We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Using Deep Networks for Scientific Discovery in Physiological Signals

Abstract: 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.
Comments: 2020 Machine Learning for Healthcare Conference
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
Cite as: arXiv:2008.10936 [stat.ML]
  (or arXiv:2008.10936v1 [stat.ML] for this version)

Submission history

From: Tom Beer [view email]
[v1] Tue, 25 Aug 2020 10:55:25 GMT (6493kb,D)

Link back to: arXiv, form interface, contact.