An EMG-based Eating Behaviour Monitoring system with haptic feedback to promote mindful eating

Area of Science:

• Biomedical Engineering
• Behavioral Science
• Human-Computer Interaction

Background:

• Mindless eating is linked to obesity and unhealthy eating behaviors.
• Current methods for monitoring eating behavior are often inaccurate or cumbersome.
• There is a need for intelligent, lightweight systems for automatic eating behavior monitoring and feedback.

Purpose of the Study:

• To develop an automated system for detecting eating behavior using wearable Electromyography (EMG) sensors.
• To apply this system with real-time wristband haptic feedback to promote mindful eating.
• To evaluate the effectiveness of the system in moderating eating behavior.

Main Methods:

• Collected EMG data from 16 participants to develop an algorithm for detecting chewing and swallowing.
• Extracted 18 features from EMG data and used Support Vector Machine (SVM) classifier.
• Conducted an experimental study with 20 participants to assess the impact of haptic feedback on eating behavior.

Main Results:

• The algorithm achieved high accuracy in detecting chewing (F1-Score = 0.95) and swallowing (F1-Score = 0.87).
• Participants showed a significantly lower chewing rate when receiving haptic feedback via wristband vibration compared to baseline conditions.
• The system demonstrated the potential for self-moderation of eating behavior through automated detection and feedback.

Conclusions:

• Wearable EMG sensors can accurately detect eating behaviors.
• Haptic feedback integrated with automated eating detection can effectively promote mindful eating.
• This technology offers a novel approach to managing eating behavior for improved health outcomes.