Information theoretic analysis of neuroimaging signals: Hands-on tutorial, Centre for Human Brain Health, Thursday, 10. October 2019

Dr. Robin Ince (University of Glasgow) will give a talk at the Centre for Human Brain Health (CHBH) seminar on the 10th October at 13:00-14:00. From his talk (seminar), you will learn the theoretical background and some applications of Information Theoretical measures in various neuroimaging techniques such as fMRI and MEG/EEG. After his talk, we are planning a hands-on session to give a more practical overview (14:00-17:00).
 
Brief Intro about Hands-on Session
Information theoretic methods have great potential for neuroimaging but they can be difficult to implement in practise. The Gaussian Copula Mutual Information (GCMI) estimator [2,3] provides a simple, robust and computationally efficient way to calculate many information theoretic quantities on typical neuroimaging data. A major advantage is that this can cover a wide range of statistical comparisons – involving discrete, continuous and multidimensional stimulus and response variables – while producing effect sizes on a common meaningful scale [1]. GCMI can be used to probe representational interactions e.g. redundant coding between responses at different time points or different sensors, or between different frequency bands, phase vs power etc. In this tutorial, we will provide a hands-on, accessible introduction to applying GCMI to neuroimaging data to perform non-parametric statistical inference and to quantify representational interactions.
[1]
[2]
[3] Ince et al. (2016) HBM
[4] Application in MEG: 

Please sign up!
We would like to know the number of attendees for room booking etc. More information (detailed program, preparations for hands-on, room etc.) will follow closer to the date and will be sent to registered attendees.
Any queries, please feel free to contact the organiser.
Please save the date in your diary!

Thursday, 10. October 2019, Centre for Human Brain Health, Information theoretic analysis of neuroimaging signals: Hands-on tutorial

Find more interesting events
Get event recommendations based on your Facebook taste. Get it now!Show me the suitable events for meNot now