhanced CMR images of arrhythmic patients based on CNNs implemented in a multi-stage pipeline. Two-stage implementation allows subdividing the task into localization of the relevant cardiac structures and segmentation of these structures from the cropped sub-regions obtained from previous step leading to efficient and effective way of automated cardiac segmentation.

TestFit: A plug-and-play one-pass test time method for medical image segmentation

Deep learning (DL) based methods have been extensively studied for medical image segmentation, mostly emphasizing the design and training of DL networks. Only few attempts were made on developing methods for applying DL models in test time. In this paper, we study whether a given off-the-shelf segmentation network can be stably improved on-the-fly during test time in an online processing-and-learning fashion. We propose a new online test-time method, called TestFit, to improve results of a given off-the-shelf DL segmentation model in test time by actively fitting the test data distribution. TestFit first creates a supplementary network (SuppNet) from the given trained off-the-shelf segmentation network (this original network is referred to as OGNet) and applies SuppNet together with OGNet for test time inference. OGNet keeps its hypothesis derived from the original training set to prevent the model from collapsing, while SuppNet seeks to fit the test data distribution. Segmentation results and supervision signals (for updating SuppNet) are generated by combining the outputs of OGNet and SuppNet on the fly. TestFit needs only one pass on each test sample - the same as the traditional test model pipeline - and requires no training time preparation. Since it is challenging to look at only one test sample and no manual annotation for model update each time, we develop a series of technical treatments for improving the stability and effectiveness of our proposed online test-time training method. TestFit works in a plug-and-play fashion, requires minimal hyper-parameter tuning, and is easy to use in practice. Experiments on a large collection of 2D and 3D datasets demonstrate the capability of our TestFit method.

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.

S-Net: a multiple cross aggregation convolutional architecture for automatic segmentation of small/thin structures for cardiovascular applications

With the success of U-Net or its variants in automatic medical image segmentation, building a fully convolutional network (FCN) based on an encoder-decoder structure has become an effective end-to-end learning approach. However, the intrinsic property of FCNs is that as the encoder deepens, higher-level features are learned, and the receptive field size of the network increases, which results in unsatisfactory performance for detecting low-level small/thin structures such as atrial walls and small arteries. To address this issue, we propose to keep the different encoding layer features at their original sizes to constrain the receptive field from increasing as the network goes deeper. Accordingly, we develop a novel S-shaped multiple cross-aggregation segmentation architecture named S-Net, which has two branches in the encoding stage, i.e., a resampling branch to capture low-level fine-grained details and thin/small structures and a downsampling branch to learn high-level discriminative knowledge. In particular, these two branches learn complementary features by residual cross-aggregation; the fusion of the complementary features from different decoding layers can be effectively accomplished through lateral connections. Meanwhile, we perform supervised prediction at all decoding layers to incorporate coarse-level features with high semantic meaning and fine-level features with high localization capability to detect multi-scale structures, especially for small/thin volumes fully. To validate the effectiveness of our S-Net, we conducted extensive experiments on the segmentation of cardiac wall and intracranial aneurysm (IA) vasculature, and quantitative and qualitative evaluations demonstrated the superior performance of our method for predicting small/thin structures in medical images.

MUE-CoT: multi-scale uncertainty entropy-aware co-training framework for left atrial segmentation

Objective.Accurate left atrial segmentation is the basis of the recognition and clinical analysis of atrial fibrillation. Supervised learning has achieved some competitive segmentation results, but the high annotation cost often limits its performance. Semi-supervised learning is implemented from limited labeled data and a large amount of unlabeled data and shows good potential in solving practical medical problems.Approach. In this study, we proposed a collaborative training framework for multi-scale uncertain entropy perception (MUE-CoT) and achieved efficient left atrial segmentation from a small amount of labeled data. Based on the pyramid feature network, learning is implemented from unlabeled data by minimizing the pyramid prediction difference. In addition, novel loss constraints are proposed for co-training in the study. The diversity loss is defined as a soft constraint so as to accelerate the convergence and a novel multi-scale uncertainty entropy calculation method and a consistency regularization term are proposed to measure the consistency between prediction results. The quality of pseudo-labels cannot be guaranteed in the pre-training period, so a confidence-dependent empirical Gaussian function is proposed to weight the pseudo-supervised loss.Main results.The experimental results of a publicly available dataset and an in-house clinical dataset proved that our method outperformed existing semi-supervised methods. For the two datasets with a labeled ratio of 5%, the Dice similarity coefficient scores were 84.94% ± 4.31 and 81.24% ± 2.4, the HD95values were 4.63 mm ± 2.13 and 3.94 mm ± 2.72, and the Jaccard similarity coefficient scores were 74.00% ± 6.20 and 68.49% ± 3.39, respectively.Significance.The proposed model effectively addresses the challenges of limited data samples and high costs associated with manual annotation in the medical field, leading to enhanced segmentation accuracy.

Segmenting computed tomograms for cardiac ablation using machine learning leveraged by domain knowledge encoding

Background

Segmentation of computed tomography (CT) is important for many clinical procedures including personalized cardiac ablation for the management of cardiac arrhythmias. While segmentation can be automated by machine learning (ML), it is limited by the need for large, labeled training data that may be difficult to obtain. We set out to combine ML of cardiac CT with domain knowledge, which reduces the need for large training datasets by encoding cardiac geometry, which we then tested in independent datasets and in a prospective study of atrial fibrillation (AF) ablation.

Methods

We mathematically represented atrial anatomy with simple geometric shapes and derived a model to parse cardiac structures in a small set of N = 6 digital hearts. The model, termed "virtual dissection," was used to train ML to segment cardiac CT in N = 20 patients, then tested in independent datasets and in a prospective study.

Results

In independent test cohorts (N = 160) from 2 Institutions with different CT scanners, atrial structures were accurately segmented with Dice scores of 96.7% in internal (IQR: 95.3%-97.7%) and 93.5% in external (IQR: 91.9%-94.7%) test data, with good agreement with experts (r = 0.99; p < 0.0001). In a prospective study of 42 patients at ablation, this approach reduced segmentation time by 85% (2.3 ± 0.8 vs. 15.0 ± 6.9 min, p < 0.0001), yet provided similar Dice scores to experts (93.9% (IQR: 93.0%-94.6%) vs. 94.4% (IQR: 92.8%-95.7%), p = NS).

