Carotid Plaque Classification Research Articles

WAL-Net: Weakly supervised auxiliary task learning network for carotid plaques classification

The classification of carotid artery ultrasound images is a crucial step for diagnosing carotid plaques, holding significant clinical relevance for predicting the risk of stroke. Recent research suggests that utilizing supervised plaque segmentation as an auxiliary task for classification can enhance performance by leveraging the correlation between segmentation and classification tasks. However, this approach relies on obtaining a substantial amount of challenging-to-acquire segmentation annotations. This paper proposes a novel weakly supervised auxiliary task learning network model (WAL-Net) to explore the interdependence between carotid plaque classification and segmentation tasks. The plaque classification task is the primary one, while the plaque segmentation task serves as the auxiliary one, providing valuable information to enhance the performance of the primary task. Weakly supervised learning is adopted in the auxiliary task in which the segmentation masks are not provided. Experiments and evaluations are conducted on a dataset comprising 1270 carotid plaque ultrasound images from Wuhan University Zhongnan Hospital. Results indicate that WAL-Net achieved an approximately 1.3% improvement in carotid plaque classification accuracy compared to the baseline network. Specifically, the accuracy of mixed-echoic plaques classification increased by approximately 3.3%, demonstrating the effectiveness of WAL-Net.

A self-supervised fusion network for carotid plaque ultrasound image classification.

Carotid plaque classification from ultrasound images is crucial for predicting ischemic stroke risk. While deep learning has shown effectiveness, it heavily relies on substantial labeled datasets. Achieving high performance with limited labeled images is essential for clinical use. Self-supervised learning (SSL) offers a potential solution; however, the existing works mainly focus on constructing the SSL tasks, neglecting the use of multiple tasks for pretraining. To overcome these limitations, this study proposed a self-supervised fusion network (Fusion-SSL) for carotid plaque ultrasound image classification with limited labeled data. Fusion-SSL consists of two SSL tasks: classifying image block order (Ordering) and predicting image rotation angle (Rotating). A dual-branch residual neural network was developed to fuse feature presentations learned by the two tasks, which can extract richer visual boundary shape and contour information than a single task. In this experiment, 1270 carotid plaque ultrasound images were collected from 844 patients at Zhongnan Hospital (Wuhan, China). The results showed that Fusion-SSL outperforms single SSL methods across different percentages of labeled training data, ranging from 10 to 100%. Moreover, with only 40% labeled training data, Fusion-SSL achieved comparable results to a single SSL method (predicting image rotation angle) with 100% labeled data. These results indicate that Fusion-SSL could be beneficial for the classification of carotid plaques and the early warning of a stroke in clinical practice.

CPTV: Classification by tracking of carotid plaque in ultrasound videos.

The risk assessment of carotid plaque is strongly related to the plaque echo status in ultrasound. However, the echo classification of carotid plaques based on ultrasound remains challenging due to the changes in plaque shape and semantics, along with the complex vascular environment. This study proposed a framework for Classification of Plaque by Tracking Videos (CPTV). To the best of our knowledge, this is the first study on plaque classification by tracking ultrasound video rather than a sonographic view, which achieves accurate localization and stable echo classification. In the tracking task, Multi-scale Decoupling Tracking (MDTrack) module including Multi-scale Dilated Encoder (MDE) and Internal-Exterior Feature Decoupling (IEFD) was proposed to solve the problems caused by shape and semantic variations to achieve accurate plaque localization in ultrasound. In the classification task, the Tracking-assisted 3D Attention (T3D-Attention) module included recombination and 3D-Attention extracted plaque features and echo-related features in the vascular environment. The experiments demonstrated that the performance of CPTV is better than current mainstream tracking and classification methods, indicating that the tracking-assistance classification is a kind of enhancement method with high universality and stability in the plaque in ultrasound.

CT angiography-based radiomics as a tool for carotid plaque characterization: a pilot study.

