SP 1.1.4 - Ultradian Monitoring in Adrenal Insufficiency

Event

EUROSENSORS 2026
2026-09-06 - 2026-09-09
Zurich

Band

Lectures

Chapter

Continuous Health Monitoring in Medicine

Author(s)

I. Botusan, K. Berinder, O. Kämpe, S. Bensing - Karolinska Institutet,Stockholm (Sweden) and Karolinska University Hospital,Stockholm (Sweden), E. Zavala - University of Manchester,Manchester (United Kingdom), T. Upton, G. Russell, S. Lightman - University of Bristol,Bristol (United Kingdom), P. Methlie, K. Simunkova, M. Grytaas, E. Husebye, M. Øksnes - University of Bergen,Bergen (Norway) and Haukeland University Hospital,Bergen (Norway), D. Vassiliadi, S. Tsagarakis - Evangelismos Hospital,Athens (Greece)

Pages

281 - 282

DOI

10.5162/eurosensors2026/SP1.1.4

ISBN

978-3-910600-12-6

Price

free

Abstract

Machine learning offers powerful opportunities to transform continuous sensor measurements into clinically or biologically predictions. However, translating raw sensor signals into robust diagnostic models requires careful consideration of temporal dynamics, feature engineering, latent-state modeling, and interpretability. In this tutorial, we present a generalizable pipeline for developing predictive diagnostic models from sensor measurements, illustrated through a real-world case study: the early detection of subclinical ketosis in dairy cows using breath acetone measured by gas sensors.