Explaining artificial neural network ensembles: A case study with electrocardiograms from chest pain patients
作者： Ulf Ekelund , Jonas Björk , Mattias Ohlsson , Lars Edenbrandt , Michael Green
DOI:
关键词:
摘要: Artificial neural networks is one of the most commonly used machine learning algorithms in medical applications. However, they are still not practice clinics partly due to their lack explanatory capacity. We compare two case-based explanation methods trained physicians on analysis electrocardiogram (ECG) data from patients with a suspected acute coronary syndrome (ACS). The median overlaps top 5 selected features between physicians, and given physician method, were initially low. Using correlation overlap increased values typically range 3-4. In conclusion, both our generate explanations similar those expert problem diagnosing ACS ECG data.
参考文章(27)
Rich Caruana, Case-Based Explanation for Artificial Neural Nets Artificial Neural Networks in Medicine and Biology. pp. 303- 308 ,(2000) , 10.1007/978-1-4471-0513-8_46
Thomas G. Dietterich, Ensemble Methods in Machine Learning Multiple Classifier Systems. pp. 1- 15 ,(2000) , 10.1007/3-540-45014-9_1
Ron Kohavi, A study of cross-validation and bootstrap for accuracy estimation and model selection international joint conference on artificial intelligence. ,vol. 2, pp. 1137- 1143 ,(1995)
Jonas Björk, Jakob L Forberg, Mattias Ohlsson, Lars Edenbrandt, Hans Öhlin, Ulf Ekelund, A simple statistical model for prediction of acute coronary syndrome in chest pain patients in the emergency department BMC Medical Informatics and Decision Making. ,vol. 6, pp. 28- 28 ,(2006) , 10.1186/1472-6947-6-28
W. Wang, P. Jones, D. Partridge, Assessing the Impact of Input Features in a Feedforward Neural Network Neural Computing and Applications. ,vol. 9, pp. 101- 112 ,(2000) , 10.1007/PL00009895
William G. Baxt, Frances S. Shofer, Frank D. Sites, Judd E. Hollander, A neural computational aid to the diagnosis of acute myocardial infarction. Annals of Emergency Medicine. ,vol. 39, pp. 366- 373 ,(2002) , 10.1067/MEM.2002.122705
Sven-Erik Olsson, Mattias Ohlsson, Hans Ohlin, Samir Dzaferagic, Marie-Louise Nilsson, Per Sandkull, Lars Edenbrandt, Decision support for the initial triage of patients with acute coronary syndromes. Clinical Physiology and Functional Imaging. ,vol. 26, pp. 151- 156 ,(2006) , 10.1111/J.1475-097X.2006.00669.X
P.J.G. Lisboa, A review of evidence of health benefit from artificial neural networks in medical intervention Neural Networks. ,vol. 15, pp. 11- 39 ,(2002) , 10.1016/S0893-6080(01)00111-3
Robert Wall, Pádraig Cunningham, Paul Walsh, Stephen Byrne, Explaining the output of ensembles in medical decision support on a case by case basis Artificial Intelligence in Medicine. ,vol. 28, pp. 191- 206 ,(2003) , 10.1016/S0933-3657(03)00056-3
J. J. Monta�o, A. Palmer, Numeric sensitivity analysis applied to feedforward neural networks Neural Computing and Applications. ,vol. 12, pp. 119- 125 ,(2003) , 10.1007/S00521-003-0377-9