Dysfunction of brain structural and functional connectivity is increasingly being recognized as playing an important role in many brain disorders. Diffusion tensor imaging (DTI) and functional magnetic resonance (fMR) imaging are widely used to infer structural and functional connectivity, respectively. How to combine structural and functional connectivity patterns for predictive modeling is an important, yet open, problem. We propose a new method, called Bayesian prediction based on multidimensional connectivity profiling (BMCP), to distinguish subjects at the individual level based on structural and functional connectivity patterns. BMCP combines finite mixture modeling and Bayesian network classification. We demonstrate its use in distinguishing young and elderly adults based on DTI and resting-state fMR data.
Tzourio-MazoyerN, LandeauB, PapathanassiouD. Automated anatomical labeling of activations in SPM using a macroscopic anatomical panellation of the MNI MRI single-subject brain.NeuroImage2002; 15 (1): 273-289. doi: 10.1006/nimg.2001.0978.
11.
KorgaonkarMS, CooperNI, WilliamsLM. Mapping inter-regional connectivity of the entire cortex to characterize major depressive disorder: a whole-brain diffusion tensor imaging tractography study.Neuroreport2012; 23 (9): 566-571. doi: 10.1097/WNR.0b013e3283546264.
12.
BehrensTE, BergHI, IbabdiS. Probabilistic diffusion tractography with multiple fibre orientations: What can we gain?Neuroimage2007; 34(1): 144-155. doi: 10.1016/j.neuroimage.2006.09.018.
13.
RosnerB. Fundamentals of biostatistics, Seventh edition. Stamford: Brooks/Cole, 2010.
14.
FraleyC, and RafteryAE. Model-based clustering, discriminant analysis, and density estimation.J Am Stat Assoc2002; 97 (458): 611-631. doi: 10.1198/016214502760047131.
DempsterAP, LairdNM, and RubinDB. Maximum likelihood from incomplete data via the EM algorithm.J Royal Stat Soc1977; 39 (1): 1-38.
17.
SchwarzGEstimating the dimension of a model.Ann Stat1978; 6(2): 461-464. doi: 10.1214/aos/1176344136.
18.
PearlI. Probabilistic Reasoning in Intelligent Systems.Morgan Kaufmann; 1988.
19.
DudaR, HartR, and StorkDG. Pattern classification, second edition. New York: Wiley, 2001.
20.
ChenR, HerskovitsEH. A Bayesian network classifier with inverse tree structure for voxel-wise MR image analysis. In the 11th conference of SIGKDD; 2005.
21.
ChenR, and HerskovitsEH. Clinical diagnosis based on bayesian classification of functional magnetic-resonance data.Neuroinformatics2007; 5 (3): 178-188. doi: 10.1007/S12021-007-0007-2.
22.
ChenR, and HerskovitsEH. Graphical-model based morphometric analysis.IEEE Trans Med Imaging2005; 24 (10): 1237-1248. doi: 10.1109/TMI.2005.854305.
23.
CooperGF, and HerskovitsEH. A Bayesian method for the induction of probabilistic networks from data.Machine Learning1992; 9 (4): 309-347. doi: 10.1023/A:1022649401552.
24.
HeckermanD, and ChickeringDM. Learning Bayesian Networks: The Combination of Knowledge and Statistical Data.Machine Learning1995; 20: 197-243.
25.
FredA, and JainAK. Combining multiple clusterings using evidence accumulation.IEEE Trans Pattern Anal Mach Intell2005; 27 (6): 835-850. doi: 10.1109/TPAMI.2005.113.
26.
BreimanL. Random Forests.Machine Learning2001; 45 (1): 5-32. doi: 10.1023/A:1010933404324.
FreundY, and SchapireR. Experiments with a new boosting algorithm.Machine Learning: Proceedings of the Thirteenth International Conference1996; 148-156.
29.
RashI. An empirical study of the naive Bayes classifier. IJCAI 2001 Workshop on Empirical Methods in Artificial Intelligence; 2001.
30.
WyattCL, LaurientiPJ. Age related atrophy dataset.Virginia Tech and Wake Forest University School of Medicine.
31.
FerreiraLK, and BusattoGF. Resting-state functional connectivity in normal brain aging.Neurosci Biobehav Rev2013; 37 (3): 384-400. doi: 10.1016/j.neubiorev.2013.01.017.
RasmussenCE, and WilliamsCKI. Gaussian processes for machine learning.Cambridge: MA: The MIT Press, 2006.
34.
VapnikVN. The nature of statistical learning theory.Springer-Verlag; 1995. doi: 10.1007/978-1-4757-2440-0.
35.
ChenR, and HerskovitsEH. Machine-learning techniques for building a diagnostic model for very mild dementia.Neuroimage2010; 52 (1): 234-244. doi: 10.1016/j.neuroimage.2010.03.084.
36.
ChenR, and HerskovitsEH. Network analysis of mild cognitive impairment.Neurolmage2006; 29 (4): 1252-1259. doi: 10.1016/j.neuroimage.2005.08.020.
37.
MoralesDA, Vives-GilabertY, Gomez-AnsonB. Predicting dementia development in Parkinson's disease using Bayesian network classifiers.Psychiatry Res2013; 213 (2): 92-98. doi: 10.1016/j.pscychresns.2012.06.001.
38.
BrookesMJ, HaleJR, ZumerJM. Measuring functional connectivity using MEG: methodology and comparison with fcMRI.Neuroimage2011; 56 (3): 1082-1104. doi: 10.1016/j.neuroimage.2011.02.054.
