Okito Yamashita, Department head of CBI in ATR

  • Affiliation
ATR  Neural Information Analysis Laboratories
Department of Computational Brain Imaging

  • Research Interest
Applications of Bayesian inference to noninvasive brain imaging data such as fMRI, MEG, EEG and NIRS.

Main applications are
Diffusion optical tomography (2010- )
- MEG/EEG Source Localization Problem  (2001-)
- Brain Computer Interface (2007-)
- fMRI decoding.  (2005-)

Main tools  in statistics are

- Time Series Analysis (AR model., Hidden Markov model, State Space model)
- Sparse Bayesian Learning (Automatic Relevance Determination Priors, Variational Bayesian algorithm).
  • Contact
ADRESS : 2-2-2, Hikaridai, Seikacho, Sorakugun, Kyoto 619-0288, Japan
TELL  : +81-774-95-1073
FAX    : +81-774-95-1259
E-mail : oyamashi(at)atr.jp
  • Employment History
2013 April -
ATR, Neural Information Analysis Laboratories, Deparment head
  • Brain network dynamics estimation with the multimodal integration approach
2010 April - 2013 March
ATR, Neural Information Analysis Laboratories, Senior Researcher
  • Cross-freuqency coupling of ECoG data (data analysis, method proposal)
  • Diffusion optical tomography (experimental validation, data analysis, toolbox)
2004 Semptember - 2009 March
ATR, Computational Neuroscience Laboratories, Researcher
  • Source localization method with MEG/EEG-fMRI mutlimodal integration (method extension, toolbox)
  • EEG/NIRS combined brain machine interface (data analysis)
  • Sparse classification for fMRI decoding and brain machine interface (method proposal, toolbox)
2004 April-August Institute of statistical mathematics,  Part-time researcher
  • dynamic source localization extended to non-stationary case (method proposal)
  • Education
Ph.D. in Statitics, 2004   The Graduate University for Advanced Studies, Department of Statistical Science
  • Application of the statistical timeseries analysis to neuroimaging data 
    • Effective connectivity estimation of fMRI   (Granger causality analysis)
    • Dynamic source localization of EEG         (High dimensional Kalman filter)
Title of Ph.D. thesis : Dynamical Inverse Problem and Causality Analysis of fMRI Data [PDF file]
M.A. in Engineering, 2001   The University of Tokyo, Faculty of Engineering, Department of Mathematical Engineering and Information Physics
  • Parameter estimation problem of  non-linear AR model 
B.A. in Engineering, 1999   The University of Tokyo, Faculty of Engineering, Department of Mathematical Engineering and Information Physics
  • Observation noise reduction of chaotic dynamical model