Automated Segmentation of Normal and Diseased Coronary Arteries - The ASOCA Challenge

A continuous-action deep reinforcement learning-based agent for coronary artery centerline extraction in coronary CT angiography images

The lumen centerline of the coronary artery allows vessel reconstruction used to detect stenoses and plaques. Discrete-action-based centerline extraction methods suffer from artifacts and plaques. This study aimed to develop a continuous-action-based method which performs more effectively in cases involving artifacts or plaques. A continuous-action deep reinforcement learning-based model was trained to predict the artery's direction and radius value. The model is based on an Actor-Critic architecture. The Actor learns a deterministic policy to output the actions made by an agent. These actions indicate the centerline's direction and radius value consecutively. The Critic learns a value function to evaluate the quality of the agent's actions. A novel DDR reward was introduced to measure the agent's action (both centerline extraction and radius estimate) at each step. The method achieved an average OV of 95.7%, OF of 93.6%, OT of 97.3%, and AI of 0.22 mm in 80 test data. In 53 cases with artifacts or plaques, it achieved an average OV of 95.0%, OF of 91.5%, OT of 96.7%, and AI of 0.23 mm. The 95% limits of agreement between the reference and estimated radius values were - 0.46 mm and 0.43 mm in the 80 test data. Experiments demonstrate that the Actor-Critic architecture can achieve efficient centerline extraction and radius estimate. Compared with discrete-action-based methods, our method performs more effectively in cases involving artifacts or plaques. The extracted centerlines and radius values allow accurate coronary artery reconstruction that facilitates the detection of stenoses and plaques.

A new dataset for measuring the performance of blood vessel segmentation methods under distribution shifts

Creating a dataset for training supervised machine learning algorithms can be a demanding task. This is especially true for blood vessel segmentation since one or more specialists are usually required for image annotation, and creating ground truth labels for just a single image can take up to several hours. In addition, it is paramount that the annotated samples represent well the different conditions that might affect the imaged tissues as well as possible changes in the image acquisition process. This can only be achieved by considering samples that are typical in the dataset as well as atypical, or even outlier, samples. We introduce VessMAP, an annotated and highly heterogeneous blood vessel segmentation dataset acquired by carefully sampling relevant images from a large non-annotated dataset containing fluorescence microscopy images. Each image of the dataset contains metadata information regarding the contrast, amount of noise, density, and intensity variability of the vessels. Prototypical and atypical samples were carefully selected from the base dataset using the available metadata information, thus defining an assorted set of images that can be used for measuring the performance of segmentation algorithms on samples that are highly distinct from each other. We show that datasets traditionally used for developing new blood vessel segmentation algorithms tend to have low heterogeneity. Thus, neural networks trained on as few as four samples can generalize well to all other samples. In contrast, the training samples used for the VessMAP dataset can be critical to the generalization capability of a neural network. For instance, training on samples with good contrast leads to models with poor inference quality. Interestingly, while some training sets lead to Dice scores as low as 0.59, a careful selection of the training samples results in a Dice score of 0.85. Thus, the VessMAP dataset can be used for the development of new active learning methods for selecting relevant samples for manual annotation as well as for analyzing the robustness of segmentation models to distribution shifts of the data.

Deep learning techniques for automated coronary artery segmentation and coronary artery disease detection: A systematic review of the last decade (2013-2024)

BACKGROUND

Coronary artery disease (CAD) is the most common cardiovascular disease, exacting high morbidity and mortality worldwide. CAD is detected on coronary artery imaging; coronary artery segmentation (CAS) of the images is essential for coronary lesion characterization. Both CAD detection and CAS require expert input, are labor-intensive, and error-prone.

OBJECTIVES

Deep learning (DL) techniques have achieved significant success in CAS and CAD detection, with many studies published recently. This study is an up-to-date systematic review of research on automated DL models for CAS and CAD detection in the past decade (2013-2024).

METHOD

Using PRISMA methodology, an initial literature search of 1,589 publications was conducted, from which 97 high-impact Q1 studies were selected based on pre-defined eligibility criteria. These studies were analyzed in terms of DL techniques employed, datasets, modalities, and performance metrics.

RESULTS

Of the 97 studies, most of which were published after 2016, 47 focused on CAS, 49 on CAD detection, and one on both tasks. CNN-based models were dominant in both domains. For CAS, CCTA was the most frequently used input modality, and U-Net was employed in 38 out of 48 studies, with recent works incorporating attention mechanisms and graph neural networks. ASOCA was the most widely used benchmark dataset. For CAD detection, ECG was the most common modality, with 45 out of 50 studies utilizing CNNs, and 20 of those relying purely on CNN architectures. Hybrid and multimodal models have become more prominent in recent years.

