Fiber optic sensor embedded smart helmet for real-time impact sensing and analysis through machine learning.

Area of Science:

• Biomedical Engineering
• Neuroscience
• Materials Science

Background:

• Mild traumatic brain injury (mTBI) is linked to chronic neurodegenerative conditions like PTSD and cognitive impairment.
• Early detection of concussive events is crucial for timely diagnosis and effective recovery strategies.
• Current methods lack real-time, detailed impact data, hindering understanding and management of head injuries.

Purpose of the Study:

• To develop a smart helmet capable of real-time sensing and analysis of blunt-force impact events.
• To utilize fiber Bragg grating (FBG) sensor technology for capturing impact magnitude and direction.
• To employ machine learning (ML) models for accurate prediction of impact characteristics.

Main Methods:

• A smart helmet prototype was engineered with an embedded fiber Bragg grating (FBG) sensor.
• Transient impact signals were captured, providing both magnitude and directional data.
• Machine learning (ML) models were trained using the FBG sensor data to predict impact events.

Main Results:

• The FBG-embedded smart helmet successfully achieved real-time sensing of concussive events.
• Impact data "fingerprints" (magnitude and direction) correlated with different impactors.
• ML models accurately predicted impact magnitudes and directions (R² ≈ 0.90) using unseen data.

Conclusions:

• The ML-assisted smart helmet provides accurate, real-time identification of concussive events.
• This system offers multi-metric impact analysis, surpassing existing device capabilities.
• The ML-FBG smart helmet system represents a novel early intervention strategy for head injuries.