Related Experiment Video
Updated: Jun 12, 2025

Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification
Published on: February 14, 2018
An accelerometry and gyroscopy-based system for detecting swallowing and coughing events
Guylian Stevens1,2, Stijn Van De Velde3, Michiel Larmuseau4
1Departement of electronics and information systems-IBiTech, Ghent University, Korneel Heymanslaan, Gent, 9000, East-Flanders, Belgium. Guylian.Stevens@ugent.be.
A new wearable sensor system accurately detects swallowing and coughing, offering potential for early pneumonia prediction and improved patient monitoring in hospitals. This simple, robust technology aids in assessing swallowing function and respiratory health.
Area of Science:
- Biomedical Engineering
- Clinical Monitoring
- Respiratory Medicine
Background:
- Spontaneous swallowing frequencies (SSF) and coughing frequencies (CF) are crucial indicators of swallowing function, dysphagia, and pneumonia risk.
- Current methods for measuring SSF and CF, like electromyography and acoustic sensors, are often complex and difficult to implement clinically.
- There is a significant need for a simple, flexible, and robust system to measure these physiological parameters.
Purpose of the Study:
- To develop a low-complexity, flexible system for measuring spontaneous swallowing and coughing frequencies.
- To create a model capable of accurately identifying swallowing and coughing actions while distinguishing them from other movements.
- To assess the potential clinical applicability of this system in hospital settings and home care.
Main Methods:
- Recruited forty healthy volunteers for the study.
- Equipped participants with two medical-grade Movesense MD sensors measuring tri-axial accelerometry and gyroscopic movements, placed on the cricoid cartilage and epigastric region.
- Instructed participants to perform various actions, including swallowing and coughing, and processed recorded signals to develop an identification algorithm.
Main Results:
- The developed model achieved 70% sensitivity and 66.7% precision for detecting swallowing.
- The model demonstrated 100% sensitivity and 80% precision for detecting coughing.
- The system showed acceptable sensitivity and precision for detecting swallowing and coughing movements using two sensors.
Conclusions:
- SSF, CF, and their temporal relationships can serve as valuable predictive tools for diagnosis and therapeutic guidance.
- The developed sensor-based model is simple, robust, and promising for widespread clinical research and application.
- This technology has potential applications in predictive models for patient management, such as weaning from ventilatory support and early pneumonia detection.
More Related Videos
07:22Minimally Invasive Murine Laryngoscopy for Close-Up Imaging of Laryngeal Motion During Breathing and Swallowing
Published on: December 1, 2023
08:32Adapting Human Videofluoroscopic Swallow Study Methods to Detect and Characterize Dysphagia in Murine Disease Models
Published on: March 1, 2015