Tissue characterisation of coronary plaques using sparse feature vectors

In a diagnosis of acute coronary syndrome, the tissue characterisation of coronary plaque is very important. A novel method is proposed for the tissue characterisation of coronary plaque by using a sparse coding. In the proposed method, an intravascular ultrasound backscattered radiofrequency (RF) signal is coded by the sparse coding, and the code patterns are used for the classification of a target tissue. Experiments have been performed to classify the tissues of coronary plaque into fibrous or lipid tissue by using these RF signals, which are observed from a rabbit coronary artery. The effectiveness of the proposed method has been verified by comparative studies with the conventional integrated backscatter analysis.