Radiomics is a quantitative method able to analyze a high-throughput extraction of minable imaging features. Herein, we aim to develop a CT angiography-based radiomics analysis and machine learning model for carotid plaques to discriminate vulnerable from no vulnerable plaques. Thirty consecutive patients with carotid atherosclerosis were enrolled in this pilot study. At surgery, a binary classification of plaques was adopted ("hard" vs "soft"). Feature extraction was performed using the R software package Moddicom. Pairwise feature interdependencies were evaluated using the Spearman rank correlation coefficient. A univariate analysis was performed to assess the association between each feature and the plaque classification and chose top-ranked features. The feature predictive value was investigated using binary logistic regression. A stepwise backward elimination procedure was performed to minimize the Akaike information criterion (AIC). The final significant features were used to build the models for binary classification of carotid plaques, including logistic regression (LR), support vector machine (SVM), and classification and regression tree analysis (CART). All models were cross-validated using fivefold cross validation. Class-specific accuracy, precision, recall and F-measure evaluation metrics were used to quantify classifier output quality. A total of 230 radiomics features were extracted from each plaque. Pairwise Spearman correlation between features reported a high level of correlations, with more than 80% correlating with at least one other feature at |ρ|> 0.8. After a stepwise backward elimination procedure, the entropy and volume features were found to be the most significantly associated with the two plaque groups (p < 0.001), with AUCs of 0.92 and 0.96, respectively. The best performance was registered by the SVM classifier with the RBF kernel, with accuracy, precision, recall and F-score equal to 86.7, 92.9, 81.3 and 86.7%, respectively. The CART classification tree model for the entropy and volume features model achieved 86.7% well-classified plaques and an AUC of 0.987. This pilot study highlighted the potential of CTA-based radiomics and machine learning to discriminate plaque composition. This new approach has the potential to provide a reliable method to improve risk stratification in patients with carotid atherosclerosis.

Object-Specific Four-Path Network for Stroke Risk Stratification of Carotid Arteries in Ultrasound Images.

Atherosclerotic carotid plaques have been shown to be closely associated with the risk of stroke. Since patients with symptomatic carotid plaques have a greater risk for stroke, stroke risk stratification based on the classification of carotid plaques into symptomatic or asymptomatic types is crucial in diagnosis, treatment planning, and medical treatment monitoring. A deep learning technique would be a good choice for implementing classification. Usually, to acquire a high-accuracy classification, a specific network architecture needs to be designed for a given classification task. In this study, we propose an object-specific four-path network (OSFP-Net) for stroke risk assessment by integrating ultrasound carotid plaques in both transverse and longitudinal sections of the bilateral carotid arteries. Each path of the OSFP-Net comprises of a feature extraction subnetwork (FE) and a feature downsampling subnetwork (FD). The FEs in the four paths use the same network structure to automatically extract features from ultrasound images of carotid plaques. The FDs use different object-specific pooling strategies for feature downsampling based on the observation that the sizes and shapes in the feature maps obtained from FEs should be different. The object-specific pooling strategies enable the network to accept arbitrarily sized carotid plaques as input and to capture a more informative context for improving the classification accuracy. Extensive experimental studies on a clinical dataset consisting of 333 subjects with 1332 carotid plaques show the superiority of our OSFP-Net against several state-of-the-art deep learning-based methods. The experimental results demonstrate better clinical agreement between the ground truth and the prediction, which indicates its great potential for use as a risk stratification and as a monitoring tool in the management of patients at risk for stroke.

Using Deep Convolutional Neural Networks to Automate Classification of Carotid Plaques from Ultrasound Imaging

Using Deep Convolutional Neural Networks to Automate Classification of Carotid Plaques from Ultrasound Imaging

Carotid Artery Plaque Identification and Display System (MRI-CAPIDS) Using Opensource Tools.

Magnetic resonance imaging (MRI) represents one modality in atherosclerosis risk assessment, by permitting the classification of carotid plaques into either high- or low-risk lesions. Although MRI is generally used for observing the impact of atherosclerosis on vessel lumens, it can also show both the size and composition of itself, as well as plaque information, thereby providing information beyond that of simple stenosis. Software systems are a valuable aid in carotid artery stenosis assessment wherein commercial software is readily available but is not accessible to all practitioners because of its often high cost. This study focuses on the development of a software system designed entirely for registration, marking, and 3D visualization of the wall and lumen, using freely available open-source tools and libraries. It was designed to be free from “feature bloat” and avoid “feature-creep.” The image loading and display module of the modified QDCM library was improved by a minimum of 10,000%. A Bezier function was used in order to smoothen the curve of the polygon (referring to the shape formed by the marked points) by interpolating additional points between the marked points. This smoother curve led to a smoother 3D view of the lumen and wall.

Comparative Study of Protein Expression Levels of Five Plaque Biomarkers and Relation with Carotid Plaque Type Classification in Patients after Carotid Endarterectomy.

Atherosclerosis is an inflammatory process resulting in local plaque deposition in the vessel wall of arteries with symptoms to various areas of vascular tree. Identification of patients with progressive advanced atherosclerotic disease is mainly based on the known characteristics of the vulnerable or recently ruptured plaque. Molecular and cellular features associated with the vulnerable plaque are considered potential diagnostic markers for plaque rupture and thrombosis. Here, protein expression levels of the metalloproteases MMP-1, MMP-9, osteopontin (OPN), and cytokines TNFα and IL-6 in tissue extracts of carotid plaques in patients after endarterectomy were estimated by Western immunoblotting, after SDS-PAGE analysis and evaluated based on the ultrasonographic plaque morphology. The gender and age effect was also examined. MMP-1, MMP-9, and IL-6 were expressed in higher levels compared to OPN and TNFa as well as in symptomatic (with type II and III carotid plaque classification) than asymptomatic (type IV) patients with differences considered statistically significant (P values <0.05). A significant positive correlation between MMP-1 and IL-6 (with Pearson correlation coefficient 0.748) is also notable. The data give further insight into the possible role of specific biomarker and enhance the need for further studies in order to clarify the proper one(s) for detection of the vulnerable plaque and help identify patients at risk for cardiovascular events.