CONCLUSION

This review identified several challenges, including limited public datasets, variability in performance metrics, and model complexity. Future studies should focus on larger, diverse datasets and lightweight models integrating explainable artificial intelligence and uncertainty quantification to improve clinical applicability.

Few-shot small vessel segmentation using a detail-preserving network enhanced by discriminator

Accurate segmentation of small vessels, such as coronary and pulmonary arteries, is crucial for early detection and treatment of vascular diseases. However, challenges persist due to the vessel's small size, complex structures, morphological variations, and limited annotated data. To address these challenges, we propose a detail-preserving network enhanced by a discriminator to improve the few-shot small vessel segmentation performance. The detail-preserving network constructs a complex module with multi-residual hybrid dilated convolution, which can enhance the network's receptive field while preserving the image's full detail features, enabling it to better capture the small vessel's structural features. Simultaneously, discriminator enhancement is incorporated into the training process through adversarial learning, effectively utilizing large amounts of unlabeled data to boost the generalization and robustness of the segmentation model. We validate the proposed method on in-house and public coronary artery datasets and public pulmonary artery datasets. Experimental results demonstrate that the proposed method significantly improves segmentation accuracy, particularly for small vessels. Compared with other state-of-the-art methods, the proposed method achieves higher accuracy, a lower false positive rate, and superior generalization capability, effectively assisting the clinical diagnosis of vessel diseases.

Reliability of characterising coronary artery flow with the flow-split outflow strategy: Comparison against the multiscale approach

BACKGROUND

In computational modelling of coronary haemodynamics, imposing patient-specific flow conditions is paramount, yet often impractical due to resource and time constraints, limiting the ability to perform a large number of simulations particularly for diseased cases.

OBJECTIVE

To compare coronary haemodynamics quantified using a simplified flow-split strategy with varying exponents against the clinically verified but computationally intensive multiscale simulations under both resting and hyperaemic conditions in arteries with varying degrees of stenosis.

METHODS

Six patient-specific left coronary artery trees were segmented and reconstructed, including three with severe (>70 %) and three with mild (<50 %) focal stenoses. Simulations were performed for the entire coronary tree to account for the flow-limiting effects from epicardial artery stenoses. Both a 0D-3D coupled multiscale model and a flow-split approach with four different exponents (2.0, 2.27, 2.33, and 3.0) were used. The resulting prominent haemodynamic metrics were statistically compared between the two methods.

RESULTS

Flow-split and multiscale simulations did not significantly differ under resting conditions regardless of the stenosis severity. However, under hyperaemic conditions, the flow-split method significantly overestimated the time-averaged wall shear stress by up to 16.8 Pa (p = 0.031) and underestimate the fractional flow reserve by 0.327 (p = 0.043), with larger discrepancies observed in severe stenoses than in mild ones. Varying the exponent from 2.0 to 3.0 within the flow-split methods did not significantly affect the haemodynamic results (p > 0.141).

CONCLUSIONS

Flow-split strategies with exponents between 2.0 and 3.0 are appropriate for modelling stenosed coronaries under resting conditions. Multiscale simulations are recommended for accurate modelling of hyperaemic conditions, especially in severely stenosed arteries.(247/250 words).

Auto-segmentation, radiomic reproducibility, and comparison of radiomics between manual and AI-derived segmentations for coronary arteries in cardiac [18F]NaF PET/CT images

BACKGROUND

[18F]NaF is a potential biomarker for assessing cardiac risk. Automated analysis of [18F]NaF positron emission tomography (PET) images, specifically through quantitative image analysis ("radiomics"), can potentially enhance diagnostic accuracy and personalised patient management. However, it is essential to evaluate the reproducibility and reliability of radiomic features to ensure their clinical applicability. This study aimed to (i) develop and evaluate an automated model for coronary artery segmentation using [18F]NaF PET and calcium scoring computed tomography (CSCT) images, (ii) assess inter- and intra-observer radiomic reproducibility from manual segmentations, and (iii) evaluate the radiomics reliability from AI-derived segmentations by comparison with manual segmentations.

RESULTS

141 patients from the "effects of Vitamin K and Colchicine on vascular calcification activity" (VikCoVac, ACTRN12616000024448) trial were included. 113 were used to train an auto-segmentation model using nnUNet on [18F]NaF PET and CSCT images. Reproducibility of inter- and intra-observer radiomics and reliability of radiomics from AI-derived segmentations was assessed using lower bound of intraclass correlation coefficient (ICC). The auto-segmentation model achieved an average Dice Similarity Coefficient of 0.61 ± 0.05, having no statistically significant difference compared to the intra-observer variability (p = 0.922). For the unfiltered images, 47(12.6%) CT and 25(7.5%) PET radiomics were inter-observer reproducible, while 133(35.8%) CT and 57(15.3%) PET radiomics were intra-observer reproducible. 7(9.7%) CT and 18(25.0%) PET first-order features, as well as 17(17.7%) CT GLCM features, were reproducible for both inter- and intra-observer analyses. 9.8% and 16.8% of radiomics from AI-derived segmentations showed excellent and good reliability. First-order features were most reliable (ICC > 0.75; 78/144[54.2%]) and shape features least (2/112[1.8%]). CT features demonstrated greater reliability (147/428[34.3%]) than PET (81/428 [18.9%]). Features from the left anterior descending (76/214[35.5%]) and right coronary artery (75/214[35.0%]) were more reliability than the circumflex (49/214[22.9%]) and left main (28/214[13.1%]) arteries.

CONCLUSIONS

An effective segmentation model for coronary arteries was developed and reproducible [18F]NaF PET/CSCT radiomics were identified through inter- and intra-observer assessments, supporting their clinical applicability. The reliability of radiomics from AI-derived segmentations compared to manual segmentations was highlighted. The novelty of [18F]NaF as a biomarker underscores its potential in providing unique insights into vascular calcification activity and cardiac risk assessment.

CLINICAL TRIAL REGISTRATION

VIKCOVAC trial ("effects of Vitamin K and Colchicine on vascular calcification activity"). Unique identifier: ACTRN12616000024448. URL: https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=368825 .

Fast and automatic coronary artery segmentation using nnU-Net for non-contrast enhanced magnetic resonance coronary angiography

Non-contrast enhanced magnetic resonance coronary angiography (MRCA) is a promising coronary heart disease screening modality. However, its clinical application is hindered by inherent limitations, including low spatial resolution and insufficient contrast between coronary arteries and surrounding tissues. These technical challenges impede fast and automatic coronary artery segmentation. To tackle these issues, we propose a self-configuring deep learning-based approach for automating the segmentation of coronary arteries in MRCA images. The nnU-Net model was trained on MRCA data from 134 subjects and tested on data from 114 subjects. Two radiologists qualitatively evaluated all segmented arteries as good to excellent. Using coronary computed tomography angiography (CCTA) data from the 114 tested subjects as the gold standard. Specifically, we compared the number of branches, the total branch length, and the distance from the base of the coronary sinus to the origin of the corresponding main coronary artery obtained from manual and artificial intelligence measurements in MRCA images with those obtained from CCTA. Experiment results demonstrated that in validation nnU-Net can accurately segment from MRCA images with the Dice score of 0.903 and 0.962 for major coronary arteries and aorta, respectively.In Testing, nnU-Net achieved the Dice score of 0.726 and 0.890 for major coronary arteries and aorta, respectively. Integrating MRCA with nnU-Net to extract coronary arteries offers a non-invasive screening tool for the detection of coronary heart disease, potentially enhancing early detection and reducing reliance from CCTA.

3DGR-CT: Sparse-view CT reconstruction with a 3D Gaussian representation

Sparse-view computed tomography (CT) reduces radiation exposure by acquiring fewer projections, making it a valuable tool in clinical scenarios where low-dose radiation is essential. However, this often results in increased noise and artifacts due to limited data. In this paper we propose a novel 3D Gaussian representation (3DGR) based method for sparse-view CT reconstruction. Inspired by recent success in novel view synthesis driven by 3D Gaussian splatting, we leverage the efficiency and expressiveness of 3D Gaussian representation as an alternative to implicit neural representation. To unleash the potential of 3DGR for CT imaging scenario, we propose two key innovations: (i) FBP-image-guided Guassian initialization and (ii) efficient integration with a differentiable CT projector. Extensive experiments and ablations on diverse datasets demonstrate the proposed 3DGR-CT consistently outperforms state-of-the-art counterpart methods, achieving higher reconstruction accuracy with faster convergence. Furthermore, we showcase the potential of 3DGR-CT for real-time physical simulation, which holds important clinical applications while challenging for implicit neural representations. Code available at: https://github.com/SigmaLDC/3DGR-CT.

VSNet: Vessel Structure-aware Network for hepatic and portal vein segmentation

Identifying and segmenting hepatic and portal veins (two predominant vascular systems in the liver, from CT scans) play a crucial role for clinicians in preoperative planning for treatment strategies. However, existing segmentation models often struggle to capture fine details of minor veins. In this article, we introduce Vessel Structure-aware Network (VSNet), a multi-task learning model with vessel-growing decoder, to address the challenge. VSNet excels at accurate segmentation by capturing the topological features of both minor veins while preserving correct connectivity from minor vessels to trucks. We also build and publish the largest dataset (303 cases) for hepatic and portal vessel segmentation. Through comprehensive experiments, we demonstrate that VSNet achieves the best Dice for hepatic vein of 0.824 and portal vein of 0.807 on our proposed dataset, significantly outperforming other popular segmentation models. The source code and dataset are publicly available at https://github.com/XXYZB/VSNet.

