Computing Publications

Publications Home » Random Forest-Based Manifold Lear...

Random Forest-Based Manifold Learning for Classification of Imaging Data in Dementia

Katherine Gray, Paul Aljabar, R. A. Heckemann, A. Hammers, Daniel Rueckert

Conference or Workshop Paper
MLMI Workshop, MICCAI 2011
2011
Lecture Notes in Computer Science
Volume 7009
pp.159–166
Springer
DOI 10.1007/978-3-642-24319-6_20
Abstract

Neurodegenerative disorders are characterized by changes in multiple biomarkers, which may provide complementary information for diagnosis and prognosis. We present a framework in which proximities derived from random forests are used to learn a low-dimensional manifold from labelled training data and then to infer the clinical labels of test data mapped to this space. The proposed method facilitates the combination of embeddings from multiple datasets, resulting in the generation of a joint embedding that simultaneously encodes information about all the available features. It is possible to combine different types of data without additional processing, and we demonstrate this key feature by application to voxel-based FDG-PET and region-based MR imaging data from the ADNI study. Classification based on the joint embedding coordinates out-performs classification based on either modality alone. Results are impressive compared with other state-of-the-art machine learning techniques applied to multi-modality imaging data.

Notes

Winner of the best paper award for the workshop

BibTEX file for the publication
 

pubs.doc.ic.ac.uk: built & maintained by Ashok Argent-Katwala.