Conclusions

Encoding cardiac geometry using mathematical models greatly accelerated training of ML to segment CT, reducing the need for large training sets while retaining accuracy in independent test data. Combining ML with domain knowledge may have broad applications.

GSM-Net: A global sequence modelling network for the segmentation of short axis CINE MRI images

Atrial Fibrillation (AF) is a disease where the atria fail to properly contract but quiver instead, due to the abnormal electrical activity of the atrial tissue. In AF patients, anatomical and functional parameters of the left atrium (LA) largely differ from that of healthy people due to LA remodelling, which can continue in many cases after the catheter ablation treatment. Therefore, it is important to follow up with AF patients to detect any recurrence. LA segmentation masks obtained from short-axis CINE MRI images are used as the gold standard for the quantification of LA parameters. Thick slices of CINE MRI images hinder the use of 3D networks for segmentation while 2D architectures often fail to model inter-slice dependencies. This study presents GSM-Net which approximates 3D networks with effective modelling of inter-slice similarities with two new modules: global slice sequence encoder (GSSE) and sequence dependent channel attention module (SdCAt). In contrast to previous work modelling only local inter-slice similarities, GSSE also models global spatial dependencies across slices. SdCAt generates a distribution of attention weights over MRI slices per channel, to better trace characteristic changes in the size of the LA or other structures across slices. We found that GSM-Net outperforms previous methods on LA segmentation and helps to identify AF recurrence patients. We believe that GSM-Net can be used as an automatic tool to estimate LA parameters such as ejection fraction to identify AF, and to follow up with patients after treatment to detect any recurrence.

Multi-modality cardiac image computing: A survey

Multi-modality cardiac imaging plays a key role in the management of patients with cardiovascular diseases. It allows a combination of complementary anatomical, morphological and functional information, increases diagnosis accuracy, and improves the efficacy of cardiovascular interventions and clinical outcomes. Fully-automated processing and quantitative analysis of multi-modality cardiac images could have a direct impact on clinical research and evidence-based patient management. However, these require overcoming significant challenges including inter-modality misalignment and finding optimal methods to integrate information from different modalities. This paper aims to provide a comprehensive review of multi-modality imaging in cardiology, the computing methods, the validation strategies, the related clinical workflows and future perspectives. For the computing methodologies, we have a favored focus on the three tasks, i.e., registration, fusion and segmentation, which generally involve multi-modality imaging data, either combining information from different modalities or transferring information across modalities. The review highlights that multi-modality cardiac imaging data has the potential of wide applicability in the clinic, such as trans-aortic valve implantation guidance, myocardial viability assessment, and catheter ablation therapy and its patient selection. Nevertheless, many challenges remain unsolved, such as missing modality, modality selection, combination of imaging and non-imaging data, and uniform analysis and representation of different modalities. There is also work to do in defining how the well-developed techniques fit in clinical workflows and how much additional and relevant information they introduce. These problems are likely to continue to be an active field of research and the questions to be answered in the future.

Deep learning-based left ventricular segmentation demonstrates improved performance on respiratory motion-resolved whole-heart reconstructions

Introduction

Deep learning (DL)-based segmentation has gained popularity for routine cardiac magnetic resonance (CMR) image analysis and in particular, delineation of left ventricular (LV) borders for LV volume determination. Free-breathing, self-navigated, whole-heart CMR exams provide high-resolution, isotropic coverage of the heart for assessment of cardiac anatomy including LV volume. The combination of whole-heart free-breathing CMR and DL-based LV segmentation has the potential to streamline the acquisition and analysis of clinical CMR exams. The purpose of this study was to compare the performance of a DL-based automatic LV segmentation network trained primarily on computed tomography (CT) images in two whole-heart CMR reconstruction methods: (1) an in-line respiratory motion-corrected (Mcorr) reconstruction and (2) an off-line, compressed sensing-based, multi-volume respiratory motion-resolved (Mres) reconstruction. Given that Mres images were shown to have greater image quality in previous studies than Mcorr images, we hypothesized that the LV volumes segmented from Mres images are closer to the manual expert-traced left ventricular endocardial border than the Mcorr images.

Method

This retrospective study used 15 patients who underwent clinically indicated 1.5 T CMR exams with a prototype ECG-gated 3D radial phyllotaxis balanced steady state free precession (bSSFP) sequence. For each reconstruction method, the absolute volume difference (AVD) of the automatically and manually segmented LV volumes was used as the primary quantity to investigate whether 3D DL-based LV segmentation generalized better on Mcorr or Mres 3D whole-heart images. Additionally, we assessed the 3D Dice similarity coefficient between the manual and automatic LV masks of each reconstructed 3D whole-heart image and the sharpness of the LV myocardium-blood pool interface. A two-tail paired Student's t-test (alpha = 0.05) was used to test the significance in this study.

Results & Discussion

The AVD in the respiratory Mres reconstruction was lower than the AVD in the respiratory Mcorr reconstruction: 7.73 ± 6.54 ml vs. 20.0 ± 22.4 ml, respectively (n = 15, p-value = 0.03). The 3D Dice coefficient between the DL-segmented masks and the manually segmented masks was higher for Mres images than for Mcorr images: 0.90 ± 0.02 vs. 0.87 ± 0.03 respectively, with a p-value = 0.02. Sharpness on Mres images was higher than on Mcorr images: 0.15 ± 0.05 vs. 0.12 ± 0.04, respectively, with a p-value of 0.014 (n = 15).

Conclusion

We conclude that the DL-based 3D automatic LV segmentation network trained on CT images and fine-tuned on MR images generalized better on Mres images than on Mcorr images for quantifying LV volumes.

SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining

Despite advances in data augmentation and transfer learning, convolutional neural networks (CNNs) difficultly generalise to unseen domains. When segmenting brain scans, CNNs are highly sensitive to changes in resolution and contrast: even within the same MRI modality, performance can decrease across datasets. Here we introduce SynthSeg, the first segmentation CNN robust against changes in contrast and resolution. SynthSeg is trained with synthetic data sampled from a generative model conditioned on segmentations. Crucially, we adopt a domain randomisation strategy where we fully randomise the contrast and resolution of the synthetic training data. Consequently, SynthSeg can segment real scans from a wide range of target domains without retraining or fine-tuning, which enables straightforward analysis of huge amounts of heterogeneous clinical data. Because SynthSeg only requires segmentations to be trained (no images), it can learn from labels obtained by automated methods on diverse populations (e.g., ageing and diseased), thus achieving robustness to a wide range of morphological variability. We demonstrate SynthSeg on 5,000 scans of six modalities (including CT) and ten resolutions, where it exhibits unparallelled generalisation compared with supervised CNNs, state-of-the-art domain adaptation, and Bayesian segmentation. Finally, we demonstrate the generalisability of SynthSeg by applying it to cardiac MRI and CT scans.

