Related Experiment Video
Updated: Aug 13, 2025
08:51
The Use of Thermal Infra-Red Imaging to Detect Delayed Onset Muscle Soreness
Published on: January 22, 2012
19.4K
Can Data-Driven Supervised Machine Learning Approaches Applied to Infrared Thermal Imaging Data Estimate Muscular
David Perpetuini1, Damiano Formenti2, Daniela Cardone3
1Department of Neurosciences, Imaging and Clinical Sciences, University "G. d'Annunzio" of Chieti-Pescara, 66100 Chieti, Italy.
Sensors (Basel, Switzerland)
|January 21, 2023
Summary
This study introduces a non-invasive method using infrared thermal imaging and machine learning to estimate muscle activity and fatigue, offering a comfortable alternative to traditional surface electromyography (sEMG). The findings suggest skin temperature changes correlate with muscle exertion, paving the way for new wearable technology.
Area of Science:
- Biomedical Engineering
- Sports Science
- Wearable Technology
Background:
- Traditional surface electromyography (sEMG) uses gel electrodes, causing skin irritation and discomfort.
- Contactless EMG devices face challenges with motion artifacts and long-term monitoring comfort.
- There is a need for non-invasive, comfortable methods to assess muscle activity and fatigue.
Purpose of the Study:
- To develop and validate a non-invasive, contactless method for estimating muscle activity and fatigue parameters.
- To utilize infrared thermal imaging (IRI) and machine learning (ML) to predict sEMG-derived metrics.
- To explore the relationship between skin temperature and muscle exertion during fatiguing exercise.
Main Methods:
- Ten healthy participants performed bodyweight squats to exhaustion.
- Muscle activity was measured using sEMG, and temperature was recorded via IRI on the vastus medialis.
- Machine learning models were applied to IRI features to estimate sEMG's Average Rectified Value (ARV) and Median Frequency (MDF).
Main Results:
- Machine learning models achieved good performance in estimating ARV (r = 0.886) and MDF (r = 0.661) from thermal imaging data.
- A significant correlation was found between skin temperature changes and muscle activity/fatigue.
- The developed method shows potential for estimating muscle fatigue indicators non-invasively.
Conclusions:
- Infrared thermal imaging combined with machine learning can estimate muscle activity and fatigue parameters.
- While not replacing sEMG, this contactless approach offers a comfortable and non-invasive alternative for monitoring muscle exertion.
- The findings support the use of IRI for developing comfortable wearable sensors in sports and clinical settings.
