Brainstorm-savvy researcher required urgently for challenging EEG pre-processing task

Our EEG project investigates the effects of different frequencies of transcutaneous electroacupuncture stimulation (TEAS) applied to the hands.

In this short-term, temporary post, the researcher will be required to use an automated Brainstorm pipeline with the usual pre-processing steps (based on ICA, not SSP), in order to obtain clean data.

We have 2004 Matlab data files derived from 5-minute 19-channel EEG recorded from 66 individuals. These files were recorded using a linked ears reference and a 500 Hz sampling rate (250 Hz in a few of them), filtered between 0.5 and 150 Hz, with a notch filter at 45-55 Hz.

The files are of two types, all recorded with eyes open – half were with pulsed biphasic square wave electrical stimulation to the hands (at 2.5, 10, 80 or 160 pulses per second), and half without the stimulation.

HERE'S THE CHALLENGE: Some of the recordings during stimulation are likely to be noisy, and as we are looking for – among other things – a local physiological response in the brain related to the stimulation frequency (rather than just a volume conduction effect), the researcher will need to take into account that using ICA without careful consideration may result in discarding precisely what we are seeking.

The following outputs will be required, for both [A] the original linked ears reference and [B] the data re-montaged using an average local reference:

(0) Cleaned-up data with artefacts and bad segments marked and/or excluded;
(1.1) Spectral power between 0.5 and 45 Hz; (1.2) Spectral power between 0.5 and 150 Hz;
(2.1) Absolute spectral power density (Welch) in standard bands; (2.2) Relative spectral power density (Welch) in standard bands;
(3.1) Absolute spectral power density (Welch) in further bands up to 150 Hz (where sampling rate permits); (3.2) Relative spectral power density (Welch) in further bands up to 150 Hz (where sampling rate permits);
(4) For each recording, based on ICA and as far as possible, counts of (4.1) vertical eye blinks, (4.2) saccades (i.e. horizontal eye movements), and (4.3) body/head movements

The researcher’s initial task would be to provide these outputs for an initial sample of 20 files.

If these prove satisfactory, the remaining files could be processed in the same way, and further connectivity analysis using Brainstorm may then also be an option.

Payment and co-authorship to be negotiated.

If you think you would be interested in this temporary position, please contact:

David Mayor
Hon member AACP

Visiting Fellow (Physiotherapy),
Department of Allied Health Professions and Midwifery,
School of Health and Social Work,
University of Hertfordshire, UK

Acupuncture practitioner,
Welwyn Garden City, UK
+44 [0]1707 320782

PLEASE NOTE: Expressions of interest to be received before 27 July 2019