Energy-efficient high-fidelity image reconstruction with memristor arrays for medical diagnosis

Medical imaging is an important tool for accurate medical diagnosis, while state-of-the-art image reconstruction algorithms raise critical challenges in massive data processing for high-speed and high-quality imaging. Here, we present a memristive image reconstructor (MIR) to greatly accelerate image reconstruction with discrete Fourier transformation (DFT) by computing-in-memory (CIM) with memristor arrays. A high-accuracy quasi-analogue mapping (QAM) method and generic complex matrix transfer (CMT) scheme was proposed to improve the mapping precision and transfer efficiency, respectively. High-fidelity magnetic resonance imaging (MRI) and computed tomography (CT) image reconstructions were demonstrated, achieving software-equivalent qualities and DICE scores after segmentation with nnU-Net algorithm. Remarkably, our MIR exhibited 153× and 79× improvements in energy efficiency and normalized image reconstruction speed, respectively, compared to graphics processing unit (GPU). This work demonstrates MIR as a promising high-fidelity image reconstruction platform for future medical diagnosis, and also largely extends the application of memristor-based CIM beyond artificial neural networks.

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/).

aiWave: Volumetric Image Compression With 3-D Trained Affine Wavelet-Like Transform

Volumetric image compression has become an urgent task to effectively transmit and store images produced in biological research and clinical practice. At present, the most commonly used volumetric image compression methods are based on wavelet transform, such as JP3D. However, JP3D employs an ideal, separable, global, and fixed wavelet basis to convert input images from pixel domain to frequency domain, which seriously limits its performance. In this paper, we first design a 3-D trained wavelet-like transform to enable signal-dependent and non-separable transform. Then, an affine wavelet basis is introduced to capture the various local correlations in different regions of volumetric images. Furthermore, we embed the proposed wavelet-like transform to an end-to-end compression framework called aiWave to enable an adaptive compression scheme for various datasets. Last but not least, we introduce the weight sharing strategies of the affine wavelet-like transform according to the volumetric data characteristics in the axial direction to reduce the number of parameters. The experimental results show that: 1) when cooperating our trained 3-D affine wavelet-like transform with a simple factorized entropy coding module, aiWave performs better than JP3D and is comparable in terms of encoding and decoding complexities; 2) when adding a context module to remove signal redundancy further, aiWave can achieve a much better performance than HEVC.

Beyond CFD: Emerging methodologies for predictive simulation in cardiovascular health and disease

Physics-based computational models of the cardiovascular system are increasingly used to simulate hemodynamics, tissue mechanics, and physiology in evolving healthy and diseased states. While predictive models using computational fluid dynamics (CFD) originated primarily for use in surgical planning, their application now extends well beyond this purpose. In this review, we describe an increasingly wide range of modeling applications aimed at uncovering fundamental mechanisms of disease progression and development, performing model-guided design, and generating testable hypotheses to drive targeted experiments. Increasingly, models are incorporating multiple physical processes spanning a wide range of time and length scales in the heart and vasculature. With these expanded capabilities, clinical adoption of patient-specific modeling in congenital and acquired cardiovascular disease is also increasing, impacting clinical care and treatment decisions in complex congenital heart disease, coronary artery disease, vascular surgery, pulmonary artery disease, and medical device design. In support of these efforts, we discuss recent advances in modeling methodology, which are most impactful when driven by clinical needs. We describe pivotal recent developments in image processing, fluid-structure interaction, modeling under uncertainty, and reduced order modeling to enable simulations in clinically relevant timeframes. In all these areas, we argue that traditional CFD alone is insufficient to tackle increasingly complex clinical and biological problems across scales and systems. Rather, CFD should be coupled with appropriate multiscale biological, physical, and physiological models needed to produce comprehensive, impactful models of mechanobiological systems and complex clinical scenarios. With this perspective, we finally outline open problems and future challenges in the field.

Learning Whole Heart Mesh Generation From Patient Images for Computational Simulations

Patient-specific cardiac modeling combines geometries of the heart derived from medical images and biophysical simulations to predict various aspects of cardiac function. However, generating simulation-suitable models of the heart from patient image data often requires complicated procedures and significant human effort. We present a fast and automated deep-learning method to construct simulation-suitable models of the heart from medical images. The approach constructs meshes from 3D patient images by learning to deform a small set of deformation handles on a whole heart template. For both 3D CT and MR data, this method achieves promising accuracy for whole heart reconstruction, consistently outperforming prior methods in constructing simulation-suitable meshes of the heart. When evaluated on time-series CT data, this method produced more anatomically and temporally consistent geometries than prior methods, and was able to produce geometries that better satisfy modeling requirements for cardiac flow simulations. Our source code and pretrained networks are available at https://github.com/fkong7/HeartDeformNets.

Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning

Accurate segmentation of the left atrial structure using magnetic resonance images provides an important basis for the diagnosis of atrial fibrillation (AF) and its treatment using robotic surgery. In this study, an image segmentation method based on sequence relationship learning and multi-scale feature fusion is proposed for 3D to 2D sequence conversion in cardiac magnetic resonance images and the varying scales of left atrial structures within different slices. Firstly, a convolutional neural network layer with an attention module was designed to extract and fuse contextual information at different scales in the image, to strengthen the target features using the correlation between features in different regions within the image, and to improve the network's ability to distinguish the left atrial structure. Secondly, a recurrent neural network layer oriented to two-dimensional images was designed to capture the correlation of left atrial structures in adjacent slices by simulating the continuous relationship between sequential image slices. Finally, a combined loss function was constructed to reduce the effect of positive and negative sample imbalance and improve model stability. The Dice, IoU, and Hausdorff distance values reached 90.73%, 89.37%, and 4.803 mm, respectively, based on the LASC2013 (left atrial segmentation challenge in 2013) dataset; the corresponding values reached 92.05%, 89.41% and 9.056 mm, respectively, based on the ASC2018 (atrial segmentation challenge at 2018) dataset.

