Diverse frequency band-based convolutional neural networks for tonic cold pain assessment using EEG

Abstract The purpose of this study is to present a novel classification framework, called diverse frequency band-based Convolutional Neural Networks (DFB-based ConvNets), which can objectively identify tonic cold pain states. To achieve this goal, scalp EEG data were recorded from 32 subjects under cold stimuli conditions. The proposed DFB-based ConvNets model is capable of classifying three classes of tonic pain: No pain, Moderate Pain, and Severe Pain. Firstly, the proposed method utilizes diverse frequency band-based inputs to learn temporal representations from different frequency bands of Electroencephalogram (EEG) which are expected to have more discriminative power. Then the derived features are concatenated to form a feature vector, which is fed into a fully-connected network for performing the classification task. Experimental results demonstrate that the proposed method successfully discriminates the tonic cold pain states. To show the superiority of the DFB-based ConvNets classifier, we compare our results with the state-of-the-art classifiers and show it has a competitive classification accuracy (97.37%). Moreover, these promising results may pave the way to use DFB-based ConvNets in clinical pain research.

[1]  S Seri,et al.  Quantitative EEG modifications during the Cold Water Pressor Test: hemispheric and hand differences. , 1994, International journal of psychophysiology : official journal of the International Organization of Psychophysiology.

[2]  S. Merkel,et al.  Pain assessment in the patient unable to self-report: position statement with clinical practice recommendations. , 2011, Pain management nursing : official journal of the American Society of Pain Management Nurses.

[3]  J. Smith,et al.  Tonic changes in alpha power during immersion of the hand in cold water. , 1991, Electroencephalography and clinical neurophysiology.

[4]  Saeid Sanei,et al.  Deep learning for epileptic intracranial EEG data , 2016, 2016 IEEE 26th International Workshop on Machine Learning for Signal Processing (MLSP).

[5]  David M. W. Powers,et al.  Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation , 2011, ArXiv.

[6]  K. Shen,et al.  Frequency-domain EEG source analysis for acute tonic cold pain perception , 2012, Clinical Neurophysiology.

[7]  M. Ingvar.  Pain and functional imaging. , 1999, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[8]  R. Høifødt,et al.  The effect of experimenter gender on autonomic and subjective responses to pain stimuli , 2007, PAIN.

[9]  F. D. Silva.  Neural mechanisms underlying brain waves: from neural membranes to networks. , 1991 .

[10]  Alon Sinai,et al.  Pain assessment by continuous EEG: Association between subjective perception of tonic pain and peak frequency of alpha oscillations during stimulation and at rest , 2010, Brain Research.

[11]  M. Kamdar,et al.  Principles of Analgesic Use in the Treatment of Acute Pain and Cancer Pain, Sixth Edition , 2010 .

[12]  Yodchanan Wongsawat,et al.  EEG-based pain estimation via fuzzy logic and polynomial kernel support vector machine , 2013, The 6th 2013 Biomedical Engineering International Conference.

[13]  Leontios J. Hadjileontiadis,et al.  EEG-Based Tonic Cold Pain Characterization Using Wavelet Higher Order Spectral Features , 2015, IEEE Transactions on Biomedical Engineering.

[14]  Alon Sinai,et al.  Tonic pain and continuous EEG: Prediction of subjective pain perception by alpha-1 power during stimulation and at rest , 2012, Clinical Neurophysiology.

[15]  Tzyy-Ping Jung,et al.  EEG-based prediction of driver's cognitive performance by deep convolutional neural network , 2016, Signal Process. Image Commun..

[16]  Ugur Halici,et al.  A novel deep learning approach for classification of EEG motor imagery signals , 2017, Journal of neural engineering.

[17]  Umberto Sabatini,et al.  A cerebral blood flow study on tonic pain activation in man , 1994, Pain.

[18]  Mohammad I. Daoud,et al.  EEG-based tonic cold pain recognition system using wavelet transform , 2017, Neural Computing and Applications.

