Detecting the Intention to Move Upper Limbs from Electroencephalographic Brain Signals

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Brain-machine interfaces (BMI) require early decoding of motor states for natural control of neuroprosthetic devices.
• Detecting movement intention before physical execution is crucial for advanced BMI applications.
• Current BMI systems face challenges in bridging the gap between neural signals and device activation.

Purpose of the Study:

• To investigate the detection of movement intention from electroencephalographic (EEG) signals prior to actual limb movement.
• To explore the potential of early motor command detection for enhancing BMI control and rehabilitation.
• To analyze brain activity patterns associated with the intention to move.

Main Methods:

• Recorded electroencephalographic (EEG) signals from six healthy participants during self-initiated upper limb reaching movements.
• Analyzed EEG data for event-related desynchronization in motor-related α and β frequency bands over the motor cortex.
• Developed a detection method to identify movement intention from pre-movement EEG activity.

Main Results:

• Significant event-related desynchronization was observed in motor-related frequency bands before and during movement execution.
• Successfully classified between states of relaxation and movement intention using EEG signals.
• Demonstrated significant detection of movement intention preceding the onset of the physical movement.

Conclusions:

• Early detection of movement intention from EEG is feasible and can be achieved by analyzing oscillatory brain activity.
• This capability can significantly improve BMI performance by reducing the delay between thought and action.
• The findings support the use of pre-movement intention detection for natural control in neuroprosthetic and rehabilitation devices.