STAMP: Simultaneous Training and Model Pruning for low data regimes in medical image segmentation

Acquisition of high quality manual annotations is vital for the development of segmentation algorithms. However, to create them we require a substantial amount of expert time and knowledge. Large numbers of labels are required to train convolutional neural networks due to the vast number of parameters that must be learned in the optimisation process. Here, we develop the STAMP algorithm to allow the simultaneous training and pruning of a UNet architecture for medical image segmentation with targeted channelwise dropout to make the network robust to the pruning. We demonstrate the technique across segmentation tasks and imaging modalities. It is then shown that, through online pruning, we are able to train networks to have much higher performance than the equivalent standard UNet models while reducing their size by more than 85% in terms of parameters. This has the potential to allow networks to be directly trained on datasets where very low numbers of labels are available.

The Medical Segmentation Decathlon

International challenges have become the de facto standard for comparative assessment of image analysis algorithms. Although segmentation is the most widely investigated medical image processing task, the various challenges have been organized to focus only on specific clinical tasks. We organized the Medical Segmentation Decathlon (MSD)-a biomedical image analysis challenge, in which algorithms compete in a multitude of both tasks and modalities to investigate the hypothesis that a method capable of performing well on multiple tasks will generalize well to a previously unseen task and potentially outperform a custom-designed solution. MSD results confirmed this hypothesis, moreover, MSD winner continued generalizing well to a wide range of other clinical problems for the next two years. Three main conclusions can be drawn from this study: (1) state-of-the-art image segmentation algorithms generalize well when retrained on unseen tasks; (2) consistent algorithmic performance across multiple tasks is a strong surrogate of algorithmic generalizability; (3) the training of accurate AI segmentation models is now commoditized to scientists that are not versed in AI model training.

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/).

Medical image analysis on left atrial LGE MRI for atrial fibrillation studies: A review

Late gadolinium enhancement magnetic resonance imaging (LGE MRI) is commonly used to visualize and quantify left atrial (LA) scars. The position and extent of LA scars provide important information on the pathophysiology and progression of atrial fibrillation (AF). Hence, LA LGE MRI computing and analysis are essential for computer-assisted diagnosis and treatment stratification of AF patients. Since manual delineations can be time-consuming and subject to intra- and inter-expert variability, automating this computing is highly desired, which nevertheless is still challenging and under-researched. This paper aims to provide a systematic review on computing methods for LA cavity, wall, scar, and ablation gap segmentation and quantification from LGE MRI, and the related literature for AF studies. Specifically, we first summarize AF-related imaging techniques, particularly LGE MRI. Then, we review the methodologies of the four computing tasks in detail and summarize the validation strategies applied in each task as well as state-of-the-art results on public datasets. Finally, the possible future developments are outlined, with a brief survey on the potential clinical applications of the aforementioned methods. The review indicates that the research into this topic is still in the early stages. Although several methods have been proposed, especially for the LA cavity segmentation, there is still a large scope for further algorithmic developments due to performance issues related to the high variability of enhancement appearance and differences in image acquisition.

Validation of a Whole Heart Segmentation from Computed Tomography Imaging Using a Deep-Learning Approach

The aim of this study is to develop an automated deep-learning-based whole heart segmentation of ECG-gated computed tomography data. After 21 exclusions, CT acquired before transcatheter aortic valve implantation in 71 patients were reviewed and randomly split in a training (n = 55 patients), validation (n = 8 patients), and a test set (n = 8 patients). A fully automatic deep-learning method combining two convolutional neural networks performed segmentation of 10 cardiovascular structures, which was compared with the manually segmented reference by the Dice index. Correlations and agreement between myocardial volumes and mass were assessed. The algorithm demonstrated high accuracy (Dice score = 0.920; interquartile range: 0.906-0.925) and a low computing time (13.4 s, range 11.9-14.9). Correlations and agreement of volumes and mass were satisfactory for most structures. Six of ten structures were well segmented. Deep-learning-based method allowed automated WHS from ECG-gated CT data with a high accuracy. Challenges remain to improve right-sided structures segmentation and achieve daily clinical application.

A Two-Stage Model for Predicting Mild Cognitive Impairment to Alzheimer's Disease Conversion

Early detection of Alzheimer's disease (AD), such as predicting development from mild cognitive impairment (MCI) to AD, is critical for slowing disease progression and increasing quality of life. Although deep learning is a promising technique for structural MRI-based diagnosis, the paucity of training samples limits its power, especially for three-dimensional (3D) models. To this end, we propose a two-stage model combining both transfer learning and contrastive learning that can achieve high accuracy of MRI-based early AD diagnosis even when the sample numbers are restricted. Specifically, a 3D CNN model was pretrained using publicly available medical image data to learn common medical features, and contrastive learning was further utilized to learn more specific features of MCI images. The two-stage model outperformed each benchmark method. Compared with the previous studies, we show that our model achieves superior performance in progressive MCI patients with an accuracy of 0.82 and AUC of 0.84. We further enhance the interpretability of the model by using 3D Grad-CAM, which highlights brain regions with high-predictive weights. Brain regions, including the hippocampus, temporal, and precuneus, are associated with the classification of MCI, which is supported by the various types of literature. Our model provides a novel model to avoid overfitting because of a lack of medical data and enable the early detection of AD.

LA-Net: A Multi-Task Deep Network for the Segmentation of the Left Atrium

Although atrial fibrillation (AF) is the most common sustained atrial arrhythmia, treatment success for this condition remains suboptimal. Information from magnetic resonance imaging (MRI) has the potential to improve treatment efficacy, but there are currently few automatic tools for the segmentation of the atria in MR images. In the study, we propose a LA-Net, a multi-task network optimised to simultaneously generate left atrial segmentation and edge masks from MRI. LA-Net includes cross attention modules (CAMs) and enhanced decoder modules (EDMs) to purposefully select the most meaningful edge information for segmentation and smoothly incorporate it into segmentation masks at multiple-scales. We evaluate the performance of LA-Net on two MR sequences: late gadolinium enhanced (LGE) atrial MRI and atrial short axis balanced steady state free precession (bSSFP) MRI. LA-Net gives Hausdorff distances of 12.43 mm and Dice scores of 0.92 on the LGE (STACOM 2018) dataset and Hausdorff distances of 17.41 mm and Dice scores of 0.90 on the bSSFP (in-house) dataset without any post-processing, surpassing previously proposed segmentation networks, including U-Net and SEGANet. Our method allows automatic extraction of information about the LA from MR images, which can play an important role in the management of AF patients.