HWA-ResMamba: automatic segmentation of coronary arteries based on residual Mamba with high-order wavelet-enhanced convolution and attention feature aggregation

Objective.Automatic segmentation of coronary arteries is a crucial prerequisite in assisting in the diagnosis of coronary artery disease. However, due to the fuzzy boundaries, small-slender branches, and significant individual variations, automatic segmentation of coronary arteries is extremely challenging.Approach.This study proposes a residual Mamba with high-order wavelet-enhanced convolution and attention feature aggregation (HWA-ResMamba) for coronary arteries segmentation. The network consists of three core modules: high-order wavelet-enhanced convolution block (HWCB), residual Mamba (ResMamba), and attention feature aggregation (AFA) module. Firstly, the HWCB captures low-frequency information of the image in the shallow layers of the network, allowing for detailed exploration of subtle changes in the boundaries of coronary arteries. Secondly, the ResMamba module establishes long-range dependencies between features in the deep layers of the encoder and at the beginning of the decoder, improving the continuity of the segmentation process. Finally, the AFA module in the decoder reduces semantic differences between the encoder and decoder, which can capture small-slender coronary artery branches and further improve segmentation accuracy.Main results.Experiments on three coronary artery segmentation datasets have shown that the HWA-ResMamba outperforms other state-of-the-art methods in performance and generalization. Specifically, in the self-built dataset, HWA-ResMamba obtained Dice of 0.8857 and Hausdorff Distance (HD) of 1.9028, outperforming nnUnet by 0.0521, and 0.5489, respectively. HWA-ResMamba obtained Dice of 0.8371, and 0.7861 in the two public datasets, outperforming nnUnet by 0.0255, and 0.0107, respectively.Significance.Our method can accurately segment coronary arteries and can contribute to improved diagnosis and assessment of CAD.

MedShapeNet - a large-scale dataset of 3D medical shapes for computer vision

OBJECTIVES

The shape is commonly used to describe the objects. State-of-the-art algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models are used. This is seen from the growing popularity of ShapeNet (51,300 models) and Princeton ModelNet (127,915 models). However, a large collection of anatomical shapes (e.g., bones, organs, vessels) and 3D models of surgical instruments is missing.

METHODS

We present MedShapeNet to translate data-driven vision algorithms to medical applications and to adapt state-of-the-art vision algorithms to medical problems. As a unique feature, we directly model the majority of shapes on the imaging data of real patients. We present use cases in classifying brain tumors, skull reconstructions, multi-class anatomy completion, education, and 3D printing.

RESULTS

By now, MedShapeNet includes 23 datasets with more than 100,000 shapes that are paired with annotations (ground truth). Our data is freely accessible via a web interface and a Python application programming interface and can be used for discriminative, reconstructive, and variational benchmarks as well as various applications in virtual, augmented, or mixed reality, and 3D printing.

CONCLUSIONS

MedShapeNet contains medical shapes from anatomy and surgical instruments and will continue to collect data for benchmarks and applications. The project page is: https://medshapenet.ikim.nrw/.

A Robust Blood Vessel Segmentation Technique for Angiographic Images Employing Multi-Scale Filtering Approach

Background: This study focuses on the critical task of blood vessel segmentation in medical image analysis, essential for diagnosing cardiovascular diseases and enabling effective treatment planning. Although deep learning architectures often produce very high segmentation results in medical images, coronary computed tomography angiography (CTA) images are more challenging than invasive coronary angiography (ICA) images due to noise and the complexity of vessel structures. Methods: Classical architectures for medical images, such as U-Net, achieve only moderate accuracy, with an average Dice score of 0.722. Results: This study introduces Morpho-U-Net, an enhanced U-Net architecture that integrates advanced morphological operations, including Gaussian blurring, thresholding, and morphological opening/closing, to improve vascular integrity, reduce noise, and achieve a higher Dice score of 0.9108, a precision of 0.9341, and a recall of 0.8872. These enhancements demonstrate superior robustness to noise and intricate vessel geometries. Conclusions: This pre-processing filter effectively reduces noise by grouping neighboring pixels with similar intensity values, allowing the model to focus on relevant anatomical structures, thus outperforming traditional methods in handling the challenges posed by CTA images.

