Carotid Ultrasound Images Research Articles

Creating a Simulated Dataset for Training Deep Convolutional Neural Networks for Use in Cardiovascular Photoacoustic Tomography

Background/Objective: Photoacoustic tomography possesses increasing potential as a non-invasive imaging method that combines optical and acoustic imaging to maximize the visualization of tissue. Determining the composition, orientation, and location of anatomical structures in multidimensional space requires maximizing image resolution and differentiation from noise and reflection artifacts. Using simulations to develop and improve methods for image resolution allows for flexibility and variation of numerous variables. 
 
 Methods: Binary masks were created from mouse common carotid ultrasound images using a graphical user interface for MATLAB. With the k-Wave toolbox, we performed time-reversal photoacoustic simulations using the masks. Medium properties for the simulations were assigned for sound speed and density for connective tissue (1540 m/s, 1027 kg/m3) and arterial walls (1569 m/s, 1102 kg/m3). The dataset was augmented through rotational and mirrored transformations and the addition of noise and reflection artifacts via Python open-source software. 
 
 Results: A set of 87 binary masks was generated from common carotid ultrasound images. These masks were used to simulate initial pressure distributions through the k-Wave toolbox to reconstruct the structure of the common carotid. Each simulation yielded graphs for initial pressure and sensor distribution, simulated sensor data, reconstructed initial pressure, and a comparison profile between the original and reconstructed pressure. Data augmentation was implemented using the reconstructed pressure output from the 87 simulations, each producing 12 distinct images from rotations and mirroring with the addition of noise and reflection artifacts. The final dataset yielded 1044 images. 
 
 Conclusion and Potential Impact: Future work will involve applying this dataset to a neural network to improve photoacoustic quality such that transfer learning can be applied on ex vivo and in vivo datasets. Thus, there is potential for use in diagnostic applications in patients with cardiovascular disease states like atherosclerosis and aneurysms that require high resolution visualization of tissue structure and composition.

Traffic-related NOx exposure and proximity to freeways are related to adverse changes in subclinical atherosclerosis measures from childhood to adulthood

Background: Chronic exposure to air pollutants is associated with increased risk of cardiovascular disease (CVD) among adults. However, little is known about how air pollution may affect the development of subclinical atherosclerosis in younger populations. Subclinical measures of atherosclerosis, including carotid artery intima-media thickness (CIMT) and arterial stiffness (CAS), may provide insight into early CVD pathogenesis. Methods: In a pilot study of 70 participants from the Southern California Children’s Health Study, we investigated subclinical atherosclerosis progression from childhood to adulthood. Using carotid artery ultrasound images obtained at age 10 and follow-up images from age 21-22, we examined associations between traffic-related pollutants and changes in CIMT and CAS. Lifetime average traffic-related NOx exposures were assigned using the CALINE4 (California Line Source) dispersion model. Distance to nearest roadways by class was assigned based on ESRI Streetmap Premium roadway data. Results: For each interquartile range increase in average lifetime NOx exposure, we observed a 12.12μm (95% CI: 1.11-23.13; p=0.03) increase in early adulthood attained CIMT and greater longitudinal changes in CIMT from childhood to adulthood (11.5μm, 95% CI: 1.29-21.73, p=0.03) over the follow-up period. Similar adverse changes in CIMT were observed with distance to nearest freeway. Changes in CAS were related to distance to freeway, such that living in closer proximity to freeways was related to greater adverse change in CAS over the follow-up period (stiffness index beta: 0.85; 95% CI: 1.50, 0.21, p=0.01; distensibility: -5.61; 95% CI: -0.93, -10.27, p=0.02).Conclusions: We observed greater adverse changes measures of carotid artery wall thickness and stiffness over time in relation to lifetime traffic-related NOx exposure and residential proximity to freeways. To our knowledge, this is the first study to evaluate the longitudinal impact of lifetime traffic-related air pollution exposure on changes in subclinical markers of atherosclerosis during the critical transition from childhood to adulthood.

Fully Automatic Measurement of Intima-Media Thickness in Ultrasound Images of the Common Carotid Artery Based on Improved Otsu's Method and Adaptive Wind Driven Optimization.