AtrialJSQnet: A New framework for joint segmentation and quantification of left atrium and scars incorporating spatial and shape information

Left atrial (LA) and atrial scar segmentation from late gadolinium enhanced magnetic resonance imaging (LGE MRI) is an important task in clinical practice. The automatic segmentation is however still challenging due to the poor image quality, the various LA shapes, the thin wall, and the surrounding enhanced regions. Previous methods normally solved the two tasks independently and ignored the intrinsic spatial relationship between LA and scars. In this work, we develop a new framework, namely AtrialJSQnet, where LA segmentation, scar projection onto the LA surface, and scar quantification are performed simultaneously in an end-to-end style. We propose a mechanism of shape attention (SA) via an implicit surface projection to utilize the inherent correlation between LA cavity and scars. In specific, the SA scheme is embedded into a multi-task architecture to perform joint LA segmentation and scar quantification. Besides, a spatial encoding (SE) loss is introduced to incorporate continuous spatial information of the target in order to reduce noisy patches in the predicted segmentation. We evaluated the proposed framework on 60 post-ablation LGE MRIs from the MICCAI2018 Atrial Segmentation Challenge. Moreover, we explored the domain generalization ability of the proposed AtrialJSQnet on 40 pre-ablation LGE MRIs from this challenge and 30 post-ablation multi-center LGE MRIs from another challenge (ISBI2012 Left Atrium Fibrosis and Scar Segmentation Challenge). Extensive experiments on public datasets demonstrated the effect of the proposed AtrialJSQnet, which achieved competitive performance over the state-of-the-art. The relatedness between LA segmentation and scar quantification was explicitly explored and has shown significant performance improvements for both tasks. The code has been released via https://zmiclab.github.io/projects.html.

A Deep Learning-Based and Fully Automated Pipeline for Thoracic Aorta Geometric Analysis and Planning for Endovascular Repair from Computed Tomography

Feasibility assessment and planning of thoracic endovascular aortic repair (TEVAR) require computed tomography (CT)-based analysis of geometric aortic features to identify adequate landing zones (LZs) for endograft deployment. However, no consensus exists on how to take the necessary measurements from CT image data. We trained and applied a fully automated pipeline embedding a convolutional neural network (CNN), which feeds on 3D CT images to automatically segment the thoracic aorta, detects proximal landing zones (PLZs), and quantifies geometric features that are relevant for TEVAR planning. For 465 CT scans, the thoracic aorta and pulmonary arteries were manually segmented; 395 randomly selected scans with the corresponding ground truth segmentations were used to train a CNN with a 3D U-Net architecture. The remaining 70 scans were used for testing. The trained CNN was embedded within computational geometry processing pipeline which provides aortic metrics of interest for TEVAR planning. The resulting metrics included aortic arch centerline radius of curvature, proximal landing zones (PLZs) maximum diameters, angulation, and tortuosity. These parameters were statistically analyzed to compare standard arches vs. arches with a common origin of the innominate and left carotid artery (CILCA). The trained CNN yielded a mean Dice score of 0.95 and was able to generalize to 9 pathological cases of thoracic aortic aneurysm, providing accurate segmentations. CILCA arches were characterized by significantly greater angulation (p = 0.015) and tortuosity (p = 0.048) in PLZ 3 vs. standard arches. For both arch configurations, comparisons among PLZs revealed statistically significant differences in maximum zone diameters (p < 0.0001), angulation (p < 0.0001), and tortuosity (p < 0.0001). Our tool allows clinicians to obtain objective and repeatable PLZs mapping, and a range of automatically derived complex aortic metrics.

JAS-GAN: Generative Adversarial Network Based Joint Atrium and Scar Segmentations on Unbalanced Atrial Targets

Automated and accurate segmentations of left atrium (LA) and atrial scars from late gadolinium-enhanced cardiac magnetic resonance (LGE CMR) images are in high demand for quantifying atrial scars. The previous quantification of atrial scars relies on a two-phase segmentation for LA and atrial scars due to their large volume difference (unbalanced atrial targets). In this paper, we propose an inter-cascade generative adversarial network, namely JAS-GAN, to segment the unbalanced atrial targets from LGE CMR images automatically and accurately in an end-to-end way. Firstly, JAS-GAN investigates an adaptive attention cascade to automatically correlate the segmentation tasks of the unbalanced atrial targets. The adaptive attention cascade mainly models the inclusion relationship of the two unbalanced atrial targets, where the estimated LA acts as the attention map to adaptively focus on the small atrial scars roughly. Then, an adversarial regularization is applied to the segmentation tasks of the unbalanced atrial targets for making a consistent optimization. It mainly forces the estimated joint distribution of LA and atrial scars to match the real ones. We evaluated the performance of our JAS-GAN on a 3D LGE CMR dataset with 192 scans. Compared with the state-of-the-art methods, our proposed approach yielded better segmentation performance (Average Dice Similarity Coefficient (DSC) values of 0.946 and 0.821 for LA and atrial scars, respectively), which indicated the effectiveness of our proposed approach for segmenting unbalanced atrial targets.

