Detecting Blinks in Healthy and Parkinson's EEG: A Deep Learning Perspective

2025-09-05

AI Summary

Accurate detection of involuntary blinks in electroencephalography (EEG) signals from healthy individuals and Parkinson’s disease (PD) patients remains challenging due to motion artifacts—especially tremor-induced noise in PD. Method: We propose an end-to-end, multi-channel EEG blink segmentation framework operating directly on raw frontal electrode signals. Systematic evaluation of CNN, RNN, TCN, Transformer, and hybrid architectures reveals that a CNN-RNN model achieves optimal performance with minimal preprocessing. Its core innovation lies in a synergistic mechanism integrating temporal modeling (via RNN) and local feature extraction (via CNN), enhancing robustness against PD-related motor interference. Results: The model attains 95.8% accuracy on healthy controls and 75.8% on PD patients—outperforming all competing architectures. It enables reliable cross-population quantification of blink rate and cognitive load, establishing a generalizable deep learning paradigm for non-invasive neural state assessment.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Blinks in electroencephalography (EEG) are often treated as unwanted artifacts. However, recent studies have demonstrated that blink rate and its variability are important physiological markers to monitor cognitive load, attention, and potential neurological disorders. This paper addresses the critical task of accurate blink detection by evaluating various deep learning models for segmenting EEG signals into involuntary blinks and non-blinks. We present a pipeline for blink detection using 1, 3, or 5 frontal EEG electrodes. The problem is formulated as a sequence-to-sequence task and tested on various deep learning architectures including standard recurrent neural networks, convolutional neural networks (both standard and depth-wise), temporal convolutional networks (TCN), transformer-based models, and hybrid architectures. The models were trained on raw EEG signals with minimal pre-processing. Training and testing was carried out on a public dataset of 31 subjects collected at UCSD. This dataset consisted of 15 healthy participants and 16 patients with Parkinson's disease allowing us to verify the model's robustness to tremor. Out of all models, CNN-RNN hybrid model consistently outperformed other models and achieved the best blink detection accuracy of 93.8%, 95.4% and 95.8% with 1, 3, and 5 channels in the healthy cohort and correspondingly 73.8%, 75.4% and 75.8% in patients with PD. The paper compares neural networks for the task of segmenting EEG recordings to involuntary blinks and no blinks allowing for computing blink rate and other statistics.

Problem

• Accurate blink detection in EEG signals using deep learning
• Evaluating various models for segmenting blinks and non-blinks
• Testing model robustness on both healthy and Parkinson's patients

Innovation

• CNN-RNN hybrid model for EEG blink detection
• Sequence-to-sequence task with minimal pre-processing
• Multiple frontal EEG electrodes for robust accuracy