The intima media thickness (IMT) of the common carotid artery (CCA) can be used to predict the risk of atherosclerosis. Many image segmentation techniques have been used for IMT measurement. However, severe noise in the ultrasound image can lead to erroneous segmentation results. To improve the robustness to noise, a fully automatic method, based on an improved Otsu's method and an adaptive wind-driven optimization technique, is proposed for estimating the IMT (denoted as "improved Otsu-AWDO"). First, an advanced despeckling filter, i.e., " Nagare's filter" is used to address the speckle noise in the carotid ultrasound images. Next, an improved fuzzy contrast method (IFC) is used to enhance the region of the intima media complex (IMC) in the blurred filtered images. Then, a new method is used for automatic extraction of the region of interest (ROI). Finally, the lumen intima interface and media adventitia interface are segmented from the IMC using improved Otsu-AWDO. Then, 156 B-mode longitudinal carotid ultrasound images of six different datasets are used to evaluate the performance of the automatic measurements. The results indicate that the absolute error of proposed method is only 10.1 ± 9.6 (mean ± std in μm). Moreover, the proposed method has a correlation coefficient as high as 0.9922, and a bias as low as 0.0007. From comparison with previous methods, we can conclude that the proposed method has strong robustness and can provide accurate IMT estimations.

Do Carotid Plaque Ulcers Heal? Potential Detection of Carotid Artery Plaque Healing by Carotid Ultrasound Imaging.

Carotid plaque ulcers confer an increased risk for stroke/ transient ischemic attacks in both symptomatic and asymptomatic carotid artery stenosis. Little is known about the healing rates of ulcers or the development of new ulcers. Carotid Duplex studies are noninvasive and easily repeatable tests to monitor progression of carotid stenosis and plaque morphology. Our aim was to determine the prevalence and healing rates of ultrasound-detected carotid plaque ulcers. We retrospectively reviewed 5837 carotid Duplex studies performed in an outpatient ultrasound laboratory affiliated with the neurological department of an academic center. A total of 3215 patients underwent a first carotid ultrasound Duplex study, and 2622 follow-up studies were done. Carotid ulcer was defined as a 2 mm deep surface indentation in a carotid plaque with a well-defined back wall, as determined by multimodal ultrasound imaging techniques. The prevalence of carotid plaque ulcers among the 3215 patients with a first ultrasound study was 3% (82/3215). The mean follow-up was 42 ± 30 months, and the median number of follow-up studies was 6. Among patients with ulcers, follow-up studies were available in 65/82 patients. During the follow-up period, 28/65 (43%) ulcers healed. Among all 2622 follow-up studies, 45 patients developed a new ulcer. Duplex-detected carotid plaque ulcer prevalence is low. The carotid ulcers healed in approximately half of patients during follow-up. Factors associated with ulcer healing and development remain poorly understood.

Performance evaluation of unpreprocessed and pre-processed ultrasound images of carotid artery using CNN algorithm

Detection of plaques is an important task especially those which are prone to rupture and may dislocate to some other body parts. Also early detection reduces the risk of cardiac and cerebrovascular anomalies. Due to its wide availability and low cost, ultrasound images of carotid artery has the ability and potential to gain the preference over other resources for plaque detection and analysis in medical practice. However, the difficulty caused in automated techniques to identify plaques is significantly due to image noise, plaque size and the complex appearance of tissues comprising a plaque. So, in this paper, we have addressed this problem by using deep learning techniques such as CNN algorithm. Here we will build a CNN (Convolutional Neural Network) that will extract features from the dataset of images thereby giving detailed information which will help the clinicians to identify the abnormalities and constituents of different plaques in an image and report the image as “normal” or “abnormal”. In this paper we have used approximately 1000 images (JPG format) of 100 cases to process and has validated the proposed convolutional neural network (CNN). The process of cross-validation with the clinical assessment showed a correlation of about 0.75 for the detection of plaque. This results indicate the potential of deep learning techniques in medical fields, here, automatic detection of anomalies in carotid ultrasound images.

Segmentation of 3D ultrasound carotid vessel wall using U-Net and segmentation average network.

Segmentation of carotid vessel wall is required in vessel wall volume (VWV) and local vessel-wall-plus-plaque thickness (VWT) quantification of the carotid artery. Manual segmentation of the vessel wall is time-consuming and prone to interobserver variability. In this paper, we proposed a convolutional neural network (CNN) to segment the common carotid artery (CCA) from 3D carotid ultrasound images. The proposed CNN involves three U-Nets that segmented the 3D ultrasound (3DUS) images in the axial, lateral and frontal orientations. The segmentation maps generated by three U-Nets were consolidated by a novel segmentation average network (SAN) we proposed in this paper. The experimental results show that the proposed CNN improved the segmentation accuracies. Compared to only using U-Net alone, the proposed CNN improved the Dice similarity coefficient (DSC) for vessel wall segmentation from 64.8% to 67.5%, the sensitivity from 63.8% to 70.5%, and the area under receiver operator characteristic curve (AUC) from 0.89 to 0.94.

Cardiovascular/stroke risk prevention: A new machine learning framework integrating carotid ultrasound image-based phenotypes and its harmonics with conventional risk factors

MotivationMachine learning (ML)-based stroke risk stratification systems have typically focused on conventional risk factors (CRF) (AtheroRisk-conventional). Besides CRF, carotid ultrasound image phenotypes (CUSIP) have shown to be powerful phenotypes risk stratification. This is the first ML study of its kind that integrates CUSIP and CRF for risk stratification (AtheroRisk-integrated) and compares against AtheroRisk-conventional. MethodsTwo types of ML-based setups called (i) AtheroRisk-integrated and (ii) AtheroRisk-conventional were developed using random forest (RF) classifiers. AtheroRisk-conventional uses a feature set of 13 CRF such as age, gender, hemoglobin A1c, fasting blood sugar, low-density lipoprotein, and high-density lipoprotein (HDL) cholesterol, total cholesterol (TC), a ratio of TC and HDL, hypertension, smoking, family history, triglyceride, and ultrasound-based carotid plaque score. AtheroRisk-integrated system uses the feature set of 38 features with a combination of 13 CRF and 25 CUSIP features (6 types of current CUSIP, 6 types of 10-year CUSIP, 12 types of quadratic CUSIP (harmonics), and age-adjusted grayscale median). Logistic regression approach was used to select the significant features on which the RF classifier was trained. The performance of both ML systems was evaluated by area-under-the-curve (AUC) statistics computed using a leave-one-out cross-validation protocol. ResultsLeft and right common carotid arteries of 202 Japanese patients were retrospectively examined to obtain 404 ultrasound scans. RF classifier showed higher improvement in AUC (~57%) for leave-one-out cross-validation protocol. Using RF classifier, AUC statistics for AtheroRisk-integrated system was higher (AUC = 0.99,p-value<0.001) compared to AtheroRisk-conventional (AUC = 0.63,p-value<0.001). ConclusionThe AtheroRisk-integrated ML system outperforms the AtheroRisk-conventional ML system using RF classifier.

Two-stage artificial intelligence model for jointly measurement of atherosclerotic wall thickness and plaque burden in carotid ultrasound: A screening tool for cardiovascular/stroke risk assessment

MotivationThe early screening of cardiovascular diseases (CVD) can lead to effective treatment. Thus, accurate and reliable atherosclerotic carotid wall detection and plaque measurements are crucial. Current measurement methods are time-consuming and do not utilize the power of knowledge-based paradigms such as artificial intelligence (AI). We present an AI-based methodology for the joint automated detection and measurement of wall thickness and carotid plaque (CP) in the form of carotid intima-media thickness (cIMT) and total plaque area (TPA), a class of AtheroEdge™ system (AtheroPoint™, CA, USA). MethodThe novel system consists of two stages, and each stage comprises an independent deep learning (DL) model. In Stage I, the first DL model segregates the common carotid artery (CCA) patches from ultrasound (US) images into the rectangular wall and non-wall patches. The characterized wall patches are integrated to form the region of interest (ROI), which is then fed into Stage II. In Stage II, the second DL model segments the far wall region. Lumen-intima (LI) and media-adventitial (MA) boundaries are then extracted from the wall region, which is then used for cIMT and PA measurement. ResultsUsing the database of 250 carotid scans, the cIMT error using the AI model is 0.0935±0.0637 mm, which is lower than those of all previous methods. The PA error is found to be 2.7939±2.3702 mm2. The system's correlation coefficient (CC) between AI and ground truth (GT) values for cIMT is 0.99 (p < 0.0001), which is higher compared with the CC of 0.96 (p < 0.0001) shown by the earlier DL method. The CC for PA between AI and GT values is 0.89 (p < 0.0001). ConclusionA novel AI-based strategy was applied to carotid US images for the joint detection of carotid wall thickness (cWT) and plaque area (PA), followed by cIMT and PA measurement. This AI-based strategy shows improved performance using the patch technique compared with previous methods using full carotid scans.