[19]  H. Beecher,et al.  Experimental pharmacology and measurement of the subjective response. , 1952 .

[20]  Fei Wang,et al.  Poster paper: Predicting seizures from electroencephalography recordings: A knowledge transfer strategy , 2016, 2016 IEEE International Conference on Healthcare Informatics (ICHI).

[21]  Wolfram Burgard,et al.  Deep learning with convolutional neural networks for EEG decoding and visualization , 2017, Human brain mapping.

[22]  Lars Arendt-Nielsen,et al.  Topographic effects of tonic cutaneous nociceptive stimulation on human electroencephalograph , 2001, Neuroscience Letters.

[23]  Andrew C. N. Chen,et al.  Human pain responsivity in a tonic pain model: Psychological determinants , 1989, Pain.

[24]  Xiangzhou Wang,et al.  The offline feature extraction of four-class motor imagery EEG based on ICA and Wavelet-CSP , 2014, CCC 2014.

[25]  Caroline Schnakers,et al.  Pain assessment and management in disorders of consciousness , 2007, Current opinion in neurology.

[26]  Asbjørn Mohr Drewes,et al.  Characterization of cortical source generators based on electroencephalography during tonic pain , 2017, Journal of pain research.

[27]  Bin He,et al.  Quantifying and Characterizing Tonic Thermal Pain Across Subjects From EEG Data Using Random Forest Models , 2017, IEEE Transactions on Biomedical Engineering.

[28]  C. Graversen,et al.  Dynamic spectral indices of the electroencephalogram provide new insights into tonic pain , 2015, Clinical Neurophysiology.

[29]  Lars Arendt-Nielsen,et al.  Comparative cerebral responses to non-painful warm vs. cold stimuli in man: EEG power spectra and coherence. , 2005, International journal of psychophysiology : official journal of the International Organization of Psychophysiology.

[30]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[31]  S. Stone-Elander,et al.  Pain-related cerebral activation is altered by a distracting cognitive task , 2000, Pain.

[32]  E. Berner,et al.  Overconfidence as a cause of diagnostic error in medicine. , 2008, The American journal of medicine.

[33]  D. Price.  Psychological and neural mechanisms of the affective dimension of pain. , 2000, Science.

[34]  F. L. D. Silva,et al.  Basic mechanisms of cerebral rhythmic activities , 1990 .

[35]  Lars Arendt-Nielsen,et al.  Dynamic changes and spatial correlation of EEG activities during cold pressor test in man , 2002, Brain Research Bulletin.

[36]  T. Sejnowski,et al.  Removal of eye activity artifacts from visual event-related potentials in normal and clinical subjects , 2000, Clinical Neurophysiology.

[37]  Joelle Pineau,et al.  Learning Robust Features using Deep Learning for Automatic Seizure Detection , 2016, MLHC.

[38]  Andrew C. N. Chen,et al.  Topographic brain measures of human pain and pain responsivity , 1989, Pain.

[39]  R. Haddad,et al.  Multiresolution Signal Decomposition: Transforms, Subbands, and Wavelets , 1992 .

[40]  Yeong-Ray Wen,et al.  A Novel Fuzzy Pain Demand Index Derived From Patient-Controlled Analgesia for Postoperative Pain , 2007, IEEE Transactions on Biomedical Engineering.

[41]  Gang Pan,et al.  Remembered or Forgotten?—An EEG-Based Computational Prediction Approach , 2016, PloS one.

[42]  Sebastian Stober,et al.  Using Convolutional Neural Networks to Recognize Rhythm Stimuli from Electroencephalography Recordings , 2014, NIPS.

[43]  Sneh Anand,et al.  Electroencephalogram based quantitative estimation of pain for balanced anaesthesia , 2015 .

[44]  Shouqian Sun,et al.  Single-trial EEG classification of motor imagery using deep convolutional neural networks , 2017 .