An automated and time-efficient framework for simulation of coronary blood flow under steady and pulsatile conditions

BACKGROUND AND OBJECTIVE

Invasive fractional flow reserve (FFR) measurement is the gold standard method for coronary artery disease (CAD) diagnosis. FFR-CT exploits computational fluid dynamics (CFD) for non-invasive evaluation of FFR, simulating coronary flow in virtual geometries reconstructed from computed tomography (CT), but suffers from cost-intensive computing process and uncertainties in the definition of patient specific boundary conditions (BCs). In this work, we investigated the use of time-averaged steady BCs, compared to pulsatile to reduce the computational time and deployed a self-adjusting method for the tuning of BCs to patient-specific clinical data.

METHODS

133 coronary arteries were reconstructed form CT images of patients suffering from CAD. For each vessel, invasive FFR was measured. After segmentation, the geometries were prepared for CFD simulation by clipping the outlets and discretizing into tetrahedral mesh. Steady BCs were defined in two steps: (i) rest BCs were extrapolated from clinical and image-derived data; (ii) hyperemic BCs were computed from resting conditions. Flow rate was iteratively adjusted during the simulation, until patient's aortic pressure was matched. Pulsatile BCs were defined exploiting the convergence values of steady BCs. After CFD simulation, lesion-specific hemodynamic indexes were computed and compared between group of patients for which surgery was indicated and not. The whole pipeline was implemented as a straightforward process, in which each single step is performed automatically.

RESULTS

Steady and pulsatile FFR-CT yielded a strong correlation (r = 0.988, p < 0.001) and correlated with invasive FFR (r = 0.797, p < 0.001). The per-point difference between the pressure and FFR-CT field predicted by the two methods was below 1 % and 2 %, respectively. Both approaches exhibited a good diagnostic performance: accuracy was 0.860 and 0.864, the AUC was 0.923 and 0.912, for steady and pulsatile case, respectively. The computational time required by steady BCs CFD was approximatively 30-folds lower than pulsatile case.

CONCLUSIONS

This work shows the feasibility of using steady BCs CFD for computing the FFR-CT in coronary arteries, as well as its computational and diagnostic performance within a fully automated pipeline.

DIAS: A dataset and benchmark for intracranial artery segmentation in DSA sequences

The automated segmentation of Intracranial Arteries (IA) in Digital Subtraction Angiography (DSA) plays a crucial role in the quantification of vascular morphology, significantly contributing to computer-assisted stroke research and clinical practice. Current research primarily focuses on the segmentation of single-frame DSA using proprietary datasets. However, these methods face challenges due to the inherent limitation of single-frame DSA, which only partially displays vascular contrast, thereby hindering accurate vascular structure representation. In this work, we introduce DIAS, a dataset specifically developed for IA segmentation in DSA sequences. We establish a comprehensive benchmark for evaluating DIAS, covering full, weak, and semi-supervised segmentation methods. Specifically, we propose the vessel sequence segmentation network, in which the sequence feature extraction module effectively captures spatiotemporal representations of intravascular contrast, achieving intracranial artery segmentation in 2D+Time DSA sequences. For weakly-supervised IA segmentation, we propose a novel scribble learning-based image segmentation framework, which, under the guidance of scribble labels, employs cross pseudo-supervision and consistency regularization to improve the performance of the segmentation network. Furthermore, we introduce the random patch-based self-training framework, aimed at alleviating the performance constraints encountered in IA segmentation due to the limited availability of annotated DSA data. Our extensive experiments on the DIAS dataset demonstrate the effectiveness of these methods as potential baselines for future research and clinical applications. The dataset and code are publicly available at https://doi.org/10.5281/zenodo.11401368 and https://github.com/lseventeen/DIAS.

A new understanding of coronary curvature and haemodynamic impact on the course of plaque onset and progression

The strong link between atherosclerosis and luminal biomechanical stresses is well established. Yet, this understanding has not translated into preventative coronary diagnostic imaging, particularly due to the under-explored role of coronary anatomy and haemodynamics in plaque onset, which we aim to address with this work. The left coronary trees of 20 non-stenosed (%diameter stenosis [%DS] = 0), 12 moderately (0 < %DS < 70) and 7 severely (%DS ≥ 70) stenosed cases were dissected into bifurcating and non-bifurcating segments for whole-tree and segment-specific comparisons, correlating nine three-dimensional coronary anatomical features, topological shear variation index (TSVI) and luminal areas subject to low time-average endothelial shear stress (%LowTAESS), high oscillatory shear index (%HighOSI) and high relative residence time (%HighRRT). We found that TSVI is the only metric consistently differing between non-stenosed and stenosed cases across the whole tree, bifurcating and non-bifurcating segments (p < 0.002, AUC = 0.876), whereas average curvature and %HighOSI differed only for the whole trees (p < 0.024) and non-bifurcating segments (p < 0.027), with AUC > 0.711. Coronary trees with moderate or severe stenoses differed only in %LowTAESS (p = 0.009) and %HighRRT (p = 0.012). This suggests TSVI, curvature and %HighOSI are potential factors driving plaque onset, with greater predictive performance than the previously recognized %LowTAESS and %HighRRT, which appears to play a role in plaque progression.

