Harmonizing Wearable Biosensor Data Streams to Test Polysubstance Detection
Fang, Hua (Julia)
Smelson, David A.
UMass Chan AffiliationsDepartment of Psychiatry
Department of Emergency Medicine
Department of Population and Quantitative Health Sciences
Document TypeConference Paper
Biomedical Devices and Instrumentation
Health Services Administration
Substance Abuse and Addiction
MetadataShow full item record
AbstractWearable biosensors, as a key component of wireless body area network (WBAN) systems, have extended the ability of health care providers to achieve continuous health monitoring. Prior research has shown the ability of externally placed, non-invasive sensors combined with machine learning algorithms to detect intoxication from a variety of substances. Such approaches have also shown limitations. The difficulties in developing a model capable of detecting intoxication generally include differences among human beings, sensors, drugs, and environments. This paper examines how approaching wireless communication advances and new paradigms in constructing distributed systems may facilitate polysubstance use detection. We perform supervised learning after harmonizing two types of offline data streams containing wearable biosensor readings from users who had taken different substances, accurately classifying 90% of samples. We examine time domain and frequency domain features and find that skin temperature and mean acceleration are the most important predictors.
Rumbut J, Fang H, Wang H, Carreiro S, Smelson D, Chapman B, Boyer E. Harmonizing Wearable Biosensor Data Streams to Test Polysubstance Detection. Int Conf Comput Netw Commun. 2020 Feb;2020:445-449. doi: 10.1109/icnc47757.2020.9049684. Epub 2020 Mar 30. PMID: 33732746; PMCID: PMC7962664. Link to article on publisher's site