Castrillon JG, Ahmadi A, Navab N, Richiardi J (2015)
Publication Type: Conference contribution
Publication year: 2015
Publisher: IEEE Computer Society
Book Volume: 2015-April
Pages Range: 608-612
Conference Proceedings Title: Conference Record - Asilomar Conference on Signals, Systems and Computers
Event location: Pacific Grove, CA, USA
ISBN: 9781479982974
DOI: 10.1109/ACSSC.2014.7094518
Neuroimaging data collection is very costly, and acquisition is commonly distributed across multiple sites. However, factors such as different noise characteristics or inhomogeneities make it difficult to successfully combine multi-site functional imaging data. Here, we show that the distribution of signal quality measures across scanners can be significantly different, and that this will have an impact on correlation estimators necessary for computing functional connectivity graphs as well as topological features extracted from the graphs. We propose to find a stable subspace by using a discriminative projection that does not only minimise site differences, but also preserves discriminative class information. We compare our method with the 'regressing-out' approach in a cross-validation setting and show that regressing out can yield very poor results.
APA:
Castrillon, J.G., Ahmadi, A., Navab, N., & Richiardi, J. (2015). Learning with multi-site fMRI graph data. In Michael B. Matthews (Eds.), Conference Record - Asilomar Conference on Signals, Systems and Computers (pp. 608-612). Pacific Grove, CA, USA: IEEE Computer Society.
MLA:
Castrillon, J. Gabriel, et al. "Learning with multi-site fMRI graph data." Proceedings of the 48th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015, Pacific Grove, CA, USA Ed. Michael B. Matthews, IEEE Computer Society, 2015. 608-612.
BibTeX: Download