Impact of minimal lumen segmentation uncertainty on patient-specific coronary simulations: A look at FFRCT

We examined the effect of minimal lumen segmentation uncertainty on Fractional Flow Reserve obtained from Coronary Computed Tomography AngiographyFFR CT . A total of 14 patient-specific coronary models with different stenosis locations and degrees of severity were enrolled in this study. The optimal segmented coronary lumens were disturbed using intra± 6 % and inter-operator± 15 % variations on the segmentation threshold.FFR CT was evaluated in each case by 3D-OD CFD simulations. The findings suggest that the sensitivity ofFFR CT to this type of uncertainty increases distally and with the stenosis severity. Cases with moderate or severe distal coronary lesions should undergo either exact and thorough segmentation operations or invasive FFR measurements, particularly if theFFR CT is close to the cutoff (0.80). Therefore, we conclude that it is crucial to consider the lesion's location and degree of severity when evaluatingFFR CT results.

WITHDRAWN: Coronary artery segmentation based on ACMA-Net and unscented Kalman filter algorithm

This article has been withdrawn at the request of the author(s) and/or editor. The Publisher apologizes for any inconvenience this may cause. The full Elsevier Policy on Article Withdrawal can be found at https://www.elsevier.com/about/policies/article-withdrawal.

A fully automated deep learning approach for coronary artery segmentation and comprehensive characterization

Coronary computed tomography angiography (CCTA) allows detailed assessment of early markers associated with coronary artery disease (CAD), such as coronary artery calcium (CAC) and tortuosity (CorT). However, their analysis can be time-demanding and biased. We present a fully automated pipeline that performs (i) coronary artery segmentation and (ii) CAC and CorT objective analysis. Our method exploits supervised learning for the segmentation of the lumen, and then, CAC and CorT are automatically quantified. 281 manually annotated CCTA images were used to train a two-stage U-Net-based architecture. The first stage employed a 2.5D U-Net trained on axial, coronal, and sagittal slices for preliminary segmentation, while the second stage utilized a multichannel 3D U-Net for refinement. Then, a geometric post-processing was implemented: vessel centerlines were extracted, and tortuosity score was quantified as the count of branches with three or more bends with change in direction forming an angle >45°. CAC scoring relied on image attenuation. CAC was detected by setting a patient specific threshold, then a region growing algorithm was applied for refinement. The application of the complete pipeline required <5 min per patient. The model trained for coronary segmentation yielded a Dice score of 0.896 and a mean surface distance of 1.027 mm compared to the reference ground truth. Tracts that presented stenosis were correctly segmented. The vessel tortuosity significantly increased locally, moving from proximal, to distal regions (p < 0.001). Calcium volume score exhibited an opposite trend (p < 0.001), with larger plaques in the proximal regions. Volume score was lower in patients with a higher tortuosity score (p < 0.001). Our results suggest a linked negative correlation between tortuosity and calcific plaque formation. We implemented a fast and objective tool, suitable for population studies, that can help clinician in the quantification of CAC and various coronary morphological parameters, which is helpful for CAD risk assessment.

An Anatomy- and Topology-Preserving Framework for Coronary Artery Segmentation

Coronary artery segmentation is critical for coronary artery disease diagnosis but challenging due to its tortuous course with numerous small branches and inter-subject variations. Most existing studies ignore important anatomical information and vascular topologies, leading to less desirable segmentation performance that usually cannot satisfy clinical demands. To deal with these challenges, in this paper we propose an anatomy- and topology-preserving two-stage framework for coronary artery segmentation. The proposed framework consists of an anatomical dependency encoding (ADE) module and a hierarchical topology learning (HTL) module for coarse-to-fine segmentation, respectively. Specifically, the ADE module segments four heart chambers and aorta, and thus five distance field maps are obtained to encode distance between chamber surfaces and coarsely segmented coronary artery. Meanwhile, ADE also performs coronary artery detection to crop region-of-interest and eliminate foreground-background imbalance. The follow-up HTL module performs fine segmentation by exploiting three hierarchical vascular topologies, i.e., key points, centerlines, and neighbor connectivity using a multi-task learning scheme. In addition, we adopt a bottom-up attention interaction (BAI) module to integrate the feature representations extracted across hierarchical topologies. Extensive experiments on public and in-house datasets show that the proposed framework achieves state-of-the-art performance for coronary artery segmentation.

