An automated EEG repair tool Kristjan-Julius Laak mission Automate - - PowerPoint PPT Presentation

▶
an automated eeg repair tool
SMART_READER_LITE
LIVE PREVIEW

An automated EEG repair tool Kristjan-Julius Laak mission Automate - - PowerPoint PPT Presentation

Dec 31, 2022 262 likes •490 views

An automated EEG repair tool Kristjan-Julius Laak mission Automate the phase of cleaning EEG data from non-brain- related activity. impossible? Step by step one goes very far (proverb) Mission possible AUTOMATE ONLY SOME ASPECTS OF THIS

slide-1
SLIDE 1

An automated EEG repair tool

Kristjan-Julius Laak

slide-2
SLIDE 2

mission

Automate the phase

  • f cleaning EEG data

from non-brain- related activity.

impossible?

slide-3
SLIDE 3

Step by step one goes very far

(proverb)

slide-4
SLIDE 4

Mission possible

AUTOMATE ONLY SOME ASPECTS OF THIS PHASE

slide-5
SLIDE 5

Who cares at all?

200 recordings .. 10 subjects ... 60 electrodes ....

slide-6
SLIDE 6

!!!!!!!!!!!!!?!!!!!!!!!!!!!!

256

slide-7
SLIDE 7

Goal

Build an automated algorithm that detects and repairs channels containing noise or artifacts.

slide-8
SLIDE 8

Prerequisites

  • Data is already preprocessed (Fieldtrip)
  • There are trials not a single continuous

recording

  • Some aspects of the data are known (e.g.

sample rate)

slide-9
SLIDE 9

General pipeline

±250µV Euclidean distance from median Probably density estimate Interpolation Visual rejection ICA

slide-10
SLIDE 10

General pipeline

±250µV Euclidean distance from median Probably density estimate Interpolation Visual rejection ICA

slide-11
SLIDE 11

Euclidean distance from median

slide-12
SLIDE 12

General pipeline

±250µV Euclidean distance from median Probably density estimate Interpolation Visual rejection ICA

slide-13
SLIDE 13

Density estimate

  • A. Probability density estimate plots for each channel per trial. B. The red line on is a reflection
  • f the left side curve from the maximum, illustrating a Gauss curve.
slide-14
SLIDE 14

General pipeline

±250µV Euclidean distance from median Probably density estimate Interpolation Visual rejection ICA

slide-15
SLIDE 15

Bad channels (automated)

slide-16
SLIDE 16

General pipeline

±250µV Euclidean distance from median Probably density estimate Interpolation Visual rejection ICA

slide-17
SLIDE 17
slide-18
SLIDE 18

General pipeline

±250µV Euclidean distance from median Probably density estimate Interpolation Visual rejection ICA

slide-19
SLIDE 19

Independent component analysis

slide-20
SLIDE 20

Results

The difference between raw data (A) and pre-processed data (B) for the same trial. Different colours mark different channels.
slide-21
SLIDE 21

Last slide.

slide-22
SLIDE 22

Intentionally black slide... Thank you!