All Data are extracted from OpenBCI Kit
- Visualization of signal in all channels
- Statistical reports (mean, std, variance, correlation..etc)
- Power spectrum and data filtering with high pass filter
- Detection of most notable frequencies in the signal
- Model in SVM and Random Forest to classify the signal
- Time Series Analysis (TODO)
- Topological map
- Correlation Between EEG and EMG
- Multiple Linear Regression EEG vs ECG and EMG
- Added Analysis for other samples
In this section, we just apply the the following algorithms on the data.
in order to classify the signal whether it belongs to rest or activity
- test_size = 0.5
- C=10E3
- kernel = linear
positive | negative | |
---|---|---|
positive | 4 | 0 |
negative | 0 | 2 |
precision | recall | f1-score | support | |
---|---|---|---|---|
0 | 1.00 | 1.00 | 1.00 | 4 |
1 | 1.00 | 1.00 | 1.00 | 2 |
accuracy | 1.00 | 6 | ||
macro avg | 1.00 | 1.00 | 1.00 | 6 |
weighted avg | 1.00 | 1.00 | 1.00 | 6 |
1.0
- random_state = 0
- iterations = 100
positive | negative | |
---|---|---|
positive | 3 | 1 |
negative | 0 | 2 |
precision | recall | f1-score | support | |
---|---|---|---|---|
0 | 1.00 | 0.75 | 0.86 | 4 |
1 | 0.67 | 1.00 | 0.80 | 2 |
accuracy | 0.83 | 6 | ||
macro avg | 0.83 | 0.88 | 0.83 | 6 |
weighted avg | 0.89 | 0.83 | 0.84 | 6 |
0.8333333333333334
The data is first transformed into suitable format using a converter code,
to be in csv format and remove redunant columns
We Visualize the time series in time domain. to make a visual check that we are loading the correct data,
and we can use it to the next steps
Here we generate a full report for each sample,
to see the correlation between signals and also the see the Statistical features of the data itself
We do this, because EEG is only useful in frequency, and it's not having useful features in time domain alone, also to apply a HPF to choose the relevant features
We combine the data from all the samples, and add relevant features, to build a small dataset for the model itself
The data is first transformed into suitable format using a converter code,
to be in csv format and remove redunant columns
We Visualize the time series in time domain. to make a visual check that we are loading the correct data,
and we can use it to the next steps
Here we generate a full report for each sample,
to see the correlation between signals and also the see the Statistical features of the data itself
We do this, because EEG is only useful in frequency, and it's not having useful features in time domain alone, also to apply a HPF to choose the relevant features
We combine the data from all the samples, and add relevant features, to build a small dataset for the model itself