The Circulating Fatty Acid Transporter Soluble CD36 Is Not Associated with Carotid Atherosclerosis in Subjects with Type 1 and Type 2 Diabetes Mellitus

This study aimed to determine the association of fatty acid transporter plasma soluble cluster of differentiation 36 (sCD36) with subclinical carotid atherosclerosis (SCA). A cross-sectional study was conducted in 1023 subjects, 225 with type 1 diabetes (T1D), 276 with type 2 diabetes (T2D) and 522 who were nondiabetic. Carotid atherosclerotic plaque (CAP) presence was determined using B-mode carotid ultrasound imaging. sCD36 were analysed by ELISA, and CD36 surface receptor and mRNA expression were measured by flow cytometry and real-time PCR. Logistic regression models were used to evaluate sCD36 as a biomarker of SCA. Up to 376 (36.75%) participants had at least one CAP, 76 T1D, 164 T2D and 136 without diabetes, while the remaining 647 (63.25%) did not have any CAP. There were no differences in sCD36 between patients with and without CAP in T1D (p = 0.287) or T2D (p = 0.513). Although nondiabetic subjects with plaques had lower sCD36 levels than those without (p = 0.023), the multivariate models revealed no association of sCD36 with CAP in any of the three study groups. No differences were found in surface CD36 or CD36 mRNA expression between the patients with and without CAP. sCD36 is not associated with SCA in type 1 or type 2 diabetic or in nondiabetic subjects.

Carotid Plaque Strain Indices Were Correlated With Cognitive Performance in a Cohort With Advanced Atherosclerosis, and Traditional Doppler Measures Showed no Association.

Traditional Doppler measures have been used to predict cognitive performance in patients with carotid atherosclerosis. Novel measures, such as carotid plaque strain indices (CPSIs), have shown associations with cognitive performance. We hypothesized that lower mean middle cerebral artery (MCA) velocities, higher bulb-internal carotid artery (ICA) velocities, the MCA pulsatility index (PI), and CPSIs would be associated with poorer cognitive performance in individuals with advanced atherosclerosis. Neurocognitive testing, carotid ultrasound imaging, transcranial Doppler imaging, and carotid strain imaging were performed on 40 patients scheduled for carotid endarterectomy. Kendall tau correlations were used to examine relationships between cognitive tests and the surgical-side maximum peak systolic velocity (PSV; from the bulb, proximal, mid, or distal ICA), mean MCA velocity and PI, and maximum CPSIs (axial, lateral, and shear strain indices used to characterize plaque deformations with arterial pulsation). Cognitive measures included age-adjusted indices of verbal fluency, verbal and visual learning/memory, psychomotor speed, auditory attention/working memory, visuospatial construction, and mental flexibility. Participants had a median age of 71.0 (interquartile range, 9.75) years; 26 were male (65%), and 14 were female (35%). Traditional Doppler parameters, PSV, mean MCA velocity, and MCA PI did not predict cognitive performance (all P > .05). Maximum CPSIs were significantly associated with cognitive performance (P < .05). Traditional velocity measurements of the maximum bulb-ICA PSV, mean MCA velocity, and PI were not associated with cognitive performance in patients with advanced atherosclerotic disease; however, maximum CPSIs were associated with cognitive performance. These findings suggest that cognition may be associated with unstable plaque rather than blood flow.

Comparison of the Diagnostic Performances of Ultrasound, High-Resolution Magnetic Resonance Imaging, and Positron Emission Tomography/Computed Tomography in a Rabbit Carotid Vulnerable Plaque Atherosclerosis Model.

