The current adoption of Medical Artificial Intelligence (AI) solutions in clinical practice suggest that despite its undeniable potential AI is not achieving this potential. A major barrier to its adoption is the lack of transparency and interpretability, and the inability of the system to explain its results. Explainable AI (XAI) is an emerging field in AI that aims to address these barriers, with the development of new or modified algorithms to enable transparency, provide explanations in a way that humans can understand and foster trust. Numerous XAI techniques have been proposed in the literature, commonly classified as model-agnostic or model-specific. In this study, we examine the application of four model-agnostic XAI techniques (LIME, SHAP, ANCHORS, inTrees) to an XGBoost classifier trained on real-life medical data for the prediction of high-risk asymptomatic carotid plaques based on ultrasound image analysis. We present and compare local explanations for selected observations in the test set. We also present global explanations generated from these techniques that explain the behavior of the entire model. Additionally, we assess the quality of the explanations, using suggested properties in the literature. Finally, we discuss the results of this comparative study and suggest directions for future work.

Despite the recent recognition of the value of Artificial Intelligence and Machine Learning in healthcare, barriers to further adoption remain, mainly due to their "black box" nature and the algorithm's inability to explain its results. In this paper we present and propose a methodology of applying argumentation on top of machine learning to build explainable AI (XAI) models. We compare our results with Random Forests and an SVM classifier that was considered best for the same dataset in [1].

Through this study we are presenting the initial steps towards a real time motion analysis system to predict the stability of carotid bifurcation plaques. The analysis is performed on B-mode video loops. Loops are analyzed in order to follow systole and diastole sections of the cardiac cycle and trace the motion of plaques during these periods. We had created a system that applies Farnebacks optical flow estimation method in order to estimate the flow between consecutive frames or frames at a predefined interval. Over each pair of video frames we measure velocities, orientation and magnitude of movement. The goal is to identify if a plaque has movement spread to different angles or at nearby angles. This can help us identify discordant or concordant movement. In order to verify our system we had created a set of simulated videos that have structures moving in a similar way as done in a cardiac cycle and videos that move and appear as an atherosclerotic artery. Following these tests the system has been tested and results are presented on two carotid plaques videos classified visually as having concordant and discordant plaque movement.

This paper aims at evaluating the changes that head rotation poses on morphological and flow characteristics of the carotid bifurcation (CB) and on the distribution of parameters that are regarded as important in atherosclerosis development, such as relative particle residence time (RRT), normalized oscillatory shear index (nOSI), and helicity (HL). Using a subject-specific approach, six healthy volunteers were MR-scanned in two head postures: supine neutral and prone with rightward head rotation. Cross-sectional flow velocity distribution was obtained using phase-contrast MRI at the common carotid artery (CCA). Our results indicate that peak systolic flow rate is reduced at the prone position in most cases for both CCAs. Morphological MR images are used to segment and construct the CB models. Numerical simulations are performed and areas exposed to high helicity or unfavorable hemodynamics are calculated. Head rotation affects the instantaneous spatial extent of high helicity regions. Posture-related observed differences in the distribution of nOSI and RRT suggest that inlet flow waveform tends to moderate geometry-induced changes in the qualitative and quantitative distribution of atherosclerosis-susceptible wall regions. Overall, presented results indicate that an individualized approach is required to fully assess the postural role in atherosclerosis development and in complications arising in stenotic and stented vessels.

The intima-media thickness (IMT) of the common carotid artery (CCA) is an established indicator of cardiovascular disease (CVD). There have been reports about the difference between the left and the right sides of the CCA IMT and its importance when evaluated with various risk factors as well as their association with the risk of stroke. In this study, we use an automated system based on snakes, for segmenting the CCA and perform measurements of the IMT of the carotid artery and provide their differences between the left and right sides. The study was performed on 205 longitudinal-section ultrasound images acquired from 87 men and 118 women at a mean±SD age of 63±10.47 years, out of which 51 had cardiovascular symptoms. A cardiovascular expert manually measured the IMT on the left CCA side (mean±standard deviation = 0.79±0.21 mm) and the right CCA side (0.76±0.33 mm). The left and right IMT automated measurements were 0.70±0.15 mm and 0.66±0.15 mm, respectively. We found no statistical significant differences: 1) between the left and right IMT measurements, for both the manual and automated measurements, and 2) between the manual and automated measurements for both sides. These findings suggest that the measurement of the CCA IMT on one side only is enough (and this is in agreement with other studies), as well as automated measurements can be used.

