A multiscale dynamic programming procedure for boundary detection in ultrasonic artery images

摘要: Ultrasonic measurements of human carotid and femoral artery walls are conventionally obtained by manually tracing interfaces between tissue layers. The drawbacks this method the interobserver variability inefficiency. Here, authors present a new automated which reduces these problems. By applying multiscale dynamic programming (DP) algorithm, approximate vessel wall positions first estimated in coarse-scale image, then guide detection boundaries fine-scale image. In both cases, DP is used for finding global optimum cost function. function weighted sum terms, fuzzy expression forms, representing image features geometrical characteristics interfaces. weights adjusted training procedure using expert tracings. Operator interventions, if needed, also take effect under framework optimality. This amount intervention and, hence, due to subjectiveness. incorporating knowledge experience, algorithm becomes more robust. A thorough evaluation clinical environment shows that evidently decreased so overall analysis time. conclude can replace manual leads an improved performance.