Özet:

Abstract It can be difficult for healthcare professionals to recognise and diagnose sliding disorder, a neurological ailment marked by a loss of coordination and control over movement. Signals from electroencephalography (EEG) have shown to be a useful method for examining brain activity and can shed light on neurological conditions. Using a hybrid deep learning framework and an ensemble machine learning classifier, we suggest a unique method in this study for the detection of sliding disorder. In the first step of our procedure, EEG signals from healthy controls and people with Sleeping disease are collected. To collect the necessary information, these signals are divided into shorter time intervals after being preprocessed to remove noise and artefacts. In order to obtain a concise representation of the EEG data, feature extraction techniques are used. This aids in highlighting significant patterns and traits connected to Sleeping disease. The proposed methodology is intended for integration into embedded devices to provide a novel and effective method for classifying sleep stages. For evaluation, the study makes use of Power Spectrum Density (PSD) Dataset. We experimented with a publicly accessible Power Spectrum Density (PSD) dataset of patients with sliding disease in order to assess the effectiveness of our suggested strategy. The outcomes show that our method outperforms both conventional machine learning algorithms and stand-alone deep learning models in terms of sliding disorder identification. The use of Hybrid Deep Learning with Ensemble Machine Learning Classifier together effectively enhance classification sensitivity of 89.06% and accuracy to 96.78%.

Anahtar Kelimeler: