Towards NIRS-based hand movement recognition

Area of Science:

• Biomedical Engineering
• Neuroscience
• Human-Computer Interaction

Background:

• Hand movement recognition is crucial for gaming, social, and medical applications.
• Existing methods like physical contact and vision techniques have limitations.
• Surface ElectroMyoGraphy (sEMG) is increasingly used but struggles with nuanced hand movements.

Purpose of the Study:

• To investigate Near InfraRed Spectroscopy (NIRS) as a complementary technique for hand movement recognition.
• To explore the potential of NIRS in capturing muscle activity at different depths.
• To evaluate multimodal approaches combining NIRS and sEMG for improved recognition performance.

Main Methods:

• Utilized Near InfraRed Spectroscopy (NIRS) to capture muscle optical properties during hand movements.
• Employed surface ElectroMyoGraphy (sEMG) to record electrical muscle activity.
• Developed and compared three distinct multimodal recognition strategies integrating NIRS and sEMG data.

Main Results:

• NIRS demonstrated capability in recognizing a selection of hand movements.
• NIRS provided insights into muscle activity at varying depths.
• Preliminary results indicate the effectiveness of multimodal approaches in enhancing hand movement recognition.

Conclusions:

• NIRS offers a valuable supplementary perspective for hand movement analysis.
• Multimodal systems integrating NIRS and sEMG show potential for robust hand movement recognition.
• Further research is warranted to fully explore the capabilities of NIRS in this domain.