Advanced

Explaining artificial neural network ensembles: A case study with electrocardiograms from chest pain patients

Green, Michael LU ; Ekelund, Ulf LU ; Edenbrandt, Lars LU ; Björk, Jonas LU ; Lundager Hansen, Jakob LU and Ohlsson, Mattias LU (2008) International Conference on Machine Learning In Proceedings of the ICML/UAI/COLT 2008 Workshop on Machine Learning for Health-Care Applications
Abstract
Artificial neural networks is one of the most commonly used machine learning algorithms in medical applications. However, they are still not used in practice in the clinics partly due to their lack of explanatory capacity. We compare two case-based explanation methods to two trained physicians on analysis of electrocardiogram (ECG) data from patients with a suspected acute coronary syndrome (ACS). The median overlaps of the top 5 selected features between the two physicians, and a given physician and a method, were initially low. Using a correlation analysis of the features the median overlap increased to values typically in the range 3-4. In conclusion, both our case-based methods generate explanations similar to those of trained expert... (More)
Artificial neural networks is one of the most commonly used machine learning algorithms in medical applications. However, they are still not used in practice in the clinics partly due to their lack of explanatory capacity. We compare two case-based explanation methods to two trained physicians on analysis of electrocardiogram (ECG) data from patients with a suspected acute coronary syndrome (ACS). The median overlaps of the top 5 selected features between the two physicians, and a given physician and a method, were initially low. Using a correlation analysis of the features the median overlap increased to values typically in the range 3-4. In conclusion, both our case-based methods generate explanations similar to those of trained expert physicians on the problem of diagnosing ACS from ECG data. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
acute coronary syndrome, case-based explanation, rule extraction, neural network ensembles
in
Proceedings of the ICML/UAI/COLT 2008 Workshop on Machine Learning for Health-Care Applications
editor
Hauskrecht, Milos
pages
8 pages
conference name
International Conference on Machine Learning
language
English
LU publication?
yes
id
4267bd79-089b-4369-9846-94c9b5f99a57 (old id 1153067)
date added to LUP
2008-07-10 09:53:40
date last changed
2016-04-16 10:51:25
@misc{4267bd79-089b-4369-9846-94c9b5f99a57,
  abstract     = {Artificial neural networks is one of the most commonly used machine learning algorithms in medical applications. However, they are still not used in practice in the clinics partly due to their lack of explanatory capacity. We compare two case-based explanation methods to two trained physicians on analysis of electrocardiogram (ECG) data from patients with a suspected acute coronary syndrome (ACS). The median overlaps of the top 5 selected features between the two physicians, and a given physician and a method, were initially low. Using a correlation analysis of the features the median overlap increased to values typically in the range 3-4. In conclusion, both our case-based methods generate explanations similar to those of trained expert physicians on the problem of diagnosing ACS from ECG data.},
  author       = {Green, Michael and Ekelund, Ulf and Edenbrandt, Lars and Björk, Jonas and Lundager Hansen, Jakob and Ohlsson, Mattias},
  editor       = {Hauskrecht, Milos},
  keyword      = {acute coronary syndrome,case-based explanation,rule extraction,neural network ensembles},
  language     = {eng},
  pages        = {8},
  series       = {Proceedings of the ICML/UAI/COLT 2008 Workshop on Machine Learning for Health-Care Applications},
  title        = {Explaining artificial neural network ensembles: A case study with electrocardiograms from chest pain patients},
  year         = {2008},
}