Alireza Roodaki

Automatic segmentation of calcified plaques and vessel borders in IVUS images

ObjectiveIntravascular ultrasound (IVUS) is a diagnostic imaging technique for tomographic visualization of coronary arteries. Automatic analysis of IVUS images is difficult due to speckle noise, artifacts of the catheter, and shadows generated by calcifications. We designed and implemented a system for automated segmentation of coronary artery IVUS images.MethodsTwo methods for automatic detection of the intima and the media-adventitia borders in IVUS coronary artery images were developed and compared. The first method uses the parametric deformable models, while the second method is based on the geometric deformable models. The initial locations of the borders are approximated using two different edge detection methods. The final borders are then defined using the two deformable models. Finally, the calcified regions between the extracted borders are identified using a Bayesian classifier. The performance of the proposed methods was evaluated using 60 different IVUS images obtained from 7 patients.ResultsSegmented images were compared with manually outlined contours. We compared the performance of calcified region characterization methods using ROC analysis and by computing the sensitivity and specificity of the Bayesian classifier, thresholding, adaptive thresholding, and textural features. The Bayesian method performed best.ConclusionThe results shows that the geometric deformable model outperforms the parametric deformable model for automated segmentation of IVUS coronary artery images.

Fisher linear discriminant based person identification using visual evoked potentials

Biometrics is the technique of uniquely recognizing a person among a group of people. It is usually performed based on one or more of humanpsilas intrinsic physical or behavioral traits. One such trait is the electroencephalogram (EEG) signal. In this paper, the feasibility of visual evoked potential (VEP) in the gamma band of EEG signal, as a physiological trait, is studied, and used to identify individuals in a group of 20 people. To this end, the parameters of the AR model together with the peak of the power spectrum density (PSD) of the gamma band VEP signal (GMVEP) are considered as main useful features. Next, the Fisherpsilas linear discriminant (FLD) is used to reduce the feature vector dimensions. Finally, the k nearest neighborhood (KNN) technique is employed to classify the data and the leave-one-out cross validation method is used for accuracy assessment. A correct classification rate of 100% is achieved.

A New Approach for Improving Coronary Plaque Component Analysis Based on Intravascular Ultrasound Images

Virtual histology intravascular ultrasound (VH-IVUS) is a clinically available technique for atherosclerosis plaque characterization. It, however, suffers from a poor longitudinal resolution due to electrocardiogram (ECG)-gated acquisition. This article presents an effective algorithm for IVUS image-based histology to overcome this limitation. After plaque area extraction within an input IVUS image, a textural analysis procedure consisting of feature extraction and classification steps is proposed. The pixels of the extracted plaque area excluding the shadow region were classified into one of the three plaque components of fibro-fatty (FF), calcification (CA) or necrotic core (NC) tissues. The average classification accuracy for pixel and region based validations is 75% and 87% respectively. Sensitivities (specificities) were 79% (85%) for CA, 81% (90%) for FF and 52% (82%) for NC. The kappa (kappa) = 0.61 and p value = 0.02 indicate good agreement of the proposed method with VH images. Finally, the enhancement in the longitudinal resolution was evaluated by reconstructing the IVUS images between the two sequential IVUS-VH images.

An IVUS image-based approach for improvement of coronary plaque characterization

Virtual Histology-Intravascular Ultrasound (VH-IVUS) is widely used for studying atherosclerosis plaque composition. However, one of the main limitations of the VH-IVUS relates to its dependence to the Electrocardiogram (ECG)-gated acquisition. To overcome this limitation, this paper proposes a robust image-based approach for characterization of the plaques using IVUS images. The proposed method consists of three main steps of (1) shadow detection: as an efficient preprocessing step to identify and remove acoustic shadow regions; (2) feature extraction: a combination of gray-scale based features and textural descriptors; and (3) classification: to classify each pixel into one of the three classes (calcium, necrotic core and fibro-fatty). In order to evaluate the efficiency of the proposed algorithm two in-vivo and ex-vivo data sets are considered. The kappa values of 0.639 on in-vivo and 0.628 on ex-vivo tests with VH-IVUS and the histology images labeled by the experts respectively indicate the effectiveness of the proposed algorithm.

A new method for characterization of coronary plaque composition via IVUS images

IVUS-derived virtual histology (VH) permits the assessment of atherosclerotic plaque morphology by using radiofrequency analysis of ultrasound signals. However, it requires the acquisition to be ECG-gated, which is a major limitation of VH. Indeed, its computation can only be performed once per cardiac cycle, which significantly decreases the longitudinal resolution of VH. To overcome this limitation, the introduction of an image-based plaque characterization is of great importance. Current IVUS image processing techniques do not allow adequate identification of the coronary artery plaques. This can be improved by defining appropriate features for the different kinds of plaques. In this paper, a novel feature extraction method based on Run-length algorithm is presented and used for improving the automated characterization of the plaques within the IVUS images. The proposed feature extraction method is applied to 200 IVUS images obtained from five patients. As a result an accuracy rate of 77% was achieved. Comparing this to the accuracy rates of 75% and 71% obtained using co-occurrence and local binary pattern methods respectively indicates the superior performance of the proposed feature extraction method.

