Polyvascular Subclinical Atherosclerosis: Correlation Between Ankle Brachial Index and Carotid Atherosclerosis in a Population-Based Sample.

We assessed the correlation between the biomarkers of lower limb atherosclerosis (eg, ankle-brachial index [ABI]) and of carotid atherosclerosis (eg, common carotid intima-media thickness (IMT) and presence of atherosclerotic plaque) in a population-based cohort from Girona (Northwest Spain) recruited in 2010. Ankle-brachial index and carotid ultrasound were performed in all participants. Generalized additive multivariable models were used to adjust a regression model of common carotid IMT on ABI. Logistic regression multivariable models were adjusted to assess the probability of carotid plaque in individuals with peripheral artery disease. We included 3307 individuals (54.2% women), mean age 60 years (standard deviation 11). Two patterns of association were observed between subclinical biomarkers of atherosclerosis at the lower limb and carotid artery. Ankle-brachial index and common carotid IMT showed a linear trend in men [beta coefficient (95% confidence interval) =-.068 (-.123; -.012); P = .016]. Women with peripheral artery disease presented with high risk of atherosclerotic plaque at the carotid artery [Odds ratio (95% confidence interval) = 2.61, (1.46; 4.69); P = .001]. Men showed a significant linear association between ABI levels and common carotid IMT values. Women with peripheral artery disease presented with high risk of atherosclerotic plaque at the carotid artery.

Eye-LRCN: A Long-Term Recurrent Convolutional Network for Eye Blink Completeness Detection.

Computer vision syndrome causes vision problems and discomfort mainly due to dry eye. Several studies show that dry eye in computer users is caused by a reduction in the blink rate and an increase in the prevalence of incomplete blinks. In this context, this article introduces Eye-LRCN, a new eye blink detection method that also evaluates the completeness of the blink. The method is based on a long-term recurrent convolutional network (LRCN), which combines a convolutional neural network (CNN) for feature extraction with a bidirectional recurrent neural network that performs sequence learning and classifies the blinks. A Siamese architecture is used during CNN training to overcome the high-class imbalance present in blink detection and the limited amount of data available to train blink detection models. The method was evaluated on three different tasks: blink detection, blink completeness detection, and eye state detection. We report superior performance to the state-of-the-art methods in blink detection and blink completeness detection, and remarkable results in eye state detection.

An end-to-end framework for intima media measurement and atherosclerotic plaque detection in the carotid artery.

BACKGROUND AND OBJECTIVES: The detection and delineation of atherosclerotic plaque are usually manually performed by medical experts on the carotid artery. Evidence suggests that this manual process is subject to errors and has a large variability between experts, equipment, and datasets. This paper proposes a robust end-to-end framework for automatic atherosclerotic plaque detection. METHODS: The proposed framework is composed of: (1) a semantic segmentation model based on U-Net, with EfficientNet as the backbone, that obtains a segmentation mask with the carotid intima-media region; and (2) a convolutional neural network designed using Bayesian optimization that simultaneously performs a regression to get the average and maximum carotid intima media thickness, and a classification to determine the presence of plaque. RESULTS: Our approach improves the state-of-the-art in both co and bulb territories in the REGICOR database, with more than 8000 images, while providing predictions in real-time. The correlation coefficient was 0.89 in the common carotid artery and 0.74 for bulb region, and the F1 score for atherosclerotic plaque detecting was 0.60 and 0.59, respectively. The experimentation carried out includes a comparison with other fully automatic methods for carotid intima media thickness estimation found in the literature. Additionally, we present an extensive experimental study to evaluate the robustness of our proposal, as well as its suitability and efficiency compared to different versions of the framework. CONCLUSIONS: The proposed end-to-end framework significantly improves the automatic characterization of atherosclerotic plaque. The generation of the segmented mask can be helpful for practitioners since it allows them to evaluate and interpret the model's results by visual inspection. Furthermore, the proposed framework overcomes the limitations of previous research based on ad-hoc post-processing, which could lead to overestimations in the case of oblique forms of the carotid artery.

Do individuals with autoimmune disease have increased risk of subclinical carotid atherosclerosis and stiffness?

To explore the role of chronic inflammation inherent to autoimmune diseases in the development of subclinical atherosclerosis and arterial stiffness, this study recruited two population-based samples of individuals with and without autoimmune disease (ratio 1:5) matched by age, sex, and education level and with a longstanding (≥6 years) diagnosis of autoimmune disease. Common carotid intima-media thickness (IMT) and arterial distensibility and compliance were assessed with carotid ultrasound. Multivariable linear and logistic regression models were adjusted for 10-year cardiovascular risk. In total, 546 individuals with and without autoimmune diseases (91 and 455, respectively) were included. The mean age was 66 years (standard deviation 12), and 240 (43.9%) were women. Arterial stiffness did not differ according to the presence of autoimmune diseases. In men, the diagnosis of autoimmune diseases significantly increased common carotid IMT [beta-coefficient (95% confidence interval): 0.058 (0.009; 0.108); p value = 0.022] and the percentage with IMT ≥ 75th percentile [1.012 (0.145; 1.880); p value = 0.022]. Women without autoimmune disease were more likely to have IMT ≥ the 75th percentile [-2.181 (-4.214; -0.149); p value = 0.035], but the analysis of IMT as a continuous variable did not yield significant results. In conclusion, subclinical carotid atherosclerosis, but not arterial stiffness, was more common in men with autoimmune diseases. Women did not show significant differences in any of these carotid features. Sex was an effect modifier in the association between common carotid IMT values and the diagnosis of autoimmune diseases.

