EEG_preprocessing

KO | EN

gitlite — search

#javascript #react #llm #go #machine-learning #deep-learning #linux #docker #macos #nodejs #windows #awesome-list #security #ios

EEG_preprocessing

(★ 7)

Performing basic preprocessing operation like filtering, removing line noise, detrending and removing artifacts in EEG data.

EEG_preprocessing Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

"# EEG_preprocessing" 
This repository contains Maltab and Python file for preprocessing EEG data. 
The Matlab file is the preprocessing step for futher analysis. The preprocessing was performed as follows:
  1. Filtering (0.5 – 30 Hz)
	2. re-Reference (Common Average Reference)
	3. Calculating ICA weights using EEGLAB ICA plugin.
	4. Finding and Selecting only actual trials (against whole dataset) for clean data
	5. Epoching from -1 to 2 seconds  was used.
	6. To reject artifactual components, SASICA was used.
	7. Detrending the data and removing line noise.
	8. Removing abnormal epochs
The above steps are the result of researching existing papers on the topic and experimental approach.

The python file is an analysis code for classifying EEG data for grasp-lift data taken from Kaggle database. The description of the dataset is given here: https://www.kaggle.com/c/grasp-and-lift-eeg-detection.
In the analysis I performed the simple transformation of the data and applying LightGbM algorithm from Microsoft. The code can be used as a template for future use on other EEG datasets. The lightGBM parameters were selected using GrigSearch and Cross-Validation. Along with lightGBM other algorithms such as logistic regression, XGboost, SVM were used, however, lightGBM has the best balance between speed (Top1) and accuracy (Top 2).

EEG_preprocessing

EEG_preprocessing Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

EEG_preprocessing

EEG_preprocessing Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README