Plaque classification using sparse features of IVUS RF signal for the diagnosis of arteriosclerosis
Proc. of International Conference on Applied Computer and Applied Computational Science (ACACOS2011), pp.99-103
S. Furukawa, E. Uchino, T. Azetsu, N. Suetake, T. Hiro, M. Matsuzaki
Oral presentation (general)
This paper proposes a tissue characterization method for coronary plaque by using a sparse coding. The sparse coding can efficiently represent a signal by a few basis functions extracted by learning. In the proposed method, the radio frequency (RF) signal obtained by the intravascular ultrasound (IVUS) method is expressed by a linear combination of the basis functions learned by the sparse coding, and the coefficient 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 and frequency analysis.