TY - GEN
T1 - Discriminative fusion of multiple brain networks for early mild cognitive impairment detection
AU - Wang, Qi
AU - Zhan, Liang
AU - Thompson, Paul M.
AU - Dodge, Hiroko H.
AU - Zhou, Jiayu
N1 - Funding Information:
Part by NIH ENIGMA Center grant U54 EB020403, supported by the Big Data to Knowledge (BD2K) Centers of Excellence program, and ONR grant N00014-14-1-0631
Publisher Copyright:
© 2016 IEEE.
PY - 2016/6/15
Y1 - 2016/6/15
N2 - In neuroimaging research, brain networks derived from different tractography methods may lead to different results and perform differently when used in classification tasks. As there is no ground truth to determine which brain network models are most accurate or most sensitive to group differences, we developed a new sparse learning method that combines information from multiple network models. We used it to learn a convex combination of brain connectivity matrices from 9 different tractography methods, to optimally distinguish people with early mild cognitive impairment from healthy control subjects, based on the structural connectivity patterns. Our fused networks outperformed the best single network model, Probtrackx (0.89 versus 0.77 cross-validated AUC), suggesting its potential for numerous connectivity analysis.
AB - In neuroimaging research, brain networks derived from different tractography methods may lead to different results and perform differently when used in classification tasks. As there is no ground truth to determine which brain network models are most accurate or most sensitive to group differences, we developed a new sparse learning method that combines information from multiple network models. We used it to learn a convex combination of brain connectivity matrices from 9 different tractography methods, to optimally distinguish people with early mild cognitive impairment from healthy control subjects, based on the structural connectivity patterns. Our fused networks outperformed the best single network model, Probtrackx (0.89 versus 0.77 cross-validated AUC), suggesting its potential for numerous connectivity analysis.
KW - Brain Connectome
KW - Classification
KW - Discriminative Fusion
KW - Magnetic Resonance Imaging
KW - Mild Cognitive Impairment
UR - http://www.scopus.com/inward/record.url?scp=84978372340&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84978372340&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2016.7493332
DO - 10.1109/ISBI.2016.7493332
M3 - Conference contribution
AN - SCOPUS:84978372340
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 568
EP - 572
BT - 2016 IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
T2 - 2016 IEEE 13th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2016
Y2 - 13 April 2016 through 16 April 2016
ER -