We propose a wearable sensor system for detecting swallows.
We collect data and evaluate the system's ability to detect medication using machine learning algorithms.
We use a probabilistic model to improve classification accuracy based on timing information.