Modified wavelet transform features for characterizing different plaque types in IVUS images; A feasibility study

Atherosclerosis is a leading cause of most cardiovascular diseases. Current intravascular ultrasound (IVUS) image processing techniques do not allow adequate and effective identification of the coronary artery plaques. This can be improved by defining more discriminative features for each kind of artery plaques. In this paper, the effectiveness of a modified wavelet transform feature extraction method and the Gabor filter were studied for automated characterization of the atherosclerosis plaques within the IVUS images. The methods are applied on 100 IVUS images obtained from five different patients. Support vector machine was employed in the classification step. As a result an accuracy rate of about 80% was achieved for all methods.

Summarizing posterior distributions in signal decomposition problems when the number of components is unknown

This paper addresses the problems of relabeling and summarizing posterior distributions that typically arise, in a Bayesian framework, when dealing with signal decomposition problems with an unknown number of components. Such posterior distributions are defined over union of subspaces of differing dimensionality and can be sampled from using modern Monte Carlo techniques, for instance, the increasingly popular RJ-MCMC method. No generic approach is available, however, to summarize the resulting variable-dimensional samples and extract from them component-specific parameters. We propose a novel approach, named Variable-dimensional Approximate Posterior for Relabeling and Summarizing (VAPoRS), to this problem, which consists of approximating the posterior distribution of interest by a “simple”-but still variable-dimensional-parametric distribution. The distance between the two distributions is measured using the Kullback-Leibler divergence, and a Stochastic EM-type algorithm, driven by the RJ-MCMC sampler, is proposed to estimate the parameters. Two signal decomposition problems are considered to show the capability of VAPoRS both for relabeling and for summarizing variable dimensional posterior distributions: the classical problem of detecting and estimating sinusoids in white Gaussian noise on the one hand, and a particle counting problem motivated by the Pierre Auger project in astrophysics on the other hand.

Comments on “Joint Bayesian Model Selection and Estimation of Noisy Sinusoids Via Reversible Jump MCMC”

Reversible jump MCMC (RJ-MCMC) sampling techniques, which allow to jointly tackle model selection and parameter estimation problems in a coherent Bayesian framework, have become increasingly popular in the signal processing literature since the seminal paper of Andrieu and Doucet [“Joint Bayesian model selection and estimation of noisy sinusoids via reversible jump MCMC,” IEEE Trans. Signal Process, vol. 47, no. 10, pp. 2667-2676, 1999]. Crucial to the implementation of any RJ-MCMC sampler is the computation of the so-called Metropolis-Hastings-Green (MHG) ratio, which determines the acceptance probability for the proposed moves. It turns out that the expression of the MHG ratio that was given in the paper of Andrieu and Doucet for “Birth-or-Death” moves is erroneous and has been reproduced in many subsequent papers dealing with RJ-MCMC sampling in the signal processing literature. This note fixes the erroneous expression and briefly discusses its cause and consequences.

Polar Run-Length Features in Segmentation of Retinal Blood Vessels

Manual segmentation of retinal blood vessels in optic fundus images is a tiresome task. Several methods have previously been proposed for the automatic segmentation of retinal blood vessels. In this paper we propose a classifier-based method. First the images are preprocessed so that the within class variability of the vessel and background classes are minimized. Next, the image is scanned with a window of a certain size. Polar run-length matrices are simply created by transforming the windows into polar coordinates and then constructing conventional run length matrices. Two features are then extracted for each gray level value in the polar run length matrix. The feature vectors are then classified using a multilayer perceptron artificial neural network. The performance of the proposed method is compared with that of the human observers and with those methods previously reported in literature.

Detection of Outer Layer of the Vessel Wall and Characterization of Calcified Plaques in IVUS Images

Intravascular Ultrasound (IVUS) is a diagnostic imaging technique that provides tomographic visualization of coronary arteries. Important challenges in analysis of IVUS images are speckle noise, artifacts of catheter and calcified shadows. In this paper, we present a method for the automated detection of outer (media-adventitia) border of vessel by the use of geometric deformable models. Speckle noise is reduced with median filter. The initial contour is extracted using Canny edge detection and finally the calcified regions are characterized by using Bayes classifier and thresholding methods. The proposed methods were evaluated on 60 IVUS images from 7 different patients. The results show that the border detection method was statistically accurate and in the range of inter observer variability (based on the used validation methods). Bayesian classifier enables us to characterize the regions of interest, with a sensitivity and specificity of 92.67% and 98.5% respectively.