Effect of computer monitor brightness on visual (subjective) carotid plaque characterization

Visual carotid plaque classification on duplex imaging is moderately reproducible, and, although the literature is controversial, research to analyze and overcome this problem remains scarce. The aim of this study was to assess the effect of computer screen brightness on the intraobserver variation of visual plaque classification. One hundred asymptomatic noncalcified carotid plaques causing carotid stenosis >40% on duplex scanning, in 84 patients, were transferred to a personal computer, normalized, and classified into four groups of increasing echogenicity (types 1-4) at 50% and 100% of maximum monitor brightness settings. Baseline intraobserver variability of visual plaque classification at maximum brightness was also assessed. Baseline intraobserver variability was moderate (Cohen's Kappa value 0.59, p < 0.001). Reduction in monitor brightness caused a systemic shift in echogenicity, with 51% of the type 2-4 plaques being classified as more echolucent. Likewise, Cohen's Kappa value decreased to 0.37, indicating poor agreement between the two classification rounds. Reduction in computer screen brightness deteriorates the intraobserver variation of visual plaque classification. This finding could explain some of the discrepancy in reproducibility results reported and should be taken into account when using visual characterization methods.

MRI plaque imaging reveals high-risk carotid plaques especially in diabetic patients irrespective of the degree of stenosis

BackgroundPlaque imaging based on magnetic resonance imaging (MRI) represents a new modality for risk assessment in atherosclerosis. It allows classification of carotid plaques in high-risk and low-risk lesion types (I-VIII). Type 2 diabetes mellitus (DM 2) represents a known risk factor for atherosclerosis, but its specific influence on plaque vulnerability is not fully understood. This study investigates whether MRI-plaque imaging can reveal differences in carotid plaque features of diabetic patients compared to nondiabetics.Methods191 patients with moderate to high-grade carotid artery stenosis were enrolled after written informed consent was obtained. Each patient underwent MRI-plaque imaging using a 1.5-T scanner with phased-array carotid coils. The carotid plaques were classified as lesion types I-VIII according to the MRI-modified AHA criteria. For 36 patients histology data was available.ResultsEleven patients were excluded because of insufficient MR-image quality. DM 2 was diagnosed in 51 patients (28.3%). Concordance between histology and MRI-classification was 91.7% (33/36) and showed a Cohen's kappa value of 0.81 with a 95% CI of 0.98-1.15. MRI-defined high-risk lesion types were overrepresented in diabetic patients (n = 29; 56.8%). Multiple logistic regression analysis revealed association between DM 2 and MRI-defined high-risk lesion types (OR 2.59; 95% CI [1.15-5.81]), independent of the degree of stenosis.ConclusionDM 2 seems to represent a predictor for the development of vulnerable carotid plaques irrespective of the degree of stenosis and other risk factors. MRI-plaque imaging represents a new tool for risk stratification of diabetic patients.See Commentary: http://www.biomedcentral.com/1741-7015/8/78/abstract

0034: Carotid Interpretation Criteria. How to Choose What You Use

The first criteria for estimation of carotid stenosis were developed by the Strandness group at the University of Washington in the 1980s. Others proposed different criteria around the same time. The major carotid surgery trials of the 1990s (NASCET, ECST and ACAS) prompted the development of CDUS criteria to match the 70% and 60% ICA diameter reducing stenosis cut-offs used in these trials. It is to be remembered that these trials used differing methods for estimating stenosis. B mode measurement of stenosis has improved substantially on modern duplex scanners. In 1988 Gray-Weale proposed a classification of carotid plaques on the basis of their echogenicity. Recently interest has returned to plaque characterization with measurement of Gray scale median and other features in an attempt to identify the “vulnerable carotid plaque” and better select asymptomatic patients who might benefit from intervention. The diagnosis of ICA occlusion is important clinically but is not 100% accurate with any of the available imaging modalities. Anatomic information regarding abnormal height of the bifurcation (high or low), the distal extent of the plaque, the diameter of the distal ICA and the presence of tortuosity or kinking of the vessels is also important information and may prompt additional imaging. Care is required in the interpretation of the CDUS and application of standard criteria in a variety of circumstances including contralateral high-grade ICA stenosis or occlusion; elevated CCA velocities and proximal stenosis; tandem ICA stenosis; altered cardiac output states and valvular heart disease, and the presence of intracranial disease. Standard criteria are employed for restenosis follow-up after CEA. Dimensions of the patched bulb should be recorded for serial follow-up. Different criteria with higher cut-off velocities for restenosis and B mode assessment of stent apposition are required following carotid stenting (CAS). A number of consensus panel criteria for carotid stenosis have been developed including the Society of Radiologists in Ultrasound Carotid Stenosis Consensus Conference Criteria (2002) and in Australia, the ASUM criteria (1999). Ideally individual vascular laboratories should validate the criteria they use with the results of alternative imaging modalities and operative pathology. The reality however is that this is increasingly difficult with a reduction in the use of conventional angiography and the dispersion of comparative imaging between CTA, MRA and angiography, each of which has its own inherent limitations for stenosis measurement. This presentation will discuss carotid interpretation criteria with particular emphasis on how the information assists the practicing clinician.

