Automatic monitoring of activities of daily living based on real-life acoustic sensor data: A preliminary study

This work examines the use of a low-power Wireless AcousticSensor Network (WASN) for the observation of clinicallyrelevant activities of daily living (ADL) (e.g. eating, personalhygiene, toilet usage, etc.) from elderly. The sensors used inthe WASN are both audio and ultrasound receivers. To thebest of our knowledge, the combination of audio and ultrasoundas a basis for ADL monitoring has not been investigated yet.This paper describes a baseline approach for ADL classificationbased on Gaussian mixture models. Preliminary results inthis work indicate that classification accuracies up to 85.0 %± 14.6 for audio and 61.7% ± 11.3 for ultrasound are alreadyachievable on realistic real-life recorded data.Index Terms: acoustic scene analysis, audio, ultrasound,acoustic scene classification, activities of daily living, automaticmonitoring Vuegen L., Van Den Broeck B., Karsmakers P., Van hamme H., Vanrumste B., ''Automatic monitoring of activities of daily living based on real-life acoustic sensor data: a preliminary study'', 4th workshop on speech and language processing for assistive technologies - SLPAT-2013, 6 pp., August 21-22, 2013, Grenoble, France. ispartof: pages:113-118 ispartof: Fourth workshop on speech and language processing for assistive technologies (SLPAT): Proceedings pages:113-118 ispartof: Workshop on speech and language processing for assistive technologies - SLPAT-2013 location:Grenoble, France date:21 Aug - 22 Aug 2013 status: published