A robust deep learning-driven framework for detecting Parkinson's disease using EEG

Parkinson's disease (PD) is a neurodegenerative condition that impairs motor functions. Accurate and early diagnosis is essential for enhancing well-being and ensuring effective treatment. This study proposes a deep learning-based approach for PD detection using EEG signals. First, a channel attention module refines the EEG data. Then, wavelet scattering transform generates time-frequency maps from EEG. Subsequently, a GAN (Generative Adversarial Network) model is designed to generate more similar time-frequency maps of PD-affected patients as well as healthy control subjects. An efficient CNN-Transformer-based model is designed and trained using the augmented time-frequency map images, achieving 99.52% accuracy.