Tissue characterization of coronary plaque by sparse coding with use of intravascular ultrasound method

Proc. of 2009 International Workshop on Nonlinear Circuits and Signal Processing (NCSP2009), pp.388-391

NCSP2009

Hawaii

2009/03

T. Azetsu, E. Uchino, S. Furukawa, N. Suetake, G. Hashimoto, T. Hiro, M. Matsuzaki

Tissue characterization of coronary plaque is important for a diagnosis of acute coronary syndrome (ACS). This paper proposes a novel method for the tissue characterization of the coronary plaque by sparse coding. The sparse coding is a feature extraction method based on the perceptual system of the mammalian visual cortex. In the proposed method, the radiofrequency (RF) signal obtained by the intravascular ultrasound (IVUS) method is expressed by the basis functions extracted by the sparse coding, and the code patterns of the basis
functions are used for the tissue characterization. The effectiveness of the proposed method has been verified by comparing it with the conventional integrated backscatter (IB) analysis.