Our study aimed to evaluate the diagnostic performances of 3 routine examination methods for cerebrovascular disease in a rabbit carotid artery atherosclerosis model. A total of 12 New Zealand rabbits were included: 4 in a control group and 8 in an experimental group. A clinically relevant atherosclerosis rabbit model was induced by left common carotid artery ligation and a 12-week high-fat diet. Atherosclerosis was further confirmed by a histopathologic analysis. Then carotid ultrasound (US) imaging, high-resolution magnetic resonance imaging (HRMRI), and positron emission tomography (PET)/computed tomography (CT) were performed on this model to evaluate the diagnostic performances. Carotid US showed plaque formation in the left common carotid artery and little plaque in the right common carotid artery in the experimental group. In addition, HRMRI showed stenosis formation in the left common carotid artery in the experimental group. At the horizontal level, plaques were found in the left common carotid artery, and no plaques were found in the right common carotid artery in the experimental group. Also, PET/CT showed local hypermetabolism and vulnerable plaques in the left common carotid artery of the experimental group, whereas no hypermetabolism was found in the right common carotid artery of the experimental group. Moreover, the soft plaques detected by carotid US were different from the vulnerable plaques detected by PET/CT. The unstable plaques on HRMRI were the same as the hypermetabolic vulnerable plaques on PET/CT. High-resolution MRI is recommended for the evaluation of neck and intracranial vascular stenosis and plaque properties in patients with stroke.

Carotid artery ultrasound image analysis: A review of the literature.

Stroke is one of the prominent causes of death in the recent days. The existence of susceptible plaque in the carotid artery can be used in ascertaining the possibilities of cardiovascular diseases and long-term disabilities. The imaging modality used for early screening of the disease is B-mode ultrasound image of the person in the artery area. The objective of this article is to give a widespread review of the imaging modes and methods used for studying the carotid artery for identifying stroke, atherosclerosis and related cardiovascular diseases. We encompass the review in methods used for artery wall tracking, intima-media, and lumen segmentation which will help in finding the extent of the disease. Due to the characteristics of the imaging modality used, the images have speckle noise which worsens the image quality. Adaptive homomorphic filtering with wavelet and contourlet transforms, Levy Shrink, gamma distribution were used for image denoising. Learning-based neural network approaches for denoising give better edge preservation. Domain knowledge-based segmentation approaches have proved to provide more accurate intima-media thickness measurements. There is a requirement of useful fully automatic segmentation approaches, 3D, 4D systems, and plaque motion analysis. Taking into consideration the image priors like geometry, imaging physics, intensity and temporal data, image analysis has to be performed. Encouragingly more research has focused on content-specific segmentation and classification techniques. With the evaluation of machine learning algorithms, classifying the image as with or without a fat deposit has gained better accuracy and sensitivity. Machine learning-based approaches like self-organizing map, k-nearest neighborhood and support vector machine achieve promising accuracy and sensitivity in classification. The literature reveals that there is more scope in identifying a patient-specific model in a fully automatic manner.

Bimodal Automated Carotid Ultrasound Segmentation Using Geometrically Constrained Deep Neural Networks

For asymptomatic patients suffering from carotid stenosis, the assessment of plaque morphology is an important clinical task which allows monitoring of the risk of plaque rupture and future incidents of stroke. Ultrasound Imaging provides a safe and non-invasive modality for this, and the segmentation of media-adventitia boundaries and lumen-intima boundaries of the Carotid artery form an essential part in this monitoring process. In this paper, we propose a novel Deep Neural Network as a fully automated segmentation tool, and its application in delineating both the media-adventitia boundary and the lumen-intima boundary. We develop a new geometrically constrained objective function as part of the Network's Stochastic Gradient Descent optimisation, thus tuning it to the problem at hand. Furthermore, we also apply a bimodal fusion of amplitude and phase congruency data proposed by us in previous work, as an input to the network, as the latter provides an intensity-invariant data source to the network. We finally report the segmentation performance of the network on transverse sections of the carotid. Tests are carried out on an augmented dataset of 81,000 images, and the results are compared to other studies by reporting the DICE coefficient of similarity, modified Hausdorff Distance, sensitivity and specificity. Our proposed modification is shown to yield improved results on the standard network over this larger dataset, with the advantage of it being fully automated. We conclude that Deep Neural Networks provide a reliable trained manner in which carotid ultrasound images may be automatically segmented, using amplitude data and intensity invariant phase congruency maps as a data source.

