This event is free and open to the general public, no registration required.
We welcome you to share this event via Facebook.
Abstract:
In-ear wearables, or hearables, have become increasingly popular over recent years. This is mainly due to two reasons: people have become accustomed to continuously wearing in-ear devices and, more importantly, when occluded, the ear becomes a portal of access to a plethora of human-produced events ranging from speech to a blink of an eye. Access to such a variety of signals coupled with advancements in artificial intelligence pave the way for automatic disease detection with hearables. This is particularly interesting for degenerative diseases such as Parkinson’s and Alzheimer’s disease because intervention at the early stages could slow down the decline of motor and cognitive abilities. The continuous individual use of hearables as well as simultaneous tracking of a diverse set of signals provides an opportunity for multimodal prediction models for early disease detection. This talk focuses on current advancements on classification models for in-ear signals and future applications of such algorithms.