Dual-Enhanced Temporal Aggregation with Aware-Scale Fuzzy Entropy for Cross-Subject EEG-Based Parkinsons Disease Detection
Dual-Enhanced Temporal Aggregation with Aware-Scale Fuzzy Entropy for Cross-Subject EEG-Based Parkinsons Disease Detection
Where did the research take place?
The study site has not been established. Author addresses may differ from where the research occurred.
A plain-language reading has not been prepared for this paper yet.
Original abstract
Parkinson's disease (PD) classification based on electroencephalography (EEG) signals remains challenging due to its limitations in capturing long-term dependency features and complex time series. To address this, the dual-enhanced temporal aggregation network (DETANet) and aware-scale fuzzy entropy (ASFE) are proposed in this paper. With its adaptive smoothing strategy and attention regulation method, it effectively enhances temporal consistency and cross-period correlation modeling capabilities. Experiments were validated using leave-one-out cross-validation (LOOCV) on public datasets, obtaining accuracy of 95.87% on the UNM dataset and 88.79% on the Oddball dataset.