Brute Force ECG Feature Extraction Applied on Discomfort Detection

This paper presents the idea of brute force feature extraction for Electrocardiography (ECG) signals applied to discomfort detection. To build an ECG Discomfort Corpus an experimental discomfort induction was conducted. 50 subjects underwent a 2 h (dis-)comfort condition in separate sessions in randomized order. ECG and subjective discomfort was recorded. 5 min ECG segments were labeled with corresponding subjective discomfort ratings, and 6365 brute force features (65 low-level descriptors, first and second order derivatives, and 47 functionals) and 11 traditional heart rate variability (HRV) parameters were extracted. Random Forest machine learning algorithm outperformed SVM and kNN approaches and achieved the best subject-dependent, 10-fold cross-validation results (\(r=.51\)). With this experiment, we are able to show that (a) brute force ECG feature sets achieved better discomfort detection than traditional HRV based ECG feature set; (b) cepstral and spectral flux based features appear to be the most promising to capture HRV phenomena.

[1]  Ilaria Gabbatore,et al.  Sincere, Deceitful, and Ironic Communicative Acts and the Role of the Theory of Mind in Childhood , 2017, Front. Psychol..

[2]  Fabio Valente,et al.  The INTERSPEECH 2013 computational paralinguistics challenge: social signals, conflict, emotion, autism , 2013, INTERSPEECH.

[3]  Mohamed Atibi,et al.  ECG signals classification using MFCC coefficients and ANN classifier , 2016, 2016 International Conference on Electrical and Information Technologies (ICEIT).

[4]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[5]  J. Thayer,et al.  Heart Rate Variability and Cardiac Vagal Tone in Psychophysiological Research – Recommendations for Experiment Planning, Data Analysis, and Data Reporting , 2017, Front. Psychol..

[6]  H. G. Rauch,et al.  The effect of short duration heart rate variability (HRV) biofeedback on cognitive performance during laboratory induced cognitive stress , 2011 .

[7]  G.B. Moody,et al.  PhysioNet: a Web-based resource for the study of physiologic signals , 2001, IEEE Engineering in Medicine and Biology Magazine.

[8]  Ioanna Chouvarda,et al.  EEG and HRV markers of sleepiness and loss of control during car driving , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[9]  George Trigeorgis,et al.  The INTERSPEECH 2017 Computational Paralinguistics Challenge: Addressee, Cold & Snoring , 2017, INTERSPEECH.

[10]  Klaus Gröschel,et al.  Holter-electrocardiogram-monitoring in patients with acute ischaemic stroke (Find-AFRANDOMISED): an open-label randomised controlled trial , 2017, The Lancet Neurology.

[11]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[12]  Björn Schuller,et al.  Opensmile: the munich versatile and fast open-source audio feature extractor , 2010, ACM Multimedia.

[13]  Elizabeth Jane M Pearson,et al.  Comfort and its measurement – A literature review , 2009, Disability and rehabilitation. Assistive technology.

[14]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[15]  Thierry Pun,et al.  DEAP: A Database for Emotion Analysis ;Using Physiological Signals , 2012, IEEE Transactions on Affective Computing.

[16]  Björn W. Schuller,et al.  Applying multiple classifiers and non-linear dynamics features for detecting sleepiness from speech , 2012, Neurocomputing.

[17]  R Aissaoui,et al.  Evaluation of the new flexible contour backrest for wheelchairs. , 2000, Journal of rehabilitation research and development.

[18]  Björn Schuller,et al.  The Computational Paralinguistics Challenge , 2012 .

[19]  Erik Cambria,et al.  A review of affective computing: From unimodal analysis to multimodal fusion , 2017, Inf. Fusion.

[20]  R Wachter,et al.  Vorhofflimmern besser erkennen, hilft Schlaganfälle zu verhindern , 2017 .

[21]  U. Rajendra Acharya,et al.  Automated diagnosis of congestive heart failure using dual tree complex wavelet transform and statistical features extracted from 2 s of ECG signals , 2017, Comput. Biol. Medicine.

[22]  Fabien Ringeval,et al.  AVEC 2016: Depression, Mood, and Emotion Recognition Workshop and Challenge , 2016, AVEC@ACM Multimedia.

[23]  Florian Eyben,et al.  Real-time Speech and Music Classification by Large Audio Feature Space Extraction , 2015 .

[24]  Yuta Suzuki,et al.  Heart rate variability as a predictive biomarker of thermal comfort , 2018, J. Ambient Intell. Humaniz. Comput..

[25]  J F Thayer,et al.  Heart rate variability and experimentally induced pain in healthy adults: A systematic review , 2014, European journal of pain.

[26]  David W. Aha,et al.  Instance-Based Learning Algorithms , 1991, Machine Learning.

[27]  Soma Bandyopadhyay,et al.  Identifying normal, AF and other abnormal ECG rhythms using a cascaded binary classifier , 2017, 2017 Computing in Cardiology (CinC).