Left atrial evaluation by cardiovascular magnetic resonance: sensitive and unique biomarkers

Left atrial (LA) imaging is still not routinely used for diagnosis and risk stratification, although recent studies have emphasized its importance as an imaging biomarker. Cardiovascular magnetic resonance is able to evaluate LA structure and function, metrics that serve as early indicators of disease, and provide prognostic information, e.g. regarding diastolic dysfunction, and atrial fibrillation (AF). MR angiography defines atrial anatomy, useful for planning ablation procedures, and also for characterizing atrial shapes and sizes that might predict cardiovascular events, e.g. stroke. Long-axis cine images can be evaluated to define minimum, maximum, and pre-atrial contraction LA volumes, and ejection fractions (EFs). More modern feature tracking of these cine images provides longitudinal LA strain through the cardiac cycle, and strain rates. Strain may be a more sensitive marker than EF and can predict post-operative AF, AF recurrence after ablation, outcomes in hypertrophic cardiomyopathy, stratification of diastolic dysfunction, and strain correlates with atrial fibrosis. Using high-resolution late gadolinium enhancement (LGE), the extent of fibrosis in the LA can be estimated and post-ablation scar can be evaluated. The LA LGE method is widely available, its reproducibility is good, and validations with voltage-mapping exist, although further scan-rescan studies are needed, and consensus regarding atrial segmentation is lacking. Using LGE, scar patterns after ablation in AF subjects can be reproducibly defined. Evaluation of 'pre-existent' atrial fibrosis may have roles in predicting AF recurrence after ablation, predicting new-onset AF and diastolic dysfunction in patients without AF. LA imaging biomarkers are ready to enter into diagnostic clinical practice.

A deep-learning approach for direct whole-heart mesh reconstruction

Automated construction of surface geometries of cardiac structures from volumetric medical images is important for a number of clinical applications. While deep-learning-based approaches have demonstrated promising reconstruction precision, these approaches have mostly focused on voxel-wise segmentation followed by surface reconstruction and post-processing techniques. However, such approaches suffer from a number of limitations including disconnected regions or incorrect surface topology due to erroneous segmentation and stair-case artifacts due to limited segmentation resolution. We propose a novel deep-learning-based approach that directly predicts whole heart surface meshes from volumetric CT and MR image data. Our approach leverages a graph convolutional neural network to predict deformation on mesh vertices from a pre-defined mesh template to reconstruct multiple anatomical structures in a 3D image volume. Our method demonstrated promising performance of generating whole heart reconstructions with as good or better accuracy than prior deep-learning-based methods on both CT and MR data. Furthermore, by deforming a template mesh, our method can generate whole heart geometries with better anatomical consistency and produce high-resolution geometries from lower resolution input image data. Our method was also able to produce temporally-consistent surface mesh predictions for heart motion from CT or MR cine sequences, and therefore can potentially be applied for efficiently constructing 4D whole heart dynamics. Our code and pre-trained networks are available at https://github.com/fkong7/MeshDeformNet.

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.

A Survey of Brain Tumor Segmentation and Classification Algorithms

A brain Magnetic resonance imaging (MRI) scan of a single individual consists of several slices across the 3D anatomical view. Therefore, manual segmentation of brain tumors from magnetic resonance (MR) images is a challenging and time-consuming task. In addition, an automated brain tumor classification from an MRI scan is non-invasive so that it avoids biopsy and make the diagnosis process safer. Since the beginning of this millennia and late nineties, the effort of the research community to come-up with automatic brain tumor segmentation and classification method has been tremendous. As a result, there are ample literature on the area focusing on segmentation using region growing, traditional machine learning and deep learning methods. Similarly, a number of tasks have been performed in the area of brain tumor classification into their respective histological type, and an impressive performance results have been obtained. Considering state of-the-art methods and their performance, the purpose of this paper is to provide a comprehensive survey of three, recently proposed, major brain tumor segmentation and classification model techniques, namely, region growing, shallow machine learning and deep learning. The established works included in this survey also covers technical aspects such as the strengths and weaknesses of different approaches, pre- and post-processing techniques, feature extraction, datasets, and models' performance evaluation metrics.

A bi-atrial statistical shape model for large-scale in silico studies of human atria: Model development and application to ECG simulations

Large-scale electrophysiological simulations to obtain electrocardiograms (ECG) carry the potential to produce extensive datasets for training of machine learning classifiers to, e.g., discriminate between different cardiac pathologies. The adoption of simulations for these purposes is limited due to a lack of ready-to-use models covering atrial anatomical variability. We built a bi-atrial statistical shape model (SSM) of the endocardial wall based on 47 segmented human CT and MRI datasets using Gaussian process morphable models. Generalization, specificity, and compactness metrics were evaluated. The SSM was applied to simulate atrial ECGs in 100 random volumetric instances. The first eigenmode of our SSM reflects a change of the total volume of both atria, the second the asymmetry between left vs. right atrial volume, the third a change in the prominence of the atrial appendages. The SSM is capable of generalizing well to unseen geometries and 95% of the total shape variance is covered by its first 24 eigenvectors. The P waves in the 12-lead ECG of 100 random instances showed a duration of 109.7±12.2 ms in accordance with large cohort studies. The novel bi-atrial SSM itself as well as 100 exemplary instances with rule-based augmentation of atrial wall thickness, fiber orientation, inter-atrial bridges and tags for anatomical structures have been made publicly available. This novel, openly available bi-atrial SSM can in future be employed to generate large sets of realistic atrial geometries as a basis for in silico big data approaches.

Supervised Domain Adaptation for Automated Semantic Segmentation of the Atrial Cavity

Atrial fibrillation (AF) is the most common cardiac arrhythmia. At present, cardiac ablation is the main treatment procedure for AF. To guide and plan this procedure, it is essential for clinicians to obtain patient-specific 3D geometrical models of the atria. For this, there is an interest in automatic image segmentation algorithms, such as deep learning (DL) methods, as opposed to manual segmentation, an error-prone and time-consuming method. However, to optimize DL algorithms, many annotated examples are required, increasing acquisition costs. The aim of this work is to develop automatic and high-performance computational models for left and right atrium (LA and RA) segmentation from a few labelled MRI volumetric images with a 3D Dual U-Net algorithm. For this, a supervised domain adaptation (SDA) method is introduced to infer knowledge from late gadolinium enhanced (LGE) MRI volumetric training samples (80 LA annotated samples) to a network trained with balanced steady-state free precession (bSSFP) MR images of limited number of annotations (19 RA and LA annotated samples). The resulting knowledge-transferred model SDA outperformed the same network trained from scratch in both RA (Dice equals 0.9160) and LA (Dice equals 0.8813) segmentation tasks.

Automated left atrial time-resolved segmentation in MRI long-axis cine images using active contours

BACKGROUND

Segmentation of the left atrium (LA) is required to evaluate atrial size and function, which are important imaging biomarkers for a wide range of cardiovascular conditions, such as atrial fibrillation, stroke, and diastolic dysfunction. LA segmentations are currently being performed manually, which is time-consuming and observer-dependent.

