An online swallow detection algorithm based on the quadratic variation of dual-axis accelerometry
IEEE Transactions on Signal Processing
Food Intake Activity Detection Using a Wearable Microphone System
IE '11 Proceedings of the 2011 Seventh International Conference on Intelligent Environments
Hi-index | 0.00 |
This paper presents a mobile food intake monitoring system. The system works based on a key observation that a person's otherwise continuous breathing signal is interrupted by a short apnea during swallowing. The breathing signal is collected using two Respiratory Inductance Plethysmography (RIP) belts on the chest and abdomen, and wirelessly transmitted to an Android smart phone via Bluetooth. Support Vector Machine estimating the posterior probability is applied on the extracted features. Through experiments on six healthy subjects, the system is demonstrated to be an effective way of monitoring food and drink intake. The key novelty of this paper includes 1) a non-invasive RIP belt system for swallow detection without any cosmetic and wearing discomfort issues, and 2) a smart phone to collect data wirelessly and execute swallow detection algorithms before uploading the results to the cloud for enabling remote health monitoring.