Course Program

Part 1 (day 1, theoretical workshop): Theoretical foundations for advanced EEG/MEG source analysis and connectivity: What are the main issues? Where there is a consensus? Where there are still controversies?

A set of brief notes with important equations and key references will be provided to all participants. The lectures will describe the material with as few equations as possible, but no fewer than necessary, emphasizing the meaning of the various terms.

Morning Session Part A: 8:50 - 10:15

Introductory remarks by the organizers Andreas A. Ioannides and Fabio Babiloni (~10 minutes).

Lecture 1: Basic concepts and their foundations (~30 minutes). 
Andreas A. Ioannides

  • What do EEG and MEG sensors measure and what are physical laws that describe the phenomena?
  • Source models and conductivity media.
  • Quasi-static approximation.
  • Lead fields.
  • Forward and inverse problem.

Lecture 2: Basic issues for the estimation of the sources with non-invasive EEG measurements (~45 minutes). 
Fabio Babiloni

  • Source models: single / multiple dipoles and distributed sources.
  • Volume conductor models: plane, spherical and realistic.
  • Forward problem: analytical, FEM and BEM approaches.
  • Inverse problem: discrete and extended source localization, basic issues.
  • Lead field concept.
  • Digitization of electrodes: how and how fast.

Coffee break: 10:15 - 10:30

Morning Session Part B: 10:30 - 12:30

Lecture 3: Mathematics for the forward and inverse problem of EEG (~45 minutes). 
Rolando Grave de Peralta

  • Forward and inverse problems for distributed sources: linear or non-linear approaches? What can be gained by non-linear approaches?
  • Popular distributed source estimation systems (LORETA, LAURA).
  • Mathematical and empirical instruments for the estimation of the quality of the source localization: resolution kernels, dipolar localization errors and etc.
  • What are the limits of localization and what determines them?

Lecture 4: Mathematics for the forward and inverse problems of MEG (~45 minutes).
Andreas A. Ioannides

  • Forward and inverse problems in MEG.
  • Why do simple models for conductivity work?
  • The role of symmetry breaking for accurate localization of deep sources.
  • Linear and non-linear algorithms for the inverse problem.
  • What are the limits of localization and what determines them in practice?

Round table discussion 1 (~30 minutes).
Chair: Andreas A. Ioannides / Fabio Babiloni

  • Comparison of EEG and MEG with each other and other techniques.
  • Mathematical obstacles (non-uniqueness) and real performance.
  • Limits of localization, what determines them and how to test?
  • General discussion.

Lunch break: 12:30 - 14:00

Afternoon Session: 14:00 - 17:00

Lecture 5: Basis of the estimation of connectivity: general principles and measures of connectivity (~40 minutes).
Laura Astolfi

  • Different definitions of connectivity: Granger causality and other measures, linear and non-linear connectivity estimates.
  • Drawbacks and advantages of linear model.
  • What can be gained from the application of non-linear connectivity methods (Mutual information, Synchronization and etc.)?

Lecture 6: Estimating of propagation measures (~40 minutes).
Maciej Kaminski

  • Directed transfer function (DTF) and direct DTF.
  • Dynamic propagation as a function of time (short-time DTF, SDTF).

Lecture 7: Brain networks from MEG data (~40 minutes). 
Vahe Poghosyan / Andreas A. Ioannides

  • Pair-wise connectivity measures from tomographic solutions.
  • Graph theoretical approaches.
  • Studies of normal and pathological brain activity and connectivity with MEG.

Lecture 8: Networks at different timescales (~40 minutes).
Fabio Babiloni / De Vico Fallani

  • How to use graph theory to synthesize the connectivity patterns generated by the estimation of connectivity measurements.
  • Measure of small-world, measure of density of the communication between the nodes of the network.

Round table discussion 2 (~20 minutes).
Chair: Katarzyna J. Blinowska

  • Relevance of connectivity measures and the networks derived from them for the study of the normal human brain function.
  • Differences in pathology.
  • Scales of functional connectivity and their relationships with the anatomical connectivity.
  • General discussion on afternoon session.
  • General discussion on the day’s course.

Bus departure for Nicosia hotels: 17:15-

 

Part 2: From the basic mathematical concepts to the practical analysis of real EEG and MEG data and their interpretation in terms of events at the neuronal and network levels.

The original set of notes from the theoretical workshop (day 1) will be used in short lectures and during practical sessions. Short lectures will ellaborate on the EEG (day 2) and MEG (day 3) related material, describe the freely available tools that will be used during the practical sessions, and provide an overview of the hands-on examples. The discussions will deal with the way mathematical assumptions and a priori constraints influence the solutions. It will also explain how features in the signal and the source activity estimates relate to normal and pathological brain function, as this is studied in invasive electrical measurements in vivo and in vitro slices.

Day 2 - short lectures and practical sessions, EEG

In practical sessions participants will use freely available software packages to go through all the steps of EEG data analysis, including basic pre-processing (e.g. removal of artifacts) and estimation of possible source locations, with an emphasize on distributed source analysis. Estimates of connectivity derived from the electrode signals will also be covered using available free packages.

 

Day 3 - short lectures and practical sessions, MEG

Participants will use the provided tools to explore how the basic concepts (source models and their implicit and explicit assumptions, conductivity models, lead fields, measures of connectivity and etc.) translate in practice, and how different choices affect the results and their interpretation.

 

Instructors and lecturers for days 2 and 3 of the sattelite course:

Laura Astolfi (La Sapienza, Italy)

Fabrizio de Vico Fallani (La Sapienza, Italy)

Janne Hämäläinen (Elekta Oy, Finland)

Liisa Helle (Elekta Oy, Finland)

Andreas A. Ioannides (AAI Scientific Cultural Services Ltd., Cyprus)

George K. Kostopoulos (University of Patras, Greece)

Maciej Kaminski (University of Warsaw, Poland)

Konrad Kwaskiewicz (University of Warsaw, Poland)

Lichan Liu (AAI Scientific Cultural Services Ltd., Cyprus)

Gijs Plomp (EPFL, Switzerland)

Vahe Poghosyan (AAI Scientific Cultural Services Ltd., Cyprus)