TY - JOUR
T1 - Random forest-based similarity measures for multi-modal classification of Alzheimer's disease
AU - Gray, Katherine R.
AU - Aljabar, Paul
AU - Heckemann, Rolf A.
AU - Hammers, Alexander
AU - Rueckert, Daniel
N1 - Funding Information:
The authors would like to thank Igor Yakushev (University of Mainz, Germany) for provision of the reference cluster image used for FDG-PET intensity normalisation, as well as Casper Nielsen and Kelvin Leung (Dementia Research Centre, University College London) for provision of the MIDAS brain masks used for MRI anatomical segmentation. Katherine R. Gray received a studentship from the Engineering and Physical Sciences Research Council. Rolf A. Heckemann was supported by a research grant from the Dunhill Medical Trust. The Image Registration Toolkit was used under licence from Ixico Ltd.
PY - 2013/1/15
Y1 - 2013/1/15
N2 - Neurodegenerative disorders, such as Alzheimer's disease, are associated with changes in multiple neuroimaging and biological measures. These may provide complementary information for diagnosis and prognosis. We present a multi-modality classification framework in which manifolds are constructed based on pairwise similarity measures derived from random forest classifiers. Similarities from multiple modalities are combined to generate an embedding that simultaneously encodes information about all the available features. Multi-modality classification is then performed using coordinates from this joint embedding. We evaluate the proposed framework by application to neuroimaging and biological data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Features include regional MRI volumes, voxel-based FDG-PET signal intensities, CSF biomarker measures, and categorical genetic information. Classification based on the joint embedding constructed using information from all four modalities out-performs the classification based on any individual modality for comparisons between Alzheimer's disease patients and healthy controls, as well as between mild cognitive impairment patients and healthy controls. Based on the joint embedding, we achieve classification accuracies of 89% between Alzheimer's disease patients and healthy controls, and 75% between mild cognitive impairment patients and healthy controls. These results are comparable with those reported in other recent studies using multi-kernel learning. Random forests provide consistent pairwise similarity measures for multiple modalities, thus facilitating the combination of different types of feature data. We demonstrate this by application to data in which the number of features differs by several orders of magnitude between modalities. Random forest classifiers extend naturally to multi-class problems, and the framework described here could be applied to distinguish between multiple patient groups in the future.
AB - Neurodegenerative disorders, such as Alzheimer's disease, are associated with changes in multiple neuroimaging and biological measures. These may provide complementary information for diagnosis and prognosis. We present a multi-modality classification framework in which manifolds are constructed based on pairwise similarity measures derived from random forest classifiers. Similarities from multiple modalities are combined to generate an embedding that simultaneously encodes information about all the available features. Multi-modality classification is then performed using coordinates from this joint embedding. We evaluate the proposed framework by application to neuroimaging and biological data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Features include regional MRI volumes, voxel-based FDG-PET signal intensities, CSF biomarker measures, and categorical genetic information. Classification based on the joint embedding constructed using information from all four modalities out-performs the classification based on any individual modality for comparisons between Alzheimer's disease patients and healthy controls, as well as between mild cognitive impairment patients and healthy controls. Based on the joint embedding, we achieve classification accuracies of 89% between Alzheimer's disease patients and healthy controls, and 75% between mild cognitive impairment patients and healthy controls. These results are comparable with those reported in other recent studies using multi-kernel learning. Random forests provide consistent pairwise similarity measures for multiple modalities, thus facilitating the combination of different types of feature data. We demonstrate this by application to data in which the number of features differs by several orders of magnitude between modalities. Random forest classifiers extend naturally to multi-class problems, and the framework described here could be applied to distinguish between multiple patient groups in the future.
KW - Alzheimer's disease
KW - Cerebrospinal fluid biomarkers
KW - Genetics
KW - Magnetic resonance imaging
KW - Mild cognitive impairment
KW - Multi-modality classification
KW - Positron emission tomography
UR - http://www.scopus.com/inward/record.url?scp=84868215748&partnerID=8YFLogxK
U2 - 10.1016/j.neuroimage.2012.09.065
DO - 10.1016/j.neuroimage.2012.09.065
M3 - Article
C2 - 23041336
AN - SCOPUS:84868215748
SN - 1053-8119
VL - 65
SP - 167
EP - 175
JO - NeuroImage
JF - NeuroImage
ER -