Multi-modal brain fingerprinting: A manifold approximation based framework

• Published in: NeuroImage (Orlando, Fla.) Vol. 183; pp. 212 - 226
• Main Authors: Kumar, Kuldeep, Toews, Matthew, Chauvin, Laurent, Colliot, Olivier, Desrosiers, Christian
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.12.2018; Elsevier Limited; Elsevier
• Subjects: Adult; Bag-of-Features; Bioengineering; Brain; Brain - anatomy & histology; Brain - diagnostic imaging; Brain - physiology; Brain fingerprinting; Brain mapping; Cerebral hemispheres; Cohort Studies; Computer Science; Computer Vision and Pattern Recognition; Connectome - methods; Diffusion Magnetic Resonance Imaging - methods; Embedding; Engineering Sciences; Female; Fingerprinting; Functional magnetic resonance imaging; Functional Neuroimaging - methods; HCP Twin data; Humans; Identification; Image Processing; Imaging; Individuality; Life Sciences; Magnetic Resonance Imaging - methods; Male; Manifold approximation; Medical Imaging; Multi-modal data; Neurobiology; Neurons and Cognition; Schizophrenia; Siblings; Signal and Image processing; sMRI-dMRI-rfMRI; Twins; Young Adult; United Kingdom > UK; Manifold approximation; Brain fingerprinting; Multi-modal data; Bag-of-Features; sMRI-dMRI-rfMRI; HCP Twin data; Brain fingerprint; Structural MRI; Hemisphere asymmetry; Manifold; Multi-modal; rfMRI netmats; Diffusion MRI; MAP
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2018.08.006

This work presents an efficient framework, based on manifold approximation, for generating brain fingerprints from multi-modal data. The proposed framework represents images as bags of local features which are used to build a subject proximity graph. Compact fingerprints are obtained by projecting this graph in a low-dimensional manifold using spectral embedding. Experiments using the T1/T2-weighted MRI, diffusion MRI, and resting-state fMRI data of 945 Human Connectome Project subjects demonstrate the benefit of combining multiple modalities, with multi-modal fingerprints more discriminative than those generated from individual modalities. Results also highlight the link between fingerprint similarity and genetic proximity, monozygotic twins having more similar fingerprints than dizygotic or non-twin siblings. This link is also reflected in the differences of feature correspondences between twin/sibling pairs, occurring in major brain structures and across hemispheres. The robustness of the proposed framework to factors like image alignment and scan resolution, as well as the reproducibility of results on retest scans, suggest the potential of multi-modal brain fingerprinting for characterizing individuals in a large cohort analysis. •A multi-modal brain fingerprinting framework is proposed.•Comparison of fingerprints based on T1w, T2w, dMRI, rfMRI, and their combinations.•Large-scale analysis and validation of framework using a cohort of 945 HCP subjects.•Fingerprints of monozygotic twins are more similar than dizygotic twins or non-twins.•Framework is computationally efficient; robust to image alignment & scan resolution.