Your browser has javascript turned off or blocked. This will lead to some parts of our website to not work properly or at all. Turn on javascript for best performance.

The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here: https://www.microsoft.com/en-us/microsoft-365/windows/end-of-ie-support).

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Mattias Ohlsson

Mattias Ohlsson

Professor

Mattias Ohlsson

Exploring new possibilities for case based explanation of artificial neural network ensembles

Author

  • Michael Green
  • Ulf Ekelund
  • Lars Edenbrandt
  • Jonas Björk
  • Jakob Lundager Hansen
  • Mattias Ohlsson

Summary, in English

Artificial neural network (ANN) ensembles have long suffered from a lack of interpretability. This has severely limited the practical usability of ANNs in settings where an erroneous decision can be disastrous. Several attempts have been made to alleviate this problem. Many of them are based on decomposing the decision boundary of the ANN into a set of rules. We explore and compare a set of new methods for this explanation process on two artificial data sets (Monks 1 and 3), and one acute coronary syndrome data set consisting of 861 electrocardiograms (ECG) collected retrospectively at the emergency department at Lund University Hospital. The algorithms managed to extract good explanations in more than 84% of the cases. More to the point, the best method provided 99% and 91% good explanations in Monks data 1 and 3 respectively. Also there was a significant overlap between the algorithms. Furthermore, when explaining a given ECG, the overlap between this method and one of the physicians was the same as the one between the two physicians in this study. Still the physicians were significantly, p-value <0.001, more similar to each other than to any of the methods. The algorithms have the potential to be used as an explanatory aid when using ANN ensembles in clinical decision support systems.

Department/s

  • Computational Biology and Biological Physics
  • Medicine, Lund
  • Nuclear medicine, Malmö
  • Centre for Economic Demography
  • Division of Occupational and Environmental Medicine, Lund University

Publishing year

2009

Language

English

Pages

75-81

Publication/Series

Neural Networks

Volume

22

Issue

1

Document type

Journal article

Publisher

Elsevier

Topic

  • Radiology, Nuclear Medicine and Medical Imaging
  • Anesthesiology and Intensive Care

Keywords

  • Neural Network Ensembles
  • Acute Coronary Syndrome
  • Case-Based Explanation
  • Sensitivity Analysis

Status

Published

Project

  • AIR Lund Chest pain - More efficient and equal emergency care with advanced medical decision support tools

Research group

  • Nuclear medicine, Malmö

ISBN/ISSN/Other

  • ISSN: 1879-2782