Original article
Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes

https://doi.org/10.1016/S0895-4356(96)00002-9Get rights and content

Abstract

Artificial neural networks are algorithms that can be used to perform nonlinear statistical modeling and provide a new alternative to logistic regression, the most commonly used method for developing predictive models for dichotomous outcomes in medicine. Neural networks offer a number of advantages, including requiring less formal statistical training, ability to implicitly detect complex nonlinear relationships between dependent and independent variables, ability to detect all possible interactions between predictor variables, and the availability of multiple training algorithms. Disadvantages include its “black box” nature, greater computational burden, proneness to overfitting, and the empirical nalure of model developmenl. An overview of the features of neural networks and logislic regression is presented, and the advantages and disadvanlages of using this modeling technique are discussed.

References (37)

  • WG Baxt

    Use of an artificial neural network for the diagnosis of myocardial infarction

    Ann Intern Med

    (1991)
  • TG Buchman et al.

    A comparison of statistical and connectionist models for the prediction of chronicity in a surgical intensive care unit

    Crit Care Med

    (1994)
  • S Derksen et al.

    Backward, forward and stepwise automated subset selection algorithms: Frequency of obtaining authentic and noise variables

    Br J Math Stat Psychol

    (1992)
  • DJ Spiegelhalter

    Probabilistic prediction in patient management and clinical trials

    Stat Med

    (1986)
  • K Liestol et al.

    Survival analysis and neural nets

    Stat Med

    (1994)
  • DE Rumelhart et al.

    Learning representations by back-propagating errors

    Nature (London)

    (1986)
  • JL McClelland et al.

    Training hidden units: The generalized delta rule

  • JM Zurada

    Introduction to Artificial Neural Systems

    (1992)
  • Cited by (1403)

    • Modeling real-world diesel car tailpipe emissions using regression-based approaches

      2024, Transportation Research Part D: Transport and Environment
    View all citing articles on Scopus

    This work was supported in part by a seed grant from the Center for Risk Analysis, Harvard School of Public Health.

    1

    Dr. Tu was supported by a Health Research Personnel Development Program Fellowship (04544) from the Ontario Ministry of Health. The results and conclusions are those of the author, and no official endorsement by the Ministry of Health is intended or should be inferred.

    View full text