An efficient Computer Aided Diagnosis (CAD) system for classification of carotid atherosclerosis into symptomatic or asymptomatic classes would be a useful adjunct tool that helps vascular surgeons in deciding the appropriate treatment regime for the patient. In this chapter, we present a patented CAD system called Atheromatic™ for symptomatic versus asymptomatic plaque classification in carotid ultrasound images. The system consists of two steps: (1) extraction of grayscale features based on Discrete Wavelet Transform (DWT) and averaging algorithms and (2) classification of the selected features using Support Vector Machine (SVM) classifier. The system was developed and evaluated using a database consisting of 150 asymptomatic and 196 symptomatic plaque regions. The gold standard ground as to whether the plaque was symptomatic or asymptomatic was decided based on the presence or absence of symptoms. Stratified threefold cross-validation data resampling protocol was adapted for developing and testing the classifier. A high accuracy of 83.7% was obtained using the SVM classifier with a polynomial kernel of order 2.
Symptomatic Versus Asymptomatique Plaque Classification in Carotid Ultrasound / Acharya, U. R.; Faust, O.; Sree, V. S.; Molinari, Filippo; Saba, L.; Nicolaides, A.; Suri, J. S. - In: Multi-Modality Atherosclerosis Imaging and Diagnosis / Luca Saba, João Miguel Sanches, Luís Mendes Pedro, Jasjit S. Suri. - [s.l] : New York, Springer Science+Business Media, 2014. - ISBN 9781461474241. - pp. 399-408 [10.1007/978-1-4614-7425-8_29]
Symptomatic Versus Asymptomatique Plaque Classification in Carotid Ultrasound
MOLINARI, FILIPPO;
2014
Abstract
An efficient Computer Aided Diagnosis (CAD) system for classification of carotid atherosclerosis into symptomatic or asymptomatic classes would be a useful adjunct tool that helps vascular surgeons in deciding the appropriate treatment regime for the patient. In this chapter, we present a patented CAD system called Atheromatic™ for symptomatic versus asymptomatic plaque classification in carotid ultrasound images. The system consists of two steps: (1) extraction of grayscale features based on Discrete Wavelet Transform (DWT) and averaging algorithms and (2) classification of the selected features using Support Vector Machine (SVM) classifier. The system was developed and evaluated using a database consisting of 150 asymptomatic and 196 symptomatic plaque regions. The gold standard ground as to whether the plaque was symptomatic or asymptomatic was decided based on the presence or absence of symptoms. Stratified threefold cross-validation data resampling protocol was adapted for developing and testing the classifier. A high accuracy of 83.7% was obtained using the SVM classifier with a polynomial kernel of order 2.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2542504
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo