The recent availability of real-time three-dimensional echocardiography offers a convenient, low-cost alternative for detection and diagnosis of heart pathologies. However, a complete description of the heart can be obtained only by combining the information provided by different acoustic windows. We present a new method for compounding 3D ultrasound scans acquired from different views. The method uses multiscale information about local structure definition and orientation to weight the contributions of the images. We propose to use image phase to obtain these image characteristics while keeping invariance to image contrast. The monogenic signal provides a convenient, integrated approach for this purpose. We have evaluated our algorithm on synthetic images and heart scans from volunteers, showing it provides a significant improvement in image quality when compared to traditional compounding methods. © Springer-Verlag Berlin Heidelberg 2005.
CITATION STYLE
Grau, V., & Noble, J. A. (2005). Adaptive multiscale ultrasound compounding using phase information. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3749 LNCS, pp. 589–596). https://doi.org/10.1007/11566465_73
Mendeley helps you to discover research relevant for your work.