On the misuses of artificial neural networks for prognostic and diagnostic classification in oncology.
作者： Guido Schwarzer , Werner Vach , Martin Schumacher
DOI: 10.1002/(SICI)1097-0258(20000229)19:4<541::AID-SIM355>3.0.CO;2-V
关键词:
摘要: The application of artificial neural networks (ANNs) for prognostic and diagnostic classification in clinical medicine has become very popular. In particular, feed-forward have been used extensively, often accompanied by exaggerated statements their potential. this paper, the essentials statistical counterparts (that is, logistic regression models) are reviewed. We point out that uncritical use ANNs may lead to serious problems, such as fitting implausible functions describe probability class membership underestimation misclassification probabilities. applications survival data, further difficulties arise. Finally, results a search medical literature from 1991 1995 on oncology some important common mistakes reported. It is concluded there no evidence so far represents real progress field diagnosis prognosis oncology.
sci-hub.se 参考文章(90)
John S. Bridle, Probabilistic Interpretation of Feedforward Classification Network Outputs, with Relationships to Statistical Pattern Recognition NATO Neurocomputing. pp. 227- 236 ,(1990) , 10.1007/978-3-642-76153-9_28
Brian D. Ripley, Statistical aspects of neural networks Springer US. pp. 40- 123 ,(1993) , 10.1007/978-1-4899-3099-6_2
B S Erler, A M Marchevsky, S N Thung, L D Ferrell, S A Geller, H M Truong, M H Huh, L M Petrovic, S S Kim, L Hsu, Image analysis and diagnostic classification of hepatocellular carcinoma using neural networks and multivariate discriminant functions. Laboratory Investigation. ,vol. 71, pp. 446- 451 ,(1994)
Domenic V Cicchetti, Neural networks and diagnosis in the clinical laboratory: state of the art. Clinical Chemistry. ,vol. 38, pp. 9- 10 ,(1992) , 10.1093/CLINCHEM/38.1.9
Peter B. Snow, Deborah S. Smith, William J. Catalona, Artificial Neural Networks in the Diagnosis and Prognosis of Prostate Cancer: A Pilot Study Journal of Urology. ,vol. 152, pp. 1923- 1926 ,(1994) , 10.1016/S0022-5347(17)32416-3
Michael L Astion, Peter Wilding, Application of neural networks to the interpretation of laboratory data in cancer diagnosis. Clinical Chemistry. ,vol. 38, pp. 34- 38 ,(1992) , 10.1093/CLINCHEM/38.1.34
David R. Cox, Regression Models and Life-Tables Springer Series in Statistics. ,vol. 34, pp. 527- 541 ,(1992) , 10.1007/978-1-4612-4380-9_37
An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. The New England Journal of Medicine. ,vol. 329, pp. 673- 682 ,(1993) , 10.1056/NEJM199309023291001
Christopher Desilva, Y. Attikiouzel, Applications of neural networks in medicine. Australasian Physical & Engineering Sciences in Medicine. ,vol. 18, pp. 158- 164 ,(1995)
S. Kullback, R. A. Leibler, On Information and Sufficiency Annals of Mathematical Statistics. ,vol. 22, pp. 79- 86 ,(1951) , 10.1214/AOMS/1177729694