A feature selection framework for white matter fiber clustering based on normalized cuts
Koppers S, Hebisch C, Merhof D (2017)
Publication Type: Conference contribution
Publication year: 2017
Journal
Publisher: Kluwer Academic Publishers
Pages Range: 110-115
Conference Proceedings Title: Informatik aktuell
Event location: Berlin, DEU
ISBN: 9783662494646
DOI: 10.1007/978-3-662-49465-3_21
Abstract
Due to its ability to automatically identify spatially and functionally related white matter fiber bundles, fiber clustering has the potential to improve our understanding of white matter anatomy. The normalized cuts (NCut) criterion has proven to be a suitable method for clustering fiber tracts. In this work, we show that the NCut value can be used for unsupervised feature selection as a measure for the quality of clustering. We further present a method how feature selection can be improved by penalizing spatially illogical clustering results, which is achieved by employing the Silhouette index for a fixed set of geometric features.
Involved external institutions
Germany (DE)How to cite
APA:
Koppers, S., Hebisch, C., & Merhof, D. (2017). A feature selection framework for white matter fiber clustering based on normalized cuts. In Thomas M. Deserno, Heinz Handels, Thomas Tolxdorff, Hans-Peter Meinzer (Eds.), Informatik aktuell (pp. 110-115). Berlin, DEU: Kluwer Academic Publishers.
MLA:
Koppers, Simon, Christoph Hebisch, and Dorit Merhof. "A feature selection framework for white matter fiber clustering based on normalized cuts." Proceedings of the Workshops on Image processing for the medicine, 2016, Berlin, DEU Ed. Thomas M. Deserno, Heinz Handels, Thomas Tolxdorff, Hans-Peter Meinzer, Kluwer Academic Publishers, 2017. 110-115.
BibTeX: Download