A database developed by transforming EEG signals into visual representations using Gramian Angular Fields.
In the realm of EEG analysis, the complex nature of EEG signals often necessitates their transformation into visual formats, thereby facilitating more accessible data analysis. Despite this practice, there is a notable scarcity of datasets available for such transformed EEG signals. Addressing this gap, we introduce ’MindPixels,’ a database developed by transforming EEG signals into visual representations using Gramian Angular Fields. ”MindPixels” stands out as a preprocessed, easily accessible repository of EEG-derived image data, effectively streamlining the traditionally time-intensive preprocessing steps. We showcase the utility of ”MindPixels” by deploying deep learning techniques for the classification of focal and nonfocal EEG GASF images. Our comprehensive evaluation of the ”MindPixels” database yields promising outcomes, reinforcing its potential as an invaluable asset in augmenting the interpretation of EEG signals.
