N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

44
N. Laskaris N. Laskaris

Transcript of N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Page 1: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

N. LaskarisN. Laskaris

Page 2: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

N. LaskarisN. Laskaris

Page 3: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

[ IEEE SP Magazine, May 2004 ]

N. Laskaris,

S. Fotopoulos, A. Ioannides

ENTER-2001

Page 4: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

new toolsnew tools

for Mining Information from for Mining Information from multichannel encephalographic recordingsmultichannel encephalographic recordings

& applications& applications

Page 5: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

What is Data Mining ?Data Mining ?

How is it applied ?

Why is it useful ?

What is the difficulty with single trials ?

How can Data MiningData Mining help ?

Which are the algorithmic steps ?

Is there a simple example ?

Is there a more elaborate example ?

What has been the gain ?

Where one can learn more ?

Page 6: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

What is Data MiningData Mining &

Knowledge DiscoveryKnowledge Discovery in databases ?

Page 7: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Data MiningData Mining is “the data-driven discovery and

modeling of hidden patterns in large volumes of data.”

It is a multidisciplinary field, borrowing and enhancing ideas from diverse areas such as statistics, image understanding, mathematical optimization, computer vision, and pattern recognition.

It is the process of nontrivialnontrivial extractionextraction of of implicitimplicit, previously unknown, and potentially useful information from voluminous datafrom voluminous data.

Page 8: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

How is it applied in the context of

multichannel multichannel encephalographic recordingsencephalographic recordings ?

Page 9: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Studying Brain’s self-organization by monitoring the dynamic pattern

formation reflecting neural activity

Page 10: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Why is it a

potentially valuable methodology for analyzing

Event-RelatedEvent-Related recordings ?

Page 11: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

The analysis of

Event-Related DynamicsEvent-Related Dynamics

aims at understanding the real-time processingreal-time processing of a stimulus

performed in the cortex

and demands tools able to deal with Multi-Trial dataMulti-Trial data

The traditional approach is based on identifying peaks in the averaged signal

-It blends everything

happened during the

recording

Page 12: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

What is the difficulty in analyzing Single-TrialSingle-Trial responses ?

Page 13: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

At the single-trial level, we are facing

ComplexComplex SpatiotemporalSpatiotemporal DynamicsDynamics

Page 14: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

How can Data MiningData Mining help to circumvent this complexity and reveal

the underlying brain mechanisms ?

Page 15: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

directed queries are formed in the Single-Trial data which are then summarized

using a very limited vocabulary of information granules

that are easily understood, accompanied by well-defined semantics and help express relationships existing in the data

Page 16: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

The information abstractioninformation abstraction is usually accomplished via clusteringclustering techniques and followed by a proper visualization schemevisualization scheme that can readily spot interesting events and trends in the experimental data.

Page 17: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

- Semantic Semantic MapsMaps

The CartographyCartography of neural function results in a topographical representation of response variation

and enables the virtual navigation in the encephalographic database

Page 18: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Which are the intermediate

algorithmic steps ?

Page 19: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

A Hybrid approach A Hybrid approach

Pattern AnalysisPattern Analysis & Graph TheoryGraph Theory

Page 20: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Step_Step_ the spatiotemporal dynamics are decomposed

Design of the spatial filterspatial filter used to extract

the temporal patternstemporal patterns conveying the

regional response dynamics

Page 21: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Step_Step_ Pattern AnalysisPattern Analysis of the extracted ST-patterns

Interactive Interactive Study Study of of

pattern variabilitypattern variability

Feature Feature extractioextractio

nn

Embedding Embedding in Feature in Feature

SpaceSpace

Clustering & Clustering & Vector Vector

QuantizationQuantization

Minimal Spanning Minimal Spanning TreeTree of the of the

codebookcodebook

MST-ordering MST-ordering of the code of the code

vectors vectors

Orderly Orderly presentation presentation

of response of response variabilityvariability

Page 22: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.
Page 23: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.
Page 24: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Step_Step_ Within-group Analysis of regional response dynamics

-

Page 25: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Step_Step_ Within-group Analysis of multichannel single-trial signals

Page 26: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Step_Step_ Within-group Analysis of single-trial MFT-solutionsMFT-solutions

Page 27: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Is there a simple example?

[ Laskaris & Ioannides, Clin. Neurophys., 2001 ]

Page 28: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Repeated stimulationRepeated stimulation

120 trials, binaural-stimulation [ 1kHz tones, 0.2s, 45 dB ], ISI: 3sec, passive listening

Task : to ‘‘explain’’ the averaged M100-response

Page 29: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

The M100-peak emerges from the stimulus-induced

phase-resetting

Page 30: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Phase reorganizationPhase reorganization of the ongoing brain waves

Page 31: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Is there a more elaborate example?

[ Laskaris et al., NeuroImage, 2003 ]

Page 32: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

A study of A study of global firing global firing patternspatterns

Page 33: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Their relation Their relation with localized with localized sources sources

Page 34: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

and ….and ….

initiating initiating eventsevents

Page 35: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

240 trials, pattern reversal, 4.5 deg , ISI: 0.7 sec, passive viewing

Single-Trial data in unorganized format

Page 36: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Single-Trial data summarized via ordered prototypes reflecting the variability of regional response dynamics

Page 37: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

‘‘‘‘The ongoing activity The ongoing activity before the stimulus-onsetbefore the stimulus-onset

is functionally coupled is functionally coupled with the with the

subsequent subsequent regional response’’regional response’’

Page 38: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Polymodal Parietal Areas BA5 & BA7 are the major sources of the observed variability

Page 39: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

There is relationship There is relationship between between N70m-response variability N70m-response variability

and activity and activity in early visual areas.in early visual areas.

Regional vs Local response dynamics :

Page 40: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

What has been the lesson, so far, from the analysis of Event-Related Dynamics ?

Page 41: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

The ‘‘dangerous’’ equationThe ‘‘dangerous’’ equation

Page 42: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.
Page 43: N. Laskaris. [ IEEE SP Magazine, May 2004 ] N. Laskaris, S. Fotopoulos, A. Ioannides ENTER-2001.

Where one can learn more

about Mining Mining InformationInformation from

encephalographic recordings ?

http://www.hbd.brain.riken.jp/

http://www.humanbraindynamics.com