Learning Better Registration to Learn Better Few-Shot Medical Image Segmentation: Authenticity, Diversity, and Robustness

In this work, we address the task of few-shot medical image segmentation (MIS) with a novel proposed framework based on the learning registration to learn segmentation (LRLS) paradigm. To cope with the limitations of lack of authenticity, diversity, and robustness in the existing LRLS frameworks, we propose the better registration better segmentation (BRBS) framework with three main contributions that are experimentally shown to have substantial practical merit. First, we improve the authenticity in the registration-based generation program and propose the knowledge consistency constraint strategy that constrains the registration network to learn according to the domain knowledge. It brings the semantic-aligned and topology-preserved registration, thus allowing the generation program to output new data with great space and style authenticity. Second, we deeply studied the diversity of the generation process and propose the space-style sampling program, which introduces the modeling of the transformation path of style and space change between few atlases and numerous unlabeled images into the generation program. Therefore, the sampling on the transformation paths provides much more diverse space and style features to the generated data effectively improving the diversity. Third, we first highlight the robustness in the learning of segmentation in the LRLS paradigm and propose the mix misalignment regularization, which simulates the misalignment distortion and constrains the network to reduce the fitting degree of misaligned regions. Therefore, it builds regularization for these regions improving the robustness of segmentation learning. Without any bells and whistles, our approach achieves a new state-of-the-art performance in few-shot MIS on two challenging tasks that outperform the existing LRLS-based few-shot methods. We believe that this novel and effective framework will provide a powerful few-shot benchmark for the field of medical image and efficiently reduce the costs of medical image research. All of our code will be made publicly available online.

Mining multi-center heterogeneous medical data with distributed synthetic learning

Overcoming barriers on the use of multi-center data for medical analytics is challenging due to privacy protection and data heterogeneity in the healthcare system. In this study, we propose the Distributed Synthetic Learning (DSL) architecture to learn across multiple medical centers and ensure the protection of sensitive personal information. DSL enables the building of a homogeneous dataset with entirely synthetic medical images via a form of GAN-based synthetic learning. The proposed DSL architecture has the following key functionalities: multi-modality learning, missing modality completion learning, and continual learning. We systematically evaluate the performance of DSL on different medical applications using cardiac computed tomography angiography (CTA), brain tumor MRI, and histopathology nuclei datasets. Extensive experiments demonstrate the superior performance of DSL as a high-quality synthetic medical image provider by the use of an ideal synthetic quality metric called Dist-FID. We show that DSL can be adapted to heterogeneous data and remarkably outperforms the real misaligned modalities segmentation model by 55% and the temporal datasets segmentation model by 8%.

Transforming medical imaging with Transformers? A comparative review of key properties, current progresses, and future perspectives

Transformer, one of the latest technological advances of deep learning, has gained prevalence in natural language processing or computer vision. Since medical imaging bear some resemblance to computer vision, it is natural to inquire about the status quo of Transformers in medical imaging and ask the question: can the Transformer models transform medical imaging? In this paper, we attempt to make a response to the inquiry. After a brief introduction of the fundamentals of Transformers, especially in comparison with convolutional neural networks (CNNs), and highlighting key defining properties that characterize the Transformers, we offer a comprehensive review of the state-of-the-art Transformer-based approaches for medical imaging and exhibit current research progresses made in the areas of medical image segmentation, recognition, detection, registration, reconstruction, enhancement, etc. In particular, what distinguishes our review lies in its organization based on the Transformer's key defining properties, which are mostly derived from comparing the Transformer and CNN, and its type of architecture, which specifies the manner in which the Transformer and CNN are combined, all helping the readers to best understand the rationale behind the reviewed approaches. We conclude with discussions of future perspectives.

Annotated computed tomography coronary angiogram images and associated data of normal and diseased arteries

Computed Tomography Coronary Angiography (CTCA) is a non-invasive method to evaluate coronary artery anatomy and disease. CTCA is ideal for geometry reconstruction to create virtual models of coronary arteries. To our knowledge there is no public dataset that includes centrelines and segmentation of the full coronary tree. We provide anonymized CTCA images, voxel-wise annotations and associated data in the form of centrelines, calcification scores and meshes of the coronary lumen in 20 normal and 20 diseased cases. Images were obtained along with patient information with informed, written consent as part of the Coronary Atlas. Cases were classified as normal (zero calcium score with no signs of stenosis) or diseased (confirmed coronary artery disease). Manual voxel-wise segmentations by three experts were combined using majority voting to generate the final annotations. Provided data can be used for a variety of research purposes, such as 3D printing patient-specific models, development and validation of segmentation algorithms, education and training of medical personnel and in-silico analyses such as testing of medical devices.

