IEEE BioCAS 2023

Slides EpiDeNet oral slides SkiLog poster handout EpiDeNet publication SkiLog publication
Delivering the EpiDeNet oral presentation at IEEE BioCAS 2023

Abstract

In Toronto I presented two pieces of our wearable EEG research: an oral talk on EpiDeNet, our gradient-boosted seizure detector with Sensitivity-Specificity Weighted Cross-Entropy, and a poster on SkiLog, a smart insole system that delivers real-time biofeedback to ski jumpers. Both projects push the boundary of what is possible on ultra-low-power hardware, highlighting how adaptive ML pipelines can move entirely on-device.

Date
19, Oct, 2023
Event
IEEE Biomedical Circuits and Systems Conference (BioCAS)
Location
Beanfield Centre, Toronto, Canada
105 Princes' Blvd, Toronto, Ontario M6K 3C3

Highlights

  • Oral spotlight: Presented EpiDeNet: An Energy-Efficient Approach to Seizure Detection for Embedded Systems (MSc thesis collaboration with Upasana Chakraborty and Xiaying Wang).
  • Poster session: Showcased SkiLog: A Smart Sensor System for Performance Analysis and Biofeedback in Ski Jumping together with Christoph Leitner.
  • Demo focus: Ultra-low-power EEG processing on GAP9-class hardware and adaptive boosting pipelines.

Photo reel

EpiDeNet oral presentation in Toronto.
EpiDeNet oral presentation in Toronto.
With Christoph Leitner at the SkiLog poster session.
With Christoph Leitner at the SkiLog poster session.

Slides & poster

BioCAS Wearable EEG Seizure Detection TinyML
Thorir Mar Ingolfsson
Thorir Mar Ingolfsson
Postdoctoral Researcher

I develop efficient machine learning systems for biomedical wearables that operate under extreme resource constraints. My work bridges foundation models, neural architecture design, and edge deployment to enable real-time biosignal analysis on microwatt-scale devices.

Related