DocumentCode :
1767159
Title :
Bilateral asymmetry in ultrasound-image-based mechanical and textural features in subjects with asymptomatic carotid artery disease
Author :
Golemati, S. ; Gastounioti, A. ; Tsiaparas, Nikolaos N. ; Nikita, Konstantina S.
Author_Institution :
Nat. Kapodistrian Univ. of Athens, Athens, Greece
fYear :
2014
fDate :
1-4 June 2014
Firstpage :
680
Lastpage :
683
Abstract :
Carotid atherosclerosis is a principal cause of stroke and transient ischaemic attack. It has recently been postulated that potential asymmetries between the left and right carotids may be important in terms of disease progression. In this work, we sought to investigate differences between the two carotid arteries using ultrasound-image-based mechanical and textural features. Sixteen arterial segments (8 from the left and 8 from the corresponding right side) of elderly subjects with bilateral asymptomatic carotid atherosclerosis were interrogated. Left and right sides had similar stenosis degrees. Mechanical features were estimated via motion analysis using adaptive block matching. They included motion amplitudes, velocities and diastole-to-systole displacements of selected regions around and adjacent to the plaque. Textural features included first- and second-order statistical as well as multiresolution measures. One texture and eleven mechanical features were significantly different between the two arterial sides, indicating that bilateral asymmetry is characterised better in terms of mechanical properties, rather than via tissue composition, which is expressed through texture analysis. These findings suggest that not only systemic, but also focal, factors are probably responsible for atherosclerosis development. The identification and definition of the role of these factors may help to clarify the pathogenesis of this disease and to improve the clinical and therapeutic approaches.
Keywords :
adaptive systems; biomechanics; biomedical ultrasonics; blood vessels; diseases; feature extraction; feature selection; geriatrics; image matching; image motion analysis; image texture; medical image processing; statistical analysis; symmetry; adaptive block matching; arterial segment; asymptomatic carotid artery disease; atherosclerosis development; bilateral asymmetry characterisation; bilateral asymptomatic carotid atherosclerosis; clinical approaches; disease progression; elderly patients; first-order statistical measures; focal factors; left carotid asymmetries; mechanical feature estimation; motion amplitudes; motion analysis; motion velocities; multiresolution measures; pathogenesis; plaque region selection; potential carotid asymmetries; region diastole-to-systole displacements; right carotid asymmetries; second-order statistical measures; stenosis degrees; stroke; systemic factors; textural feature; texture analysis; therapeutic approaches; tissue composition; transient ischaemic attack; ultrasound image-based features; Atherosclerosis; Carotid arteries; Diseases; Image resolution; Standards; Ultrasonic imaging;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical and Health Informatics (BHI), 2014 IEEE-EMBS International Conference on
Conference_Location :
Valencia
Type :
conf
DOI :
10.1109/BHI.2014.6864455
Filename :
6864455
Link To Document :
بازگشت