Semantic segmentation with DenseNets for carotid artery ultrasound plaque segmentation and CIMT estimation.

BACKGROUND AND OBJECTIVE: The measurement of carotid intima media thickness (CIMT) in ultrasound images can be used to detect the presence of atherosclerotic plaques. Usually, the CIMT estimation strategy is semi-automatic, since it requires: (1) a manual examination of the ultrasound image for the localization of a region of interest (ROI), a fast and useful operation when only a small number of images need to be measured; and (2) an automatic delineation of the CIM region within the ROI. The existing efforts for automating the process have replicated the same two-step structure, resulting in two consecutive independent approaches. In this work, we propose a fully automatic single-step approach based on semantic segmentation that allows us to segment the plaque and to estimate the CIMT in a fast and useful manner for large data sets of images. METHODS: Our single-step approach is based on densely connected convolutional neural networks (DenseNets) for semantic segmentation of the whole image. It has two remarkable characteristics: (1) it avoids ROI definition, and (2) it captures multi-scale contextual information in the complete image interpretation, due to the concatenation of feature maps carried out in DenseNets. Once the input image is segmented, a straightforward method for CIMT estimation and plaque detection is applied. RESULTS: The proposed method has been validated with a large data set (REGICOR) of more than 8000 images, corresponding to two territories of the carotid artery: common carotid artery (CCA) and bulb. Among them, a subset of 331 images has been used to evaluate the performance of semantic segmentation (≈90% for train, ≈10% for test). The experimental results demonstrated that our method outperforms other deep models and shallow approaches found in the literature. In particular, our CIMT estimation reaches a correlation coefficient of 0.81, and a CIMT mean error of 0.02 and 0.06 mm in CCA and Bulb images, respectively. Furthermore, the accuracy for plaque detection is 96.45% and 78.09% in CCA and Bulb, respectively. To test the generalization power, the method has also been tested with another data set (NEFRONA) that includes images acquired with different equipment. CONCLUSIONS: The validation carried out demonstrates that the proposed method is accurate and objective for both plaque detection and CIMT measurement. Moreover, the robustness and generalization capacity of the method have been proven with two different data sets.

A review of feature selection methods in medical applications.

Feature selection is a preprocessing technique that identifies the key features of a given problem. It has traditionally been applied in a wide range of problems that include biological data processing, finance, and intrusion detection systems. In particular, feature selection has been successfully used in medical applications, where it can not only reduce dimensionality but also help us understand the causes of a disease. We describe some basic concepts related to medical applications and provide some necessary background information on feature selection. We review the most recent feature selection methods developed for and applied in medical problems, covering prolific research fields such as medical imaging, biomedical signal processing, and DNA microarray data analysis. A case study of two medical applications that includes actual patient data is used to demonstrate the suitability of applying feature selection methods in medical problems and to illustrate how these methods work in real-world scenarios.

Parallel definition of tear film maps on distributed-memory clusters for the support of dry eye diagnosis.

BACKGROUND AND OBJECTIVES: The analysis of the interference patterns on the tear film lipid layer is a useful clinical test to diagnose dry eye syndrome. This task can be automated with a high degree of accuracy by means of the use of tear film maps. However, the time required by the existing applications to generate them prevents a wider acceptance of this method by medical experts. Multithreading has been previously successfully employed by the authors to accelerate the tear film map definition on multicore single-node machines. In this work, we propose a hybrid message-passing and multithreading parallel approach that further accelerates the generation of tear film maps by exploiting the computational capabilities of distributed-memory systems such as multicore clusters and supercomputers. METHODS: The algorithm for drawing tear film maps is parallelized using Message Passing Interface (MPI) for inter-node communications and the multithreading support available in the C++11 standard for intra-node parallelization. The original algorithm is modified to reduce the communications and increase the scalability. RESULTS: The hybrid method has been tested on 32 nodes of an Intel cluster (with two 12-core Haswell 2680v3 processors per node) using 50 representative images. Results show that maximum runtime is reduced from almost two minutes using the previous only-multithreaded approach to less than ten seconds using the hybrid method. CONCLUSIONS: The hybrid MPI/multithreaded implementation can be used by medical experts to obtain tear film maps in only a few seconds, which will significantly accelerate and facilitate the diagnosis of the dry eye syndrome.

Evaluation of an automatic dry eye test using MCDM methods and rank correlation.

