Model selection for a medical diagnostic decision
support system: a breast cancer detection case
, Vivian West
Department of Decision Sciences, College of Business Administration, East Carolina University,
Greenville, NC 27836, USA
East Carolina University Center for Health Sciences Communication, Greenville, NC 27858, USA
School of Nursing, University of North Carolina, Chapel Hill, NC 27599, USA
Received 28 October 1999; received in revised form 28 February 2000; accepted 6 March 2000
There are a number of different quantitative models that can be used in a medical diagnostic
decision support system (MDSS) including parametric methods (linear discriminant analysis or
logistic regression), non-parametric models (K nearest neighbor, or kernel density) and several
neural network models. The complexity of the diagnostic task is thought to be one of the prime
determinants of model selection. Unfortunately, there is no theory available to guide model
selection. Practitioners are left to either choose a favorite model or to test a small subset using cross
validation methods. This paper illustrates the use of a self-organizing map (SOM) to guide model
selection for a breast cancer MDSS. The topological ordering properties of the SOM are used to
de®ne targets for an ideal accuracy level similar to a Bayes optimal level. These targets can then be
used in model selection, variable reduction, parameter determination, and to assess the adequacy of the
clinical measurement system. These ideas are applied to a successful model selection for a real-world
breast cancer database. Diagnostic accuracy results are reported for individual models, for ensembles of
neural networks, and for stacked predictors. # 2000 Published by Elsevier Science B.V.
Keywords: Self-organizing map; Model selection; Decision support system; Neural network; Stacked
Medical diagnostic decision support systems (MDSS's) have become an established
component of medical technology and their use will continue to grow, fueled by electronic
medical records and automatic data capture [26,37]. The purpose of an MDSS is to
Artificial Intelligence in Medicine 20 (2000) 183±204
Corresponding author. Tel.: 1-252-328-6370; fax: 1-252-321-6195.
E-mail addresses: firstname.lastname@example.org, email@example.com (D. West).
0933-3657/00/$ ± see front matter # 2000 Published by Elsevier Science B.V.
PII: S 0933-3657(00)00063-4