Research Projects of the Division of Medical and Biological Informatics
Model-based segmentation
Segmentation of the abdomen based on CT data facilitates a detailed 3D reconstruction for operation planning (here using the pancreas as an example).
| © dkfz.de Segmentation, i.e. the partitioning of images into meaningful regions, is the basis for many applications in the area of medical image processing, such as surgery planning, radiation therapy and image-assisted diagnosis. In order to reduce the work load on the radiologists and technical assistants involved, the segmentation process should be automated to the greatest extent possible. At the same time, the option of quickly and easily correcting by hand automatically extracted regions needs to be retained. In view of the continually increasing resolution of imaging modalities such as CT and MRT and, resulting from this, the constant rise in the number of individual tomographic images, we are developing segmentation techniques that may be directly applied to 3D data
Selected Publications
- [P15-10] Schwarz T, Heimann T, Lossnitzer D, Mohrhardt C, Steen H, Rietdorf U, Wolf I, Meinzer HP. Multi-Object Segmentation using Coupled Shape Space Models. Proc. SPIE Vol. 7623, Medical Imaging 2010: Image Processing, 762332; DOI: 10.1117/12.844192.
- [P14-08] Schwarz T, Heimann T, Tetzlaff R, Rau AM, Wolf I, Meinzer HP. Interactive Surface Correction for 3D Shape Based Segmentation. Proc. SPIE Int. Soc. Opt. Eng. 6914, 69143O (2008), DOI:10.1117/12.770350.
- [P17-07] Heimann T, Münzing S, Meinzer HP, Wolf I. A Shape-Guided Deformable Model with Evolutionary Algorithm Initialization for 3D Soft Tissue Segmentation. In Karssemeijer N, Lelieveldt B (eds). IPMI 2007, LNCS 4584. Berlin: Springer (2007) 1-12.