Video loops of B-mode ultrasound images of 35 carotid bifurcation plaques were obtained (4 symptomatic and 31 asymptomatic) from patients with carotid bifurcation atherosclerosis. Video loops were classified visually as showing concordant (n=22) or discordant motion (n=13). Concordant plaques were characterized by uniform orientation of motion throughout the cardiac cycle. Discordant plaques exhibited significant spread in motion orientation at different parts of the cardiac cycle, especially at systole. We developed a real-time motion analysis system that applies Farneback's method to estimate velocities between consecutive video frames. For our purposes, we allow a 100msec time interval between the video frames used in the analysis. This approach allows us to analyze significant motions associated with a larger time interval. Over each video frame, we measure the spread of the motion orientation around the dominant orientation. For each video, we look at the spreads of the motion orientations for different motion magnitudes. Using these motion-spread measurements, we can quantify discordant movement. The sum of maximum fan widths for the median pixel motions 5 to 3 (SMFW5to3) had a median value of 100 degrees and interquartile range (IQR) of (80, 110) degrees for the concordant plaques and 270, (230, 430) for the discordant plaques (P < 0.001). Thus, we have a new tool to differentiate between concordant and discordant plaques.

The aim of this paper is to describe a novel and completely automated technique for carotid artery (CA) recognition, far (distal) wall segmentation, and intima-media thickness (IMT) measurement, which is a strong clinical tool for risk assessment for cardiovascular diseases. The architecture of completely automated multiresolution edge snapper (CAMES) consists of the following two stages: 1) automated CA recognition based on a combination of scale-space and statistical classification in a multiresolution framework and 2) automated segmentation of lumen-intima (LI) and media-adventitia (MA) interfaces for the far (distal) wall and IMT measurement. Our database of 365 B-mode longitudinal carotid images is taken from four different institutions covering different ethnic backgrounds. The ground-truth (GT) database was the average manual segmentation from three clinical experts. The mean distance ± standard deviation of CAMES with respect to GT profiles for LI and MA interfaces were 0.081 ± 0.099 and 0.082 ± 0.197 mm, respectively. The IMT measurement error between CAMES and GT was 0.078 ± 0.112 mm. CAMES was benchmarked against a previously developed automated technique based on an integrated approach using feature-based extraction and classifier (CALEX). Although CAMES underestimated the IMT value, it had shown a strong improvement in segmentation errors against CALEX for LI and MA interfaces by 8% and 42%, respectively. The overall IMT measurement bias for CAMES improved by 36% against CALEX. Finally, this paper demonstrated that the figure-of-merit of CAMES was 95.8% compared with 87.4% for CALEX. The combination of multiresolution CA recognition and far-wall segmentation led to an automated, low-complexity, real-time, and accurate technique for carotid IMT measurement. Validation on a multiethnic/multi-institutional data set demonstrated the robustness of the technique, which can constitute a clinically valid IMT measurement for assistance in atherosclerosis disease management.

Carotid plaques have been associated with ipsilateral neurological symptoms. High-resolution ultrasound can provide information not only on the degree of carotid artery stenosis but also on the characteristics of the arterial wall including the size and consistency of atherosclerotic plaques. The aim of this study is to determine whether the addition of ultrasonic plaque texture features to clinical features in patients with asymptomatic internal carotid artery stenosis (ACS) improves the ability to identify plaques that will produce stroke. 1121 patients with ACS have been scanned with ultrasound and followed for a mean of 4 years. It is shown that the combination of texture features based on second-order statistics spatial gray level dependence matrices (SGLDM) and clinical factors improves stroke prediction (by correctly predicting 89 out of the 108 cases that were symptomatic). Here, the best classification results of 77 ±1.8% were obtained from the use of the SGLDM texture features with support vector machine classifiers. The combination of morphological features with clinical features gave slightly worse classification results of 76 ±2.6% . These findings need to be further validated in additional prospective studies.

Noninvasive ultrasound imaging of carotid plaques allows for the development of plaque-image analysis methods associated with the risk of stroke. This paper presents several plaque-image analysis methods that have been developed over the past years. The paper begins with a review of clinical methods for visual classification that have led to standardized methods for image acquisition, describes methods for image segmentation and denoising, and provides an overview of the several texture-feature extraction and classification methods that have been applied. We provide a summary of emerging trends in 3-D imaging methods and plaque-motion analysis. Finally, we provide a discussion of the emerging trends and future directions in our concluding remarks.

