When
Tuesday Demo session
Abstract
The BioPoint is a new wireless and wearable device, targeting both the ambulatory and on-site monitoring of biosignals for research purposes. It incorporates the following six sensors i) A Photoplethysmograph (Blue, Green, Red, Infrared) at 50 Hz, ii) a 6-axis Inertial Measurement Unit (IMU) at 200 Hz, iii) a skin temperature sensor at 1 Hz, iv) an BioImpedance (EDA/GSR) at 30 Hz, v) an Electrocardiograph at 500 Hz and vi) an Electromyograph at 2000 Hz. Importantly, all of these sensors can work simultaneously without adverse effects on their signal quality and the raw data can easily be streamed/recorded. The BioPoint also features a 2GB onboard memory while also being capable of providing visual and vibrotactile feedback. All of that, with a full-day battery autonomy and in the size of a smartwatch. Thus, the goal of this project is to facilitate the acquisition of as most of a holistic view as possible from a single point on the body to inform about the behaviour and physiological response of a person to various stimuli (both internal and external). The BioPoint also comes with a 2-way python API to easily integrate it in a custom project which requires both recording and responsivity from the sensors. We are also currently integrating it within Lab Streaming Layer (LSL) so that it can be used and synchronized with other devices such as speakers, other sensors or musical instruments.
Bio
Ulysse received his PhD at Laval University (Canada) and did his Postdoc at the University of Oslo (Norway). His PhD research focused on the development of a human-computer interface to control a robotic arm, based on the user’s muscle activity (EMG). During his postdoctoral studies, he worked within the mental health sphere to develop biosensor-based mood-state detection for Bipolar Disorder Type I. Throughout his research, he has been especially interested in how to apply transfer learning, unsupervised domain adaptation and domain generalization to address the inherent challenges of a biosignal-based human-computer interface.