Annas lab notebook¶

week 1 (03.05.-09.05.)¶

literature search:

openknowledgemaps.org
terms: EEG AND schizophrenia; EEG AND schizophrenia AND language; EEG AND schizophrenia AND N400
naming the project

created github repository “LASERPIN”
- updated README.md file
- updated project.md
- updated labnotebook
- updated data*.md
- created a timeline on GitKraken
search for raw EEG data
writing mails to different people to ask for their EEG data
- Melissa Larsen and Ilvana Dzafic for the raw data of the Monash Study
- Shupin Tan (https://www.frontiersin.org/articles/10.3389/fninf.2019.00004/full)

writing again an email to Ilvana -> Ilvana told me, that she is in contact with the IT to solve the downloading problem
writing an e-mail to Daniel Hermens -> „BMRI Electroencephalography (EEG) clinical research data“ from the University of Sydney -> still waiting for an answer
trying out commands on docker and conda

talk with Alex about the answer from Elzbieta Olejarczyk
have a look on the paper and the eeg data from the study: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0188629
now my jupyter notebook works with my PC and my command system
programmed a calculator with Python code in Alex’ jupyter notebook in cooperation with Alex
after that I programmed my own calculator

downloaded the EEG data files from https://repod.icm.edu.pl/dataset.xhtml?persistentId=doi:10.18150/repod.0107441
updated data*.md
read the related publication to the EEG data files: Olejarczyk E, Jernajczyk W (2017) Graph-based analysis of brain connectivity in schizophrenia. PLoS ONE 12(11): e0188629. https://doi.org/10.1371/journal.pone.0188629
meeting with José to talk about the first steps of preprocessing
creating a jupyter notebook for preprocessing -> first steps on preprocessing to get a first look on the data (28 .edf files)
get to know MNE as a tool for EEG data

updating github repository
creating eog and ecg epochs to evaluate interfering components in the eeg signal
first steps for the ICA (independent component analysis)

meeting with José to solve problems with the preprocessing
completion of the preprocessing for one subject
creation of a loop on the jupyter notebook to preprocess all subjects
- take Fp1 and Fp2 as reference for eye artifacts
- ICA did not work for every subject: s02, s06, s07, s11, s12, h06, h07, h10, h11, h14

update “_toc.yml” on github
meeting with José
completion of preprocessing by creating an own template for filtering eye artifacts in the eeg signal
control the ICA for all subjects
- it worked better than our first ICA
- some artifacts could not be filtered: s06, s09, h04, h11
- in one subject it could not be detected a component: s12

we decide to make a frequency band analysis for our eeg data because of the missing of stimulus events
doing the epochs analysis for all subject and plotting the results
- adaptation of the scales for the different frequencies
averaging the data separately for the frequencies (delta, theta, alpha, beta and gamma)
- averaging the activation for each person
- averaging the activation for the two groups

plotting the averages for all frequencies and the two groups and making a short presentation about the results
discuss the results with Alex
looking for prospective statistic analysis
literature research because of the new aim of our project:
- frequency-band analysis
  - detecting and defining critical electrodes to compare schizophrenic patients with healthy persons
  - observing greater differences in delta and gamma oscillations
  - independent t-test for critical electrodes
creation of “code”-folder on github