Texture Analysis for the Classification of Carotid Plaques (original) (raw)
Multi-feature texture analysis for the classification of carotid plaques
1999
Abstract 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.
Selection of parameters for texture analysis for the classification of carotid plaques
IEEE Trans Med Imaging, 2003
Abstract: Texture features extracted from high-resolution ultrasound images of carotid plaques can be used for the identification of patients at risk of stroke. This work explores the selection of the parameters for the computation of the texture features, which will yield the best class separation between symptomatic and asymptomatic subjects. The following texture algorithms were investigated on 230 carotid plaque images (recorded from 115 symptomatic and 115 asymptomatic subjects): Spatial Gray Level Dependence Matrices ( ...
Texture-based classification of atherosclerotic carotid plaques
2003
Abstract 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.
There are indications that the texture of certain components of atherosclerotic carotid plaques in the common carotid artery (CCA), obtained by high resolution ultrasound imaging, may have additional prognostic implication for the risk of stroke. The objective of this study was to perform texture analysis of the middle component of atherosclerotic carotid plaques in 230 CCA plaque ultrasound images (115 asymptomatic and 115 symptomatic). These were manually delineated by a neurovascular expert after normalization and despeckle filtering using the linear despeckle filter (DsFlsmv). Texture features were extracted from the middle plaque component. We found statistical significant differences for some of the texture features extracted, between asymptomatic and symptomatic subjects. The results showed that it may be possible to identify a group of patients at risk of stroke (asymptomatic versus symptomatic) based on texture features extracted from the middle component of the atheroscler...
Electrical Engineering & Applied Signal Processing Series, 2005
There are indications that the texture of certain components of atherosclerotic carotid plaques in the common carotid artery (CCA), obtained by high resolution ultrasound imaging, may have additional prognostic implication for the risk of stroke. The objective of this study was to perform texture analysis of the middle component of atherosclerotic carotid plaques in 230 CCA plaque ultrasound images (115 asymptomatic and 115 symptomatic). These were manually delineated by a neurovascular expert after normalization and despeckle filtering using the linear despeckle filter (DsFlsmv). Texture features were extracted from the middle plaque component. We found statistical significant differences for some of the texture features extracted, between asymptomatic and symptomatic subjects. The results showed that it may be possible to identify a group of patients at risk of stroke (asymptomatic versus symptomatic) based on texture features extracted from the middle component of the atherosclerotic carotid plaque in ultrasound images of the CCA.
2003
The extraction of features characterizing the structure of atherosclerotic carotid plaques, obtained by high-resolution ultrasound imaging is important for the correct plaque classification and the estimation of the risk of stroke. In this study morphological features were extracted and compared with the well-known texture features spatial gray level dependence matrices (SGLDM), gray level difference statistics (GLDS) and the first order statistics (FOS) for the classification of 330 carotid plaques.
2003
The extraction of features characterizing the structure of atherosclerotic carotid plaques, obtained by high-resolution ultrasound imaging is important for the correct plaque classification and the estimation of the risk of stroke. In this study morphological features were extracted and compared with the well-known texture features spatial gray level dependence matrices (SGLDM), gray level difference statistics (GLDS) and the first order statistics (FOS) for the classification of 330 carotid plaques. For the classification the neural self-organizing map (SOM) classifier and the statistical k-nearest neighbor (KNN) classifier were used. The results showed that morphological and other texture features are comparable, with the morphological and the GLDS feature sets to perform slightly better than the SGLDM and the FOS features. The highest diagnostic yield was achieved with the GLDS feature set and it was about 70%.
Computer based analysis of ultrasound images for assessing carotid artery plaque risk
2003
To design and implement a computer based image analysis system, employing pattern recognition methods on ultrasound images, for assessing carotid plaque risk of causing brain infarcts. Materials and methods: Sixty-one ultrasound images displaying carotid artery stenosis were selected by an HDI-3000 ATL digital ultrasound system. Plaques were categorized on the basis of the gray scale median (GSM) as echolucent (GSMSSO gray level), with high risk of causing brain infarcts, and echogenic (GSM>SO gray level) and in accordance with the physician's assessment and final clinical outcome. Thirty-eight textural features were calculated ?om the carotid plaque's image, 4 ?om the image histogram, 28 j?om the co-occurrence matrix, and I O ?om the run-length matrix. Two classi$er.v. the least squares minimum distance (LSMDI and the support vector machines (SVM) were employed for comparison reasons. Textural features were employed as input to the class$ers. which were trained to characterize plaques as either high risk or low risk of causing brain infarcts. Results: SVM classr$cation accuracy was 96.7% employing the following twa textural features: (a) mean gray-level and (b) image contrast, a measure of local variations present in the image, from !he co-occurrence matrix. Comparatively, LSMD classification precision, employing the same textural feature combination, was 86.9%. Conclusion: The proposed image analysis system, employing textural features and the SVM classifier, may be indicative of carotid plaque risk of causing brain infarcts and may be of value to patient management.
2011
Abstract 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 was to determine cerebrovascular risk stratification based on ultrasonic plaque texture features and clinical features in patients with asymptomatic internal carotid artery (ICA) stenosis.