Dry eye is an increasingly common disease in modern society which affects a wide range of population and has a negative impact on their daily activities, such as working with computers or driving. It can be diagnosed through an automatic clinical test for tear film lipid layer classification based on color and texture analysis. Up to now, researchers have mainly focused on the improvement of the image analysis step. However, there is still large room for improvement on the machine learning side. This paper presents a methodology to optimize this problem by means of class binarization, feature selection, and classification. The methodology can be used as a baseline in other classification problems to provide several solutions and evaluate their performance using a set of representative metrics and decision-making methods. When several decision-making methods are used, they may offer disagreeing rankings that will be solved by conflict handling in which rankings are merged into a single one. The experimental results prove the effectiveness of the proposed methodology in this domain. Also, its general purpose allows to adapt it to other classification problems in different fields such as medicine and biology.

iDEAS: A web-based system for dry eye assessment.

BACKGROUND AND OBJECTIVES: Dry eye disease is a public health problem, whose multifactorial etiology challenges clinicians and researchers making necessary the collaboration between different experts and centers. The evaluation of the interference patterns observed in the tear film lipid layer is a common clinical test used for dry eye diagnosis. However, it is a time-consuming task with a high degree of intra- as well as inter-observer variability, which makes the use of a computer-based analysis system highly desirable. This work introduces iDEAS (Dry Eye Assessment System), a web-based application to support dry eye diagnosis. METHODS: iDEAS provides a framework for eye care experts to collaboratively work using image-based services in a distributed environment. It is composed of three main components: the web client for user interaction, the web application server for request processing, and the service module for image analysis. Specifically, this manuscript presents two automatic services: tear film classification, which classifies an image into one interference pattern; and tear film map, which illustrates the distribution of the patterns over the entire tear film. RESULTS: iDEAS has been evaluated by specialists from different institutions to test its performance. Both services have been evaluated in terms of a set of performance metrics using the annotations of different experts. Note that the processing time of both services has been also measured for efficiency purposes. CONCLUSIONS: iDEAS is a web-based application which provides a fast, reliable environment for dry eye assessment. The system allows practitioners to share images, clinical information and automatic assessments between remote computers. Additionally, it save time for experts, diminish the inter-expert variability and can be used in both clinical and research settings.

CASDES: A Computer-Aided System to Support Dry Eye Diagnosis Based on Tear Film Maps.

Dry eye syndrome is recognized as a growing health problem, and one of the most frequent reasons for seeking eye care. Its etiology and management challenge clinicians and researchers alike, and several clinical tests can be used to diagnose it. One of the most frequently used tests is the evaluation of the interference patterns of the tear film lipid layer. Based on this clinical test, this paper presents CASDES, a computer-aided system to support the diagnosis of dry eye syndrome. Furthermore, CASDES is also useful to support the diagnosis of other eye diseases, such as meibomian gland dysfunction, since it provides a tear film map with highly useful information for eye practitioners. Experiments demonstrate the robustness of this novel tool, which outperforms the previous attempts to create tear film maps and provides reliable results in comparison with the clinicians' annotations. Note that the processing time is noticeably reduced with the proposed method, which will help to promote its clinical use in the diagnosis and treatment of dry eye.

Correlation between tear osmolarity and tear meniscus.

PURPOSE: To examine the relationship between tear meniscus height (TMH) and subjective meniscus grading (subjective tear meniscus [TM]) with tear osmolarity. METHODS: Tear osmolarity measurements (using TearLab) and digital images of the TM were obtained in 177 consecutive patients undergoing an eye examination at our optometry clinic (Universidad de Santiago de Compostela, Spain) who fulfilled the study's inclusion criteria. Participants were also administered the McMonnies and Ocular Surface Disease Index questionnaires for the detection of dry eye disease. The lower TM was videotaped by a digital camera attached to a slit lamp in its central portion without fluorescein instillation. After the study, a masked observer extracted an image from each video and measured the TMH using open source software (NIH ImageJ). Subsequently, the masked observer subjectively graded the appearance of each meniscus. For statistical analysis, subjects were stratified by age and by dry eye symptoms as indicated by their scores in the two questionnaires. RESULTS: In the whole study population, a significant relationship was observed between osmolarity and TMH (r = -0.41, p < 0.001) and osmolarity and subjective TM (r = 0.35, p < 0.001). A cluster analysis revealed similar correlations when subjects were stratified by age or dry eye symptoms, these correlations being more pronounced in older and more symptomatic subjects. Objective TMH measurements and subjective meniscus quality were also correlated (r = -0.75, p < 0.001). CONCLUSIONS: Osmolarity and both objective TMH measurements and subjective interpretation of the meniscus showed high correlation, especially in older symptomatic subjects.

A methodology for improving tear film lipid layer classification.

Dry eye is a symptomatic disease which affects a wide range of population and has a negative impact on their daily activities. Its diagnosis can be achieved by analyzing the interference patterns of the tear film lipid layer and by classifying them into one of the Guillon categories. The manual process done by experts is not only affected by subjective factors but is also very time consuming. In this paper we propose a general methodology to the automatic classification of tear film lipid layer, using color and texture information to characterize the image and feature selection methods to reduce the processing time. The adequacy of the proposed methodology was demonstrated since it achieves classification rates over 97% while maintaining robustness and provides unbiased results. Also, it can be applied in real time, and so allows important time savings for the experts.