A Multi-channel Eye-Mask-based Wearable System for Accurate Eye Movement Recognition and Saccadic Angle Estimation

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Wearable Technology

Background:

• Advanced wearable systems are crucial for developing accurate human-computer interaction (HCI) solutions.
• Existing systems often face limitations in precision, comfort, or reliability for capturing eye movement patterns.

Purpose of the Study:

• To propose and validate a Multichannel Eye-mask-based Wearable (MEW) system for high-sensitivity eye movement recognition (EMR) and saccadic angle estimation (SAE).
• To demonstrate the system's effectiveness in portable, real-time eye tracking applications.

Main Methods:

• Development of a MEW system integrating multi-channel eye-mask electrodes for high-spatial-resolution electrooculography (EOG) sensing.
• Utilization of a high-precision EOG acquisition system with an STM32 microcontroller and BLE transceiver for wireless data transmission.
• Implementation of automatic algorithms for EMR and SAE, validated with 20 subjects performing 8-directional tasks.

Main Results:

• The EMR task achieved an average accuracy of 87.2%.
• The SAE task demonstrated a mean absolute error (MAE) of 4.19° and a root mean square error (RMSE) of 5.27°.
• The system effectively captured and decoded multiple eye movement directions and angles.

Conclusions:

• The proposed MEW system provides a reliable, portable, and high-sensitive solution for eye movement tracking.
• The system shows significant potential for practical applications in EOG-based HCI.