P832 ESCAPE Stroke: first results

Abstract Objective "ESCAPE Stroke" (Enhancement, Segmentation and Characterization of Atheroma PlaquE for Stroke Prevention) is a multidisciplinary project that aims to develop a new software tool to obtain maximum information about carotid atherosclerosis, through the (semi)automatic analysis of the morphological characteristics of the carotid plaque. We present the first results for evaluating this software. Methods For the morphological analysis of the characteristics of the atheroma plaques, a database of 117 carotid ultrasound images was used. The extracted morphological features refer to the degree of echogenicity, texture and the surface of the atheroma plaque. Regarding to echogenicity of the atheroma plaques, the implemented software algorithm divides the plaques into four classes as follows: • Anechogenic A type, the echogenic areas representing less than 25% of the plaque. The 117 plaques subjected to computerized analysis were classified as: 54 Type E plaques, 57 Type E / 2 plaques, 6 Type A / 2 plaques and no Type A plaques • A / 2 type: predominantly anechogenic, anechogenic areas accounting for more than 50% of the plaque but less than 75%; • Type E / 2: predominantly echogenic, echogenic areas accounting for more than 50% of the plaque but less than 75%; • Type E: echogenic, an-echogenic areas representing less than 25% of the plaque; (Figure 1) On the subject of texture of atheroma plaques, the implemented software algorithm studies the homogeneity. In order to quantify the degree of homogeneity of the atheroma plaques, from the Spatial Gray Level Dependency Matrix, a number of functions have been calculated that have become the homogeneity-classification parameters. The 117 plaques were divided into three different classes (Figure 1) as follows: • 28 plaques type 1, with uniform consistency regardless of their echogenicity; • 51 plaques type 2, with an average non-uniform consistency, the distribution of gray levels in the atheroma plaque varies but does not go through the entire interval from hypo-echogenic zones to hyper-echogenic zones; • 38 plaques type 3, with high non-uniform consistency, with hypo-echogenic and hyper-echogenic zones. With regard to the surface of the atheroma plaques, it is defined as smooth and regular, slightly irregular or ulcerated (Figure 1). Using different contour parameters, the 117 atheroma plaques subjected to computerized analysis were classified as follows: • 48 plaques with smooth and regular surface; • 55 plaques with slightly irregular surface; • 14 plaques with ulcerative surface; The overall agreement between software analysis and "eyeball" assessment made by 2 independent physicians was 69% for echogenicity, 58% for homogeneity and 74% for plaque surface evaluation. Conclusion Our automatic/semiautomatic method for plaque characterization could add important information in atheroma plaque assessment and further research is needed in order to confirm its value. Abstract P832 Figure1

Prevalence of Thyroid Incidentalomas from 1995 to 2016: A Single-Center, Retrospective Cohort Study.

Incidental detection of thyroid nodules on nonthyroid imaging may contribute to increased diagnosis of thyroid cancer. We investigated the prevalence of thyroid incidentalomas across imaging modalities among a predominately male veteran population. Thyroid nodules were identified on nonthyroid-directed radiology reports using natural language processing. All reports from 1995 to 2016 for chest computed tomography (CT), carotid ultrasound (US), and neck magnetic resonance imaging (MRI) were reviewed. Individuals with multiple studies were included at their initial study and duplicates removed. A total of 25 763 carotid US, 23 526 chest CTs with contrast, 39 262 noncontrast chest CTs, and 9503 MRIs were reviewed. With duplicates removed, 14 642 carotid US, 12 923 chest CTs with contrast, 17 416 noncontrast chest CTs, and 6926 MRIs were included. Mean age was 66.2 years and 1834 were female (3.53%). Thyroid nodules were reported on 0.84% carotid US, 3.45% MRIs, 5.84% chest CTs with contrast, and 5.14% noncontrast chest CTs. Women had a higher rate of thyroid nodules on MRI (6.46% vs 3.20%, P = .003), chest CT with contrast (9.80% vs 5.72%, P = .007), and noncontrast chest CT (8.77% vs 5.02%, P = .002), but not on carotid US (1.99% vs 0.81%, P = .12). Incidentaloma prevalence increased with age on MRI, chest CT with or without contrast, but not on carotid US, and were more commonly reported from 2007 to 2016 compared to before 2007 across all modalities. Thyroid incidentalomas are commonly reported, are more common among women, and increase with age. The rate of reported incidental thyroid nodules is increasing, likely contributing to the increase in thyroid cancer.