METHODS

This study presents an automated image processing algorithm for time-resolved LA segmentation in cardiac magnetic resonance imaging (MRI) long-axis cine images of the 2-chamber (2ch) and 4-chamber (4ch) views using active contours. The proposed algorithm combines mitral valve tracking, automated threshold calculation, edge detection on a radially resampled image, edge tracking based on Dijkstra's algorithm, and post-processing involving smoothing and interpolation. The algorithm was evaluated in 37 patients diagnosed mainly with paroxysmal atrial fibrillation. Segmentation accuracy was assessed using the Dice similarity coefficient (DSC) and Hausdorff distance (HD), with manual segmentations in all time frames as the reference standard. For inter-observer variability analysis, a second observer performed manual segmentations at end-diastole and end-systole on all subjects.

RESULTS

The proposed automated method achieved high performance in segmenting the LA in long-axis cine sequences, with a DSC of 0.96 for 2ch and 0.95 for 4ch, and an HD of 5.5 mm for 2ch and 6.4 mm for 4ch. The manual inter-observer variability analysis had an average DSC of 0.95 and an average HD of 4.9 mm.

CONCLUSION

The proposed automated method achieved performance on par with human experts analyzing MRI images for evaluation of atrial size and function. Video Abstract.

Optimisation of Left Atrial Feature Tracking Using Retrospective Gated Computed Tomography Images

Retrospective gated cardiac computed tomography (CCT) images can provide high contrast and resolution images of the heart throughout the cardiac cycle. Feature tracking in retrospective CCT images using the temporal sparse free-form deformations (TSFFDs) registration method has previously been optimised for the left ventricle (LV). However, there is limited work on optimising nonrigid registration methods for feature tracking in the left atria (LA). This paper systematically optimises the sparsity weight (SW) and bending energy (BE) as two hyperparameters of the TSFFD method to track the LA endocardium from end-diastole (ED) to end-systole (ES) using 10-frame retrospective gated CCT images. The effect of two different control point (CP) grid resolutions was also investigated. TSFFD optimisation was achieved using the average surface distance (ASD), directed Hausdorff distance (DHD) and Dice score between the registered and ground truth surface meshes and segmentations at ES. For baseline comparison, the configuration optimised for LV feature tracking gave errors across the cohort of 0.826 ± 0.172mm ASD, 5.882 ± 1.524mm DHD, and 0.912 ± 0.033 Dice score. Optimising the SW and BE hyperparameters improved the TSFFD performance in tracking LA features, with case specific optimisations giving errors across the cohort of 0.750 ± 0.144mm ASD, 5.096 ± 1.246mm DHD, and 0.919 ± 0.029 Dice score. Increasing the CP resolution and optimising the SW and BE further improved tracking performance, with case specific optimisation errors of 0.372 ± 0.051mm ASD, 2.739 ± 0.843mm DHD and 0.949 ± 0.018 Dice score across the cohort. We therefore show LA feature tracking using TSFFDs is improved through a chamber-specific optimised configuration.

Electrocardiographic Imaging for Atrial Fibrillation: A Perspective From Computer Models and Animal Experiments to Clinical Value

Electrocardiographic imaging (ECGI) is a technique to reconstruct non-invasively the electrical activity on the heart surface from body-surface potential recordings and geometric information of the torso and the heart. ECGI has shown scientific and clinical value when used to characterize and treat both atrial and ventricular arrhythmias. Regarding atrial fibrillation (AF), the characterization of the electrical propagation and the underlying substrate favoring AF is inherently more challenging than for ventricular arrhythmias, due to the progressive and heterogeneous nature of the disease and its manifestation, the small volume and wall thickness of the atria, and the relatively large role of microstructural abnormalities in AF. At the same time, ECGI has the advantage over other mapping technologies of allowing a global characterization of atrial electrical activity at every atrial beat and non-invasively. However, since ECGI is time-consuming and costly and the use of electrical mapping to guide AF ablation is still not fully established, the clinical value of ECGI for AF is still under assessment. Nonetheless, AF is known to be the manifestation of a complex interaction between electrical and structural abnormalities and therefore, true electro-anatomical-structural imaging may elucidate important key factors of AF development, progression, and treatment. Therefore, it is paramount to identify which clinical questions could be successfully addressed by ECGI when it comes to AF characterization and treatment, and which questions may be beyond its technical limitations. In this manuscript we review the questions that researchers have tried to address on the use of ECGI for AF characterization and treatment guidance (for example, localization of AF triggers and sustaining mechanisms), and we discuss the technological requirements and validation. We address experimental and clinical results, limitations, and future challenges for fruitful application of ECGI for AF understanding and management. We pay attention to existing techniques and clinical application, to computer models and (animal or human) experiments, to challenges of methodological and clinical validation. The overall objective of the study is to provide a consensus on valuable directions that ECGI research may take to provide future improvements in AF characterization and treatment guidance.

Harnessing Machine Intelligence in Automatic Echocardiogram Analysis: Current Status, Limitations, and Future Directions

Echocardiography (echo) is a critical tool in diagnosing various cardiovascular diseases. Despite its diagnostic and prognostic value, interpretation and analysis of echo images are still widely performed manually by echocardiographers. A plethora of algorithms has been proposed to analyze medical ultrasound data using signal processing and machine learning techniques. These algorithms provided opportunities for developing automated echo analysis and interpretation systems. The automated approach can significantly assist in decreasing the variability and burden associated with manual image measurements. In this paper, we review the state-of-the-art automatic methods for analyzing echocardiography data. Particularly, we comprehensively and systematically review existing methods of four major tasks: echo quality assessment, view classification, boundary segmentation, and disease diagnosis. Our review covers three echo imaging modes, which are B-mode, M-mode, and Doppler. We also discuss the challenges and limitations of current methods and outline the most pressing directions for future research. In summary, this review presents the current status of automatic echo analysis and discusses the challenges that need to be addressed to obtain robust systems suitable for efficient use in clinical settings or point-of-care testing.