Towards automated coronary artery segmentation: A systematic review

BACKGROUND AND OBJECTIVE

Vessel segmentation is the first processing stage of 3D medical images for both clinical and research use. Current segmentation methods are tedious and time consuming, requiring significant manual correction and hence are infeasible to use in large data sets.

METHODS

Here, we review and analyse available coronary artery segmentation methods, focusing on fully automated methods capable of handling the rapidly growing medical images available. All manuscripts published since 2010 are systematically reviewed, categorised into different groups based on the approach taken, and characteristics of the different approaches as well as trends over the past decade are explored.

RESULTS

The manuscripts were divided intro three broad categories, consisting of region growing, voxelwise prediction and partitioning approaches. The most common approach overall was region growing, particularly using active contour models, however these have had a sharp fall in popularity in recent years with convolutional neural networks becoming significantly more popular.

CONCLUSIONS

The systematic review of current coronary artery segmentation methods shows interesting trends, with rising popularity of machine learning methods, a focus on efficient methods, and falling popularity of computationally expensive processing steps such as vesselness and multiplanar reformation.

Deep learning methods for lesion detection on mammography images: a comparative analysis

Automatic lesion segmentation in mammography images assists in the diagnosis of breast cancer, which is the most common type of cancer especially among women. The robust segmentation of mammography images has been considered a backbreaking task due to: i) the low contrast of the lesion boundaries; ii) the extremely variable lesions' sizes and shapes; and iii) some extremely small lesions on the mammogram image. To overcome these drawbacks, Deep Learning methods have been implemented and have shown impressive results when applied to medical image segmentation. This work presents a benchmark for breast lesion segmentation in mammography images, where six state-of-the-art methods were evaluated on 1692 mammograms from a public dataset (CBIS-DDSM), and compared considering the following six metrics: i) Dice coefficient; ii) Jaccard index; iii) accuracy; iv) recall; v) specificity; and vi) precision. The base U-Net, UNETR, DynUNet, SegResNetVAE, RF-Net, MDA-Net architectures were trained with a combination of the cross-entropy and Dice loss functions. Although the networks presented Dice scores superior to 86%, two of them managed to distinguish themselves. In general, the results demonstrate the efficiency of the MDA-Net and DynUnet with Dice scores of 90.25% and 89.67%, and accuracy of 93.48% and 93.03%, respectively. Clinical Relevance--- The presented comparative study allowed to identify the current performance of deep learning strategies on the segmentation of breast lesions.

A new and automated risk prediction of coronary artery disease using clinical endpoints and medical imaging-derived patient-specific insights: protocol for the retrospective GeoCAD cohort study

INTRODUCTION

Coronary artery disease (CAD) is the leading cause of death worldwide. More than a quarter of cardiovascular events are unexplained by current absolute cardiovascular disease risk calculators, and individuals without clinical risk factors have been shown to have worse outcomes. The 'anatomy of risk' hypothesis recognises that adverse anatomical features of coronary arteries enhance atherogenic haemodynamics, which in turn mediate the localisation and progression of plaques. We propose a new risk prediction method predicated on CT coronary angiography (CTCA) data and state-of-the-art machine learning methods based on a better understanding of anatomical risk for CAD. This may open new pathways in the early implementation of personalised preventive therapies in susceptible individuals as a potential key in addressing the growing burden of CAD.

METHODS AND ANALYSIS

GeoCAD is a retrospective cohort study in 1000 adult patients who have undergone CTCA for investigation of suspected CAD. It is a proof-of-concept study to test the hypothesis that advanced image-derived patient-specific data can accurately predict long-term cardiovascular events. The objectives are to (1) profile CTCA images with respect to variations in anatomical shape and associated haemodynamic risk expressing, at least in part, an individual's CAD risk, (2) develop a machine-learning algorithm for the rapid assessment of anatomical risk directly from unprocessed CTCA images and (3) to build a novel CAD risk model combining traditional risk factors with these novel anatomical biomarkers to provide a higher accuracy CAD risk prediction tool.

ETHICS AND DISSEMINATION

The study protocol has been approved by the St Vincent's Hospital Human Research Ethics Committee, Sydney-2020/ETH02127 and the NSW Population and Health Service Research Ethics Committee-2021/ETH00990. The project outcomes will be published in peer-reviewed and biomedical journals, scientific conferences and as a higher degree research thesis.