Insightful stress detection from physiology modalities using Learning Vector Quantization

J. J. G. (Gert-Jan) de Vries*, Steffen C. Pauws, Michael Biehl

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

27 Citations (Scopus)

Abstract

Stress in daily life can lead to severe conditions as burn-out and depression and has a major impact on society. Being able to measure mental stress reliably opens up the ability to intervene in an early stage. We performed a large-scale study in which skin conductance, respiration and electrocardiogram were measured in semi-controlled conditions. Using Learning Vector Quantization techniques, we obtained up to 88% accuracy in the classification task to separate stress from relaxation. Relevance learning was used to identify the most informative features, indicating that most information is embedded in the cardiac signals. In addition to commonly used features, we also explored various novel features, of which the very-high frequency band of the power spectrum was found to be a very relevant addition. (C) 2014 Elsevier B.V. All rights reserved.

Original languageEnglish
Pages (from-to)873-882
Number of pages10
JournalNeurocomputing
Volume151
DOIs
Publication statusPublished - 5 Mar 2015
Externally publishedYes

Keywords

  • Mental stress
  • Classification
  • Learning Vector Quantization
  • HEART-RATE-VARIABILITY
  • RESPIRATORY SINUS ARRHYTHMIA
  • EMOTION RECOGNITION
  • MENTAL STRESS
  • CLINICAL-USE
  • SIGNALS
  • SENSORS
  • FREQUENCY
  • STANDARDS
  • ACCURACY

Cite this