Neural signals-based respiratory motion tracking: a proof-of-concept study

Area of Science:

• Biomedical Engineering
• Neuroscience
• Medical Physics

Background:

• Accurate respiratory motion tracking is crucial for thoracoabdominal radiotherapy and robotic surgery.
• Conventional imaging-based techniques suffer from system latency, impacting treatment precision.
• There is a need for advanced methods to overcome these time lags.

Purpose of the Study:

• To investigate the precursor time of respiratory-related neural signals.
• To analyze the potential of using neural signals for respiratory motion tracking.
• To address the system latency in current medical technologies.

Main Methods:

• Simultaneous acquisition of neural signals (256-channel EEG) and respiratory motion from 18 healthy volunteers.
• Preprocessing of EEG data using MNE Python to extract respiratory-related neural signals.
• Cross-correlation analysis to determine precursor time and correlation coefficients.

Main Results:

• Respiratory-related neural signals were detectable via non-invasive EEG, preceding actual respiratory motion.
• The average precursor time of these neural signals was 0.68 seconds.
• Cross-correlation coefficients ranged from 0.22 to 0.87, indicating a significant relationship.

Conclusions:

• Neural signals show potential to compensate for system latency in medical linear accelerators and surgical robots.
• Neural signal-based respiratory motion tracking is a promising solution for improving accuracy.
• This approach could significantly benefit thoracoabdominal radiotherapy and robotic surgery.