Classifying normal and abnormal vascular tissues using photoacoustic signals

In this paper a new method is proposed to classify vascular tissues in the range from normal to different degrees of abnormality based on the Photo-Acoustic (PA) signals generated by different categories of vasculatures. The classification of the vasculatures is achieved based on the statistical features of the photoacoustic radiofrequency (RF) signals such as energy, variance, and entropy in the wavelet domain. A feature vector for each category of vasculature is provided and the distance between feature vectors are computed as the measure of similarity between vasculatures. The distances are mapped in two-dimensional space depicting the proximities of the different categories of the vasculatures. The method proposed in this paper can help both detecting abnormal tissues and monitoring the treatment progress by measuring the similarity between vascular tissues in different stages of treatment. The method is applied to simulated data as well as in vivo data from tumor bearing mice to detect cancer treatment effects.