Segmentation of Dual-Echo MR Head Data
Multiecho acquisition in Magnetic Resonance Imaging (MRI) allows better discrimination of different tissue types and anatomical functional units because specific characteristics are enhanced in multiple spectral channels.
Multivariate statistical classification techniques are applied to dual-echo MR data to segment volume head data into anatomical objects and tissue categories (brain white and gray matter, ventricular system, outer csf space, bone structure, tumor).
To overcome the sensitivity of voxel-based classification to noise we applied a preprocessing technique based on anisotropic diffusion. This preprocessing increases the separability of clusters. We illustrate the robustness of supervised classification with the segmentation of a series of MR head data in a research study.
For a given set of MR parameters we show that the configuration of clusters in feature space is comparable between studies. This allows us to develop an automated clustering technique that considers a priori knowledge about cluster attributes and their configuration in feature space. The automated classification technique omits subjective criteria in the training stage of supervised classification and yields reproducible segmentation results.
Combined 3D views of multiple anatomical objects are shown. They clarify the perception of 3D relationship and highlight the locations and types of structural abnormalities.
KeywordsSupervise Classification Ventricular System Anisotropic Diffusion Filter Supervise Classification Technique Computer Assist Radiology
Unable to display preview. Download preview PDF.
- Vannier, M.W., Butterfield, R.L., Jordan, D., et.al. Multispectral Analysis of Magnetic Resonance Images, Radiology 154, 1985, pp. 221–224Google Scholar
- Vannier, M.W. et.al., Validation of Magnetic Resonance Imaging (MRI) Multispectral Tissue Classification, Proc. of 9th Conf. on Pattern Recognition, Rome, Italy, Nov. 1988, pp. 1182–1186Google Scholar
- Merickel, M.B. et.al., Multispectral Pattern Recognition of MR Imagery for the Noninvasive Analysis of Atherosclerosis, Proc. of 9th Int. Conf. on Pattern recognition, Rome, Italy, Nov. 1988, pp. 1192–1197Google Scholar
- G. Gerig, W. Kuoni, R. Kikinis and O. Kiibler, Medical Imaging and Computer Vision: An integrated approach for diagnosis and planning, 11. DAGM-Symposium Mustererkennung, 2.-4. Oct. 1989, Informatik Fachberichte IFB 219, Springer Verlag Berlin, pp. 425–432Google Scholar
- G. Gerig, R. Kikinis, O. Kübler, Significant Improvement of MR Image Data Quality using Anisotropie Diffusion Filtering, Technical Report BIWI-TR-112, Institute for Communication Technology, Image Science Division, ETH-Zurich, Switzerland, March 1990Google Scholar
- Duda R.O. and Hart, P.E., Pattern Classification and Scene Analysis, by John Wiley & Sons,Inc, 1973Google Scholar
- R. Kikinis, F. Jolesz, G. Gerig, T. Sandor, H. Cline, W. Lorensen, M. Halle, St. Benton, 3D Morphometric and Morphologic Information derived from Clinical Brain MR Images, in: 3D Imaging in Medicine, Höhne K.H., Fuchs H., Pizer St.M. editors, NATO ASI Series, Serie F: Computer and Systems Science Vol. 60, Springer-Verlag, Proceedings of the NATO Advanced Research Workshop, 25–29 June 1990, Travemiinde, FRG, pp. 441–454Google Scholar