Classification of Carotid Artery Intima Media Thickness Ultrasound Images with Deep Learning.

Cerebrovascular accident due to carotid artery disease is the most common cause of death in developed countries following heart disease and cancer. For a reliable early detection of atherosclerosis, Intima Media Thickness (IMT) measurement and classification are important. A new method for decision support purpose for the classification of IMT was proposed in this study. Ultrasound images are used for IMT measurements. Images are classified and evaluated by experts. This is a manual procedure, so it causes subjectivity and variability in the IMT classification. Instead, this article proposes a methodology based on artificial intelligence methods for IMT classification. For this purpose, a deep learning strategy with multiple hidden layers has been developed. In order to create the proposed model, convolutional neural network algorithm, which is frequently used in image classification problems, is used. 501 ultrasound images from 153 patients were used to test the model. The images are classified by two specialists, then the model is trained and tested on the images, and the results are explained. The deep learning model in the study achieved an accuracy of 89.1% in the IMT classification with 89% sensitivity and 88% specificity. Thus, the assessments in this paper have shown that this methodology performs reasonable results for IMT classification.

CARDIOVASCULAR RISK FACTORS IN SYSTEMIC LUPUS ERYTHEMATOSUS WITH HYPERTENSION

Objective: Systemic lupus erythematosus (SLE) is a systemic disease that affects the young women with predilection. It is associated with a high cardiovascular risk related to atherosclerosis accelerated by different mechanisms, thus increasing morbidity and mortality during SLE. Design and method: This is a retrospective study of the 67 patients with SLE hospitalized at least once in the internal medicine department of Sahloul Hospital in Sousse between January 1st, 2011 and January 1st, 2017. Results: Twenty-four patients were found with hypertension: 22 women and 2 men. The occurrence of hypertension was on average 5.4 [0 - 21] years after of the diagnosis of SLE. The plausible etiology was attributable to SLE in 14 cases, corticoid-induced in 17 cases and secondary to renal failure in 7 cases. SLE has been evolving in our patients for an average of 12 years [3–28]. The occurrence after the age of 40 was noted in seven cases. Lupus nephropathy was present in 15 cases. Treatment was based on long-term corticosteroid therapy (n = 24), synthetic anti-malarials (n = 20), immuno-suppressants (n = 18), and biotherapy (n = 2). Eight patients were taking non-steroidal anti-inflammatory drugs for at least 3 years. Several other cardiovascular risk factors were found in this hypertensive population. Two patients were smoking. Seven were diabetic with an average evolution of 7.3 years. Obesity was noted in 12 patients. Dyslipidemia was found in 9 cases and hyper-homocysteinemia in 6 cases. A sedentary lifestyle was reported in 23 cases. Chronic renal failure was found in 7 patients. The number of cardiovascular risk factors was 4.9 [2 - 8] per patient. Ten patients had atherosclerosis. In 4 of them, it was documented by a carotid ultrasound imaging. The remaining six had ischemic heart disease (n = 5) and / or cerebrovascular accident (n = 3). Conclusions: SLE is associated with accelerated atherosclerosis and a high risk of developing cardiovascular disease in young patients, especially when associated with hypertension. A better understanding of the pathophysiology of hypertension during SLE would improve the prognosis.

Cardinal Points Symmetry Landmarks Distribution Model for Segmentation of Region of Interest in Simulated B-Mode Ultrasound Carotid Artery Images

Measurement accuracy and understanding of geometry of Common Carotid Artery (CCA) play an important role in carotid atherosclerosis assessment and management. In order to device techniques that can accurately analyze the effects of plaques on the carotid artery, a model that can sufficiently segment the Region of Interest (ROI) in the B-mode ultrasound image of carotid artery is needed. In this paper, a new Cardinal Points Symmetry Landmarks Distribution Model (CPS-LDM) to sufficiently segment the ROI in the carotid artery B-mode ultrasound imaged in the transverse plane is developed. The developed model employs a combination of fixed landmarks (FLs) and movable landmarks (MLs) to obtain the total landmarks (TLs) that can sufficiently segment the shape of the ROI of the carotid artery. Simulated ultrasound images are used. Four FLs are fixed on each of the four ROIs of the simulated carotid artery determined by the cardinal points North (N), South (S), East (E) and West (W) drawn on the ROIs of the carotid artery. The MLs are determined by the inter-cardinal directions such as North-East (NE), North-West (NW) and so on. The CPS-LDM equation developed allows us to visualize graphically the optimum number of points that can sufficiently segment the ROIs. ImageJ2 software was used to generate the Cartesian coordinates for each landmark which were then used to generate the Shape Space Pattern (SSP) of the carotid artery ready for further statistical analysis. The results showed that the CPS-LD model is sufficiently generic and adaptable to a variety of carotid artery B-mode ultrasound image simulated under different scenarios

