Using Deep Networks for Scientific Discovery in Physiological Signals

Deep neural networks (DNN) have shown remarkable success in the classification of physiological signals. In this study we propose a method for examining to what extent does a DNN's performance rely on rediscovering existing features of the signals, as opposed to discovering genuinely new features. Moreover, we offer a novel method of "removing" a hand-engineered feature from the network's hypothesis space, thus forcing it to try and learn representations which are different from known ones, as a method of scientific exploration. We then build on existing work in the field of interpretability, specifically class activation maps, to try and infer what new features the network has learned. We demonstrate this approach using ECG and EEG signals. With respect to ECG signals we show that for the specific task of classifying atrial fibrillation, DNNs are likely rediscovering known features. We also show how our method could be used to discover new features, by selectively removing some ECG features and "rediscovering" them. We further examine how could our method be used as a tool for examining scientific hypotheses. We simulate this scenario by looking into the importance of eye movements in classifying sleep from EEG. We show that our tool can successfully focus a researcher's attention by bringing to light patterns in the data that would be hidden otherwise.

[1]  Jieyu Zhao,et al.  Balanced Datasets Are Not Enough: Estimating and Mitigating Gender Bias in Deep Image Representations , 2018, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[2]  Uri Shalit,et al.  Robust learning with the Hilbert-Schmidt independence criterion , 2019, ICML.

[3]  Bernhard Schölkopf,et al.  Measuring Statistical Dependence with Hilbert-Schmidt Norms , 2005, ALT.

[4]  Yong Jae Lee,et al.  Don’t Judge an Object by Its Context: Learning to Overcome Contextual Bias , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Madalena Costa,et al.  Multiscale entropy analysis of biological signals. , 2005, Physical review. E, Statistical, nonlinear, and soft matter physics.

[6]  Danny Eytan,et al.  Atrial fibrillation classification using step-by-step machine learning , 2018 .

[7]  Heiga Zen,et al.  WaveNet: A Generative Model for Raw Audio , 2016, SSW.

[8]  Danny Eytan,et al.  Towards Understanding ECG Rhythm Classification Using Convolutional Neural Networks and Attention Mappings , 2018, MLHC.

[9]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[10]  Jerome M. Siegel,et al.  Sleep viewed as a state of adaptive inactivity , 2009, Nature Reviews Neuroscience.

[11]  Anna Goldenberg,et al.  What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use , 2019, MLHC.

[12]  Stefan Carlsson,et al.  CNN Features Off-the-Shelf: An Astounding Baseline for Recognition , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

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

[14]  Tiago H. Falk,et al.  Deep learning-based electroencephalography analysis: a systematic review , 2019, Journal of neural engineering.

[15]  Gilles Vandewalle,et al.  Sleep slow wave changes during the middle years of life , 2011, The European journal of neuroscience.

[16]  Gal Chechik,et al.  A causal view of compositional zero-shot recognition , 2020, NeurIPS.

[17]  Eliyahu Kiperwasser,et al.  Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations , 2016, TACL.

[18]  Bernhard Schölkopf,et al.  Regression by dependence minimization and its application to causal inference in additive noise models , 2009, ICML '09.

[19]  Miad Faezipour,et al.  Sleep Stage Classification Using EEG Signal Analysis: A Comprehensive Survey and New Investigation , 2016, Entropy.

[20]  Sean L. Hill,et al.  The Sleep Slow Oscillation as a Traveling Wave , 2004, The Journal of Neuroscience.

[21]  S. Quan,et al.  Recommended Amount of Sleep for Pediatric Populations: A Consensus Statement of the American Academy of Sleep Medicine. , 2016, Journal of clinical sleep medicine : JCSM : official publication of the American Academy of Sleep Medicine.

[22]  A. Chesson,et al.  The American Academy of Sleep Medicine (AASM) Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications , 2007 .

[23]  Giovanni Calcagnini,et al.  P-wave Variability and Atrial Fibrillation , 2016, Scientific Reports.

[24]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[25]  Bolei Zhou,et al.  Learning Deep Features for Discriminative Localization , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[26]  Isaac Fernández-Varela,et al.  A Convolutional Network for the Classification of Sleep Stages , 2018, Proceedings.

[27]  Junmo Kim,et al.  Learning Not to Learn: Training Deep Neural Networks With Biased Data , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[28]  J. Samet,et al.  The Sleep Heart Health Study: design, rationale, and methods. , 1997, Sleep.

[29]  Andre Esteva,et al.  A guide to deep learning in healthcare , 2019, Nature Medicine.

[30]  Yoav Goldberg,et al.  Adversarial Removal of Demographic Attributes from Text Data , 2018, EMNLP.

[31]  Qiao Li,et al.  AF classification from a short single lead ECG recording: The PhysioNet/computing in cardiology challenge 2017 , 2017, 2017 Computing in Cardiology (CinC).

[32]  Andrew Y. Ng,et al.  CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning , 2017, ArXiv.

[33]  A. Chesson,et al.  The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology, and Techinical Specifications , 2007 .

[34]  Amjed S. Al-Fahoum,et al.  Methods of EEG Signal Features Extraction Using Linear Analysis in Frequency and Time-Frequency Domains , 2014, ISRN neuroscience.