Thumbnail Image

Volume Segmentation of 3-dimensional Images

Fiorio, Christophe; Gustedt, Jens

Inst. Mathematik

We present a practical method to segment large medical images that takes the whole 3-dimensional structure into account. We use a Union-Find data structure to record and maintain the necessary information during the segmentation process. Due to the large data size, we are forced to divide our process in two parts: a "weak segmentation" of the individual sections and a global integration of all the data. This method shows good results on computer tomographies.