Cardinal Points Symmetry Landmarks Distribution Model for Segmentation of Region of Interest in Simulated B-Mode Ultrasound Carotid Artery Images

Measurement accuracy and understanding of geometry of Common Carotid Artery (CCA) play an important role in carotid atherosclerosis assessment and management. In order to device techniques that can accurately analyze the effects of plaques on the carotid artery, a model that can sufficiently segment the Region of Interest (ROI) in the B-mode ultrasound image of carotid artery is needed. In this paper, a new Cardinal Points Symmetry Landmarks Distribution Model (CPS-LDM) to sufficiently segment the ROI in the carotid artery B-mode ultrasound imaged in the transverse plane is developed. The developed model employs a combination of fixed landmarks (FLs) and movable landmarks (MLs) to obtain the total landmarks (TLs) that can sufficiently segment the shape of the ROI of the carotid artery. Simulated ultrasound images are used. Four FLs are fixed on each of the four ROIs of the simulated carotid artery determined by the cardinal points North (N), South (S), East (E) and West (W) drawn on the ROIs of the carotid artery. The MLs are determined by the inter-cardinal directions such as North-East (NE), North-West (NW) and so on. The CPS-LDM equation developed allows us to visualize graphically the optimum number of points that can sufficiently segment the ROIs. ImageJ2 software was used to generate the Cartesian coordinates for each landmark which were then used to generate the Shape Space Pattern (SSP) of the carotid artery ready for further statistical analysis. The results showed that the CPS-LD model is sufficiently generic and adaptable to a variety of carotid artery B-mode ultrasound image simulated under different scenarios

Long-Term Exposure to Ambient Ozone and Progression of Subclinical Arterial Disease: The Multi-Ethnic Study of Atherosclerosis and Air Pollution.

Background:Long-term ozone () exposure is associated with cardiovascular mortality, but little is known about the associations between and subclinical arterial disease.Objectives:We studied the longitudinal association of exposure to and progression of key subclinical arterial markers in adults: intima-media thickness of common carotid artery (), carotid plaque (CP) burden, and coronary artery calcification (CAC).Methods:CAC was measured one to four times at baseline and at follow-up exams (1999–2012) by computed tomography (CT) in 6,619 healthy adults, recruited at age 45–84 y without cardiovascular disease (CVD), over a mean of 6.5 y (standard deviation: 3.5 y). and CP burden were quantified in 3,392 participants using carotid artery ultrasound imaging acquired over a mean of 9 y (1.7 y). Over 91% and 89% participants had at least one follow-up and CAC measurement, respectively. Residence-specific concentrations were estimated by a validated spatiotemporal model spanning from 1999 to 2012. This model relied on comprehensive monitoring data and geographical variables to predict individualized long-term average concentrations since baseline. Linear mixed models and logistic regression model were used to evaluate relationships of long-term average exposure to with longitudinal change in , CAC, and CP formation, respectively.Results:Mean progression rates of and CAC were and . CP formation was identified in 55% of the subjects. A increase in long-term average exposure was associated with a [95% confidence interval (CI): 1.4, 9.7] greater increase in over 10 y. A increase in was also associated with new CP formation [odds ratio (OR): 1.2 (95% CI: 1.1, 1.4)] but not CAC progression [ (95% CI: , 2)]. Associations were robust in the analysis with extended covariate adjustment, including copollutants, i.e., nitrogen oxides () and particulate matter with diameter ().Conclusion:Over almost a decade of follow-up, outdoor concentrations were associated with increased rate of carotid wall thickness progression and risk of new plaque formation, suggesting arterial injury in this cohort. https://doi.org/10.1289/EHP3325