Bayesian Estimation of Geometric Morphometric Landmarks for Simultaneous Localization of Multiple Anatomies in Cardiac CT Images

We propose a robust method to simultaneously localize multiple objects in cardiac computed tomography angiography (CTA) images. The relative prior distributions of the multiple objects in the three-dimensional (3D) space can be obtained through integrating the geometric morphological relationship of each target object to some reference objects. In cardiac CTA images, the cross-sections of ascending and descending aorta can play the role of the reference objects. We employed the maximum a posteriori (MAP) estimator that utilizes anatomic prior knowledge to address this problem of localizing multiple objects. We propose a new feature for each pixel using the relative distances, which can define any objects that have unclear boundaries. Our experimental results targeting four pulmonary veins (PVs) and the left atrial appendage (LAA) in cardiac CTA images demonstrate the robustness of the proposed method. The method could also be extended to localize other multiple objects in different applications.

A global benchmark of algorithms for segmenting the left atrium from late gadolinium-enhanced cardiac magnetic resonance imaging

Segmentation of medical images, particularly late gadolinium-enhanced magnetic resonance imaging (LGE-MRI) used for visualizing diseased atrial structures, is a crucial first step for ablation treatment of atrial fibrillation. However, direct segmentation of LGE-MRIs is challenging due to the varying intensities caused by contrast agents. Since most clinical studies have relied on manual, labor-intensive approaches, automatic methods are of high interest, particularly optimized machine learning approaches. To address this, we organized the 2018 Left Atrium Segmentation Challenge using 154 3D LGE-MRIs, currently the world's largest atrial LGE-MRI dataset, and associated labels of the left atrium segmented by three medical experts, ultimately attracting the participation of 27 international teams. In this paper, extensive analysis of the submitted algorithms using technical and biological metrics was performed by undergoing subgroup analysis and conducting hyper-parameter analysis, offering an overall picture of the major design choices of convolutional neural networks (CNNs) and practical considerations for achieving state-of-the-art left atrium segmentation. Results show that the top method achieved a Dice score of 93.2% and a mean surface to surface distance of 0.7 mm, significantly outperforming prior state-of-the-art. Particularly, our analysis demonstrated that double sequentially used CNNs, in which a first CNN is used for automatic region-of-interest localization and a subsequent CNN is used for refined regional segmentation, achieved superior results than traditional methods and machine learning approaches containing single CNNs. This large-scale benchmarking study makes a significant step towards much-improved segmentation methods for atrial LGE-MRIs, and will serve as an important benchmark for evaluating and comparing the future works in the field. Furthermore, the findings from this study can potentially be extended to other imaging datasets and modalities, having an impact on the wider medical imaging community.

Fully Automatic Atrial Fibrosis Assessment Using a Multilabel Convolutional Neural Network

Supplemental Digital Content is available in the text.

Background:

Pathological atrial fibrosis is a major contributor to sustained atrial fibrillation. Currently, late gadolinium enhancement (LGE) scans provide the only noninvasive estimate of atrial fibrosis. However, widespread adoption of atrial LGE has been hindered partly by nonstandardized image processing techniques, which can be operator and algorithm dependent. Minimal validation and limited access to transparent software platforms have also exacerbated the problem. This study aims to estimate atrial fibrosis from cardiac magnetic resonance scans using a reproducible operator-independent fully automatic open-source end-to-end pipeline.

Methods:

A multilabel convolutional neural network was designed to accurately delineate atrial structures including the blood pool, pulmonary veins, and mitral valve. The output from the network removed the operator dependent steps in a reproducible pipeline and allowed for automated estimation of atrial fibrosis from LGE-cardiac magnetic resonance scans. The pipeline results were compared against manual fibrosis burdens, calculated using published thresholds: image intensity ratio 0.97, image intensity ratio 1.61, and mean blood pool signal +3.3 SD.

Results:

We validated our methods on a large 3-dimensional LGE-cardiac magnetic resonance data set from 207 labeled scans. Automatic atrial segmentation achieved a 91% Dice score, compared with the mutual agreement of 85% in Dice seen in the interobserver analysis of operators. Intraclass correlation coefficients of the automatic pipeline with manually generated results were excellent and better than or equal to interobserver correlations for all 3 thresholds: 0.94 versus 0.88, 0.99 versus 0.99, 0.99 versus 0.96 for image intensity ratio 0.97, image intensity ratio 1.61, and +3.3 SD thresholds, respectively. Automatic analysis required 3 minutes per case on a standard workstation. The network and the analysis software are publicly available.

Conclusions:

Our pipeline provides a fully automatic estimation of fibrosis burden from LGE-cardiac magnetic resonance scans that is comparable to manual analysis. This removes one key source of variability in the measurement of atrial fibrosis.

Evaluation of multislice inputs to convolutional neural networks for medical image segmentation

PURPOSE

When using convolutional neural networks (CNNs) for segmentation of organs and lesions in medical images, the conventional approach is to work with inputs and outputs either as single slice [two-dimensional (2D)] or whole volumes [three-dimensional (3D)]. One common alternative, in this study denoted as pseudo-3D, is to use a stack of adjacent slices as input and produce a prediction for at least the central slice. This approach gives the network the possibility to capture 3D spatial information, with only a minor additional computational cost.

METHODS

In this study, we systematically evaluate the segmentation performance and computational costs of this pseudo-3D approach as a function of the number of input slices, and compare the results to conventional end-to-end 2D and 3D CNNs, and to triplanar orthogonal 2D CNNs. The standard pseudo-3D method regards the neighboring slices as multiple input image channels. We additionally design and evaluate a novel, simple approach where the input stack is a volumetric input that is repeatably convolved in 3D to obtain a 2D feature map. This 2D map is in turn fed into a standard 2D network. We conducted experiments using two different CNN backbone architectures and on eight diverse data sets covering different anatomical regions, imaging modalities, and segmentation tasks.

RESULTS

We found that while both pseudo-3D methods can process a large number of slices at once and still be computationally much more efficient than fully 3D CNNs, a significant improvement over a regular 2D CNN was only observed with two of the eight data sets. triplanar networks had the poorest performance of all the evaluated models. An analysis of the structural properties of the segmentation masks revealed no relations to the segmentation performance with respect to the number of input slices. A post hoc rank sum test which combined all metrics and data sets yielded that only our newly proposed pseudo-3D method with an input size of 13 slices outperformed almost all methods.

CONCLUSION

In the general case, multislice inputs appear not to improve segmentation results over using 2D or 3D CNNs. For the particular case of 13 input slices, the proposed novel pseudo-3D method does appear to have a slight advantage across all data sets compared to all other methods evaluated in this work.