Genetic-optimized classifier ensemble for cortisol salivary measurement mapping to electrocardiogram features for stress evaluation

Authors:
Chu Kiong Loo;Soon Fatt Cheong;Margaret A. Seldon;Ali Afzalian Mand;Kalaiarasi Sonai Muthu;Wei Shiung Liew;Einly Lim
Affiliations:
Faculty of Computer Science and Information Technology, University of Malaya, Kuala Lumpur, Malaysia;Faculty of Information Science and Technology, Multimedia University, Melaka, Malaysia;Faculty of Information Science and Technology, Multimedia University, Melaka, Malaysia;Faculty of Information Science and Technology, Multimedia University, Melaka, Malaysia;Faculty of Information Science and Technology, Multimedia University, Melaka, Malaysia;Faculty of Computer Science and Information Technology, University of Malaya, Kuala Lumpur, Malaysia;Faculty of Biomedical Engineering, University of Malaya, Kuala Lumpur, Malaysia
Venue:
PRICAI'12 Proceedings of the 12th Pacific Rim international conference on Trends in Artificial Intelligence
Year:
2012

Citing 6
Cited 0

Performance analysis of pattern classifier combination by plurality voting

Pattern Recognition Letters
Accurate and Reliable Diagnosis and Classification Using Probabilistic Ensemble Simplified Fuzzy ARTMAP

IEEE Transactions on Knowledge and Data Engineering
Using Heart Rate Monitors to Detect Mental Stress

BSN '09 Proceedings of the 2009 Sixth International Workshop on Wearable and Implantable Body Sensor Networks
Biased ART: A neural architecture that shifts attention toward previously disregarded features following an incorrect prediction

Neural Networks
Out of the lab and into the fray: towards modeling emotion in everyday life

Pervasive'10 Proceedings of the 8th international conference on Pervasive Computing
Monitoring of mental workload levels during an everyday life office-work scenario

Personal and Ubiquitous Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work presents our findings to map salivary cortisol measurements to electrocardiogram (ECG) features to create a physiological stress identification system. An experiment modelled on the Trier Social Stress Test (TSST) was used to simulate stress and control conditions, whereby salivary measurements and ECG measurements were obtained from student volunteers. The salivary measurements of stress biomarkers were used as objective stress measures to assign a three-class labelling (Low-Medium-High stress) to the extracted ECG features. The labelled features were then used for training and classification using a genetic-ordered ARTMAP with probabilistic voting for analysis on the efficacy of the ECG features used for physiological stress recognition. The ECG features include time-domain features of the heart rate variability and the ECG signal, and frequency-domain analysis of specific frequency bands related to the autonomic nervous activity. The resulting classification method scored approximately 60-69% success rate for predicting the three stress classes.