A Comparison of Cox Regression and Neural Networks for Risk Stratification in Cases of Acute Lymphoblastic Leukaemia in Children

For most diseases there is considerable interest in the problem of classification, both in relation to medical diagnosis and for prognosis. Multivariate statistical methods are conventionally used as an aid to clinical decision making. Neural Networks (NNs) offer an alternative approach to this type of classification problem. Exploiting 1271 cases from the United Kingdom Medical Research Council UKALL X trial for childhood Acute Lymphoblastic Leukaemia (ALL), cases were stratified as ‘high risk’ or ‘standard risk’ using both the survival analysis technique of Cox Regression and trained neural networks. Based on 10 random trials with a further 300 cases, and predicting overall five year survival from age, sex and white cell count only, there was no significant difference between the two approaches in terms of mean Receiver Operating Characteristic area, though the regression model was slightly superior to a single neural network at high sensitivity (Wilcoxon signed rank test; p = 0.033). A composite of two networks, one of which included additional prognostic factors, restored the position of no significant difference. It was concluded that in the UKALL X dataset, factors predictive of outcome are fully described by a Cox regression analysis, and that a neural network-based analysis identified no additional prognostic features. The value of the network analysis lay in suggesting that the maximum amount of prognostic information has been extracted from the database.

[1]  David G. Bounds,et al.  A comparison of neural network and other pattern recognition approaches to the diagnosis of low back disorders , 1990, Neural Networks.

[2]  C. C. Bailey,et al.  Intensification of treatment and survival in all children with lymphoblastic leukaemia: results of UK Medical Research Council trial UKALL X , 1995, The Lancet.

[3]  J. Peto,et al.  Results of Medical Research Council Childhood Leukaemia Trial UKALL VIII (Report to the Medical Research Council on behalf of the Working Party on Leukaemia in Childhood) , 1991, British journal of haematology.

[4]  M De Laurentiis,et al.  A technique for using neural network analysis to perform survival analysis of censored data. , 1994, Cancer letters.

[5]  R. Gelber,et al.  Comparative results of two intensive treatment programs for childhood acute lymphoblastic leukemia: The Berlin-Frankfurt-Münster and Dana-Farber Cancer Institute protocols. , 1991, Annals of oncology : official journal of the European Society for Medical Oncology.

[6]  D Cox,et al.  The contribution of statistical methods to cancer research , 1991 .

[7]  C C Bailey,et al.  Gender and treatment outcome in childhood lymphoblastic leukaemia: report from the MRC UKALL trials * , 1995, British journal of haematology.

[8]  Yoshua Bengio,et al.  Pattern Recognition and Neural Networks , 1995 .

[9]  D Faraggi,et al.  A neural network model for survival data. , 1995, Statistics in medicine.

[10]  D. Kleinbaum,et al.  Survival Analysis: A Self-Learning Text. , 1996 .

[11]  R Mastrangelo,et al.  Report and recommendations of the Rome workshop concerning poor-prognosis acute lymphoblastic leukemia in children: biologic bases for staging, stratification, and treatment. , 1986, Medical and pediatric oncology.

[12]  J. Wyatt,et al.  Commentary: Prognostic models: clinically useful or quickly forgotten? , 1995 .

[13]  W. Baxt,et al.  Prospective validation of artificial neural network trained to identify acute myocardial infarction , 1996, The Lancet.

[14]  L. Frankel,et al.  Ploidy of lymphoblasts is the strongest predictor of treatment outcome in B-progenitor cell acute lymphoblastic leukemia of childhood: a Pediatric Oncology Group study. , 1992, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[15]  J. M. Chessells,et al.  A Comparison of Cox Regression and Neural Networks for Risk Stratification in Cases of Acute Lymphoblastic Leukaemia in Children , 1999, Neural computing & applications (Print).

[16]  Martin Bland,et al.  An Introduction to Medical Statistics , 1987 .

[17]  M. S. Chesters,et al.  Human visual perception and ROC methodology in medical imaging. , 1992, Physics in medicine and biology.

[18]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..