The intima-media thickness (IMT) of the common carotid artery (CCA) is widely used as an early indicator of cardiovascular disease (CVD). It was proposed but not thoroughly investigated that the media layer (ML), its composition and texture, may be indicative for identifying the risk of stroke and differentiating between patients of high and low risk. In this study we investigate the usefulness of texture analysis of the ML of the CCA. The study was performed on 100 longitudinal ultrasound images acquired from asymptomatic subjects at risk of atherosclerosis. The images were separated into three different age groups, namely below 50, 50 to 60, and above 60 years old. A total of 61 different texture features were extracted from the intima-media complex (IMC), ML and the intima layer (IL). The IMC and ML were segmented manually by a neurovascular expert and automatically by a snakes segmentation system. It was shown that texture features extracted from the IL, ML and IMC are significantly different (mean, gray scale median (GSM), standard deviation, contrast, difference variance, periodicity) and that some of them can be associated with the increase (difference variance, entropy) or decrease (GSM) of patientpsilas age. It was also shown that the GSM of the ML falls linearly with increasing ML thickness (MLT) and with increasing age. Further research on more subjects is required for estimating other features that may provide information for patients at risk of stroke.

In this paper, we propose and evaluate an integrated system for the segmentation of atherosclerotic plaque in ultrasound imaging of the carotid artery based on normalization, speckle reduction filtering, and four different snakes segmentation methods. These methods are the Williams and Shah, Balloon, Lai and Chin, and the gradient vector flow (GVF) snake. The performance of the four different plaque snakes segmentation methods was tested on 80 longitudinal ultrasound images of the carotid artery using receiver operating characteristic (ROC) analysis and the manual delineations of an expert. All four methods were very satisfactory and similar in all measures evaluated, with no significant differences between them; however, the Lai and Chin snakes segmentation method gave slightly better results. Concluding, it is proposed that the integrated system investigated in this study could be used successfully for the automated segmentation of the carotid plaque.

There are indications that the morphology of atherosclerotic carotid plaques, obtained by high-resolution ultrasound imaging, has prognostic implications. The objective of this study was to develop a computer-aided system that will facilitate the characterization of carotid plaques for the identification of individuals with asymptomatic carotid stenosis at risk of stroke. A total of 230 plaque images were collected which were classified into two types: symptomatic because of ipsilateral hemispheric symptoms, or asymptomatic because they were not connected with ipsilateral hemispheric events. Ten different texture feature sets were extracted from the manually segmented plaque images using the following algorithms: first-order statistics, spatial gray level dependence matrices, gray level difference statistics, neighborhood gray tone difference matrix, statistical feature matrix, Laws texture energy measures, fractal dimension texture analysis, Fourier power spectrum and shape parameters. For the classification task a modular neural network composed of self-organizing map (SOM) classifiers, and combining techniques based on a confidence measure were used. Combining the classification results of the ten SOM classifiers inputted with the ten feature sets improved the classification rate of the individual classifiers, reaching an average diagnostic yield (DY) of 73.1%. The same modular system was implemented using the statistical k-nearest neighbor (KNN) classifier. The combined DY for the KNN system was 68.8%. The results of this paper show that it is possible to identify a group of patients at risk of stroke based on texture features extracted from ultrasound images of carotid plaques. This group of patients may benefit from a carotid endarterectomy whereas other patients may be spared from an unnecessary operation.

The objective of this work is to develop a system that will facilitate the automated characterization of ultrasonic imaging carotid plaques for the identification of individuals with asymptomatic carotid stenosis at risk of stroke. A total of 166 images were collected which were classified into: symptomatic because of ipsilateral hemispheric symptoms, or asymptomatic because they were not connected with ipsilateral hemispheric events. Ten different texture feature sets were extracted: first order statistics, spatial gray level dependence matrices, gray level difference statistics, neighbourhood gray tone difference matrix, statistical feature matrix, Laws texture energy measures, fractal dimension texture analysis, Fourier power spectrum and shape parameters. A modular neural network classifier was developed composed of self-organizing map (SOM) classifiers, achieving an overall diagnostic yield of 76.4%. The results of this work show that it is possible to identify a group of patients at risk of stroke based on texture features.

We develop a computer aided system which will facilitate the automated characterisation of carotid plaques recorded from high resolution ultrasound images for the identification of individuals with asymptomatic carotid stenosis at risk of stroke. The plaques were classified into: symptomatic or asymptomatic. Ten different texture feature sets were extracted from the segmented plaque image. Although the statistics of all features extracted for the two classes indicated a high degree of overlap, a classification of the plaques was possible using the unsupervised self-organizing feature map (SOFM) classifier and combining techniques. The classification results of the different feature sets were combined using the majority voting and weighted averaging based on a confidence measure derived from the SOFM. Combining the classification results of the ten different feature sets improved significantly the classification results obtained by the individual feature sets, reaching an average diagnostic yield of 75%.