Carotid plaque classification: Defining the certainty with which plaque components can be differentiated

Multicontrast-weighted MRI has the potential to become a powerful tool for assessment of atherosclerotic plaque. However, similarities in MR properties across plaque components limit the certainty with which these components can be differentiated. An understanding of MRI's underlying limitations in distinguishing atherosclerotic plaque components, and optimization of key parameters (including the set of components investigated and contrast weightings used) are required. In this study we analyzed endarterectomy specimens using multicontrast MRI and compared the results with matching histological findings to determine the probability of error, an unbiased measure of the underlying error caused by similarity in the spectral characteristics of components. The total error was >40% when five distinct components were investigated, but this was halved when components with similar functions and intensities were grouped together. When three contrast weightings were used to view plaque, diffusion-weighted imaging (DWI) proved valuable for separating hemorrhage from necrotic core, and "hemorrhage + necrotic" from "loose connective tissue + fibrous tissue." A two-way interaction between contrast weightings and components demonstrated that the value of a contrast can be exploited or marginalized depending on the choice of contrast weightings used.

Resolution and SNR effects on carotid plaque classification

Multicontrast-weighted MRI, which is increasingly being used in combination with automatic classification algorithms, has the potential to become a powerful tool for assessing plaque composition. The current literature, however, does not address the relationship between imaging conditions and segmentation viability well. In this study 13 carotid endarterectomy samples were imaged with a 156-microm in-plane resolution and high signal-to-noise ratio (SNR) using proton density (PD), T1, T2, and diffusion weightings. The maximum likelihood (ML) algorithm was used to classify plaque components, with sets of three contrast weighting intensities used as features. The resolution and SNR of the images were then degraded. Classification accuracy was found to be independent of in-plane resolution between 156 microm and 1250 microm, but dependent on SNR. Accuracy decreased less than 10% for degradation in SNR down to 25% of original values, and decreased sharply thereafter. The robustness of automatic classifiers makes them applicable to a wide range of imaging conditions, including standard in vivo carotid imaging scenarios.

Effect of Image Normalization on Carotid Plaque Classification and the Risk of Ipsilateral Hemispheric Ischemic Events: Results from the Asymptomatic Carotid Stenosis and Risk of Stroke Study

The aim of this study was to determine the effect of image normalization on plaque classification and the risk of ipsilateral ischemic neurologic events in patients with asymptomatic carotid stenosis. The first 1,115 patients recruited to the Asymptomatic Carotid Stenosis and Risk of Stroke (ACSRS) study with a follow-up of 6 to 84 months (mean 37.1 months) were included in this study. Duplex ultrasonography was used for grading the degree of internal carotid artery stenosis and for plaque characterization (types 1-5), which was performed before and after image normalization. One hundred sixteen ipsilateral ischemic hemispheric events occurred. Image normalization resulted in 60% of plaques being reclassified. Before image normalization, a high event rate was associated with all types of plaque. After image normalization, 109 (94%) of the events occurred in patients with plaque types 1 to 3. For patients with European Carotid Stenosis Trial (ECST) 70 to 99% diameter stenosis (equivalent to North American Symptomatic Carotid Endarterectomy Trial [NASCET] 50-99%) with plaque types 1 to 3, the cumulative stroke rate was 14% at 7 years (2% per year), and for patients with plaque types 4 and 5, the cumulative stroke rate was 0.9% at 7 years (0.14% per year). The results suggest that asymptomatic patients with plaque types 4 and 5 classified as such after image normalization are at low risk irrespective of the degree of stenosis.

Effect of Image Normalization on Carotid Plaque Classification and the Risk of Ipsilateral Hemispheric Ischemic Events: Results from the Asymptomatic Carotid Stenosis and Risk of Stroke Study

Effect of Image Normalization on Carotid Plaque Classification and the Risk of Ipsilateral Hemispheric Ischemic Events: Results from the Asymptomatic Carotid Stenosis and Risk of Stroke Study