Abstract
Purpose. For individualized followup, accurate prediction of locoregional recurrence (LRR) and second primary (SP) breast cancer risk is required. Current prediction models employ regression, but with large data sets, machinelearning techniques such as Bayesian Networks (BNs) may be better alternatives. In this study, logistic regression was compared with different BNs, built with network classifiers and constraint and scorebased algorithms. Methods. Women diagnosed with early breast cancer between 2003 and 2006 were selected from the Netherlands Cancer Registry (NCR) (N = 37,320). BN structures were developed using 1) Bayesian network classifiers, 2) correlation coefficients with different cutoffs, 3) constraintbased learning algorithms, and 4) scorebased learning algorithms. The different models were compared with logistic regression using the area under the receiver operating characteristic curve, an external validation set obtained from the NCR from 2007 and 2008 (N = 12,308), and subgroup analyses for a high and lowrisk group. Results. The BNs with the most links showed the best performance in both LRR and SP prediction (cstatistic of 0.76 for LRR and 0.69 for SP). In the external validation, logistic regression generally outperformed the BNs in both SP and LRR (cstatistic of 0.71 for LRR and 0.64 for SP). The differences were nonetheless small. Although logistic regression performed best on most parts of the subgroup analysis, BNs outperformed regression with respect to average risk for SP prediction in low and highrisk groups. Conclusions. Although estimates of regression coefficients depend on other independent variables, there is no assumed dependence relationship between coefficient estimators and the change in value of other variables as in the case of BNs. Nonetheless, this analysis suggests that regression is still more accurate or at least as accurate as BNs for risk estimation for both LRRs and SP tumors.
Original language  English 

Pages (fromto)  822833 
Number of pages  12 
Journal  Medical Decision Making 
Volume  38 
Issue number  7 
DOIs  
Publication status  Published  1 Oct 2018 
Event  38th Annual North American Meeting  Vancouver, Canada Duration: 23 Oct 2016 → 26 Oct 2016 http://smdm.org/meeting/38thannualnorthamericanmeeting 
Bibliographical note
Presented at the 38th Annual North American Meeting, Vancouver, CanadaKeywords
 Bayesian network
 breast cancer
 locoregional recurrence
 logistic regression
 machine learning
 risk prediction
 second primary
 Breast Neoplasms/pathology
 Humans
 Middle Aged
 Neoplasm Recurrence, Local
 Risk Assessment/statistics & numerical data
 Logistic Models
 Machine Learning
 Netherlands
 Algorithms
 Bayes Theorem
 Female
 ROC Curve
 Registries
Fingerprint
Dive into the research topics of 'Comparison of logistic regression and Bayesian networks for risk prediction of breast cancer recurrence'. Together they form a unique fingerprint.Datasets

Comparison of Logistic Regression and Bayesian Networks for Risk Prediction of Breast Cancer Recurrence
Witteveen, A. (Creator), Nane, G. F. (Creator), Vliegen, I. M. H. (Creator), Siesling, S. (Creator) & IJzerman, M. J. (Creator), Figshare, 2018
DOI: 10.25384/sage.c.4208183, https://figshare.com/collections/Comparison_of_Logistic_Regression_and_Bayesian_Networks_for_Risk_Prediction_of_Breast_Cancer_Recurrence/4208183 and one more link, https://figshare.com/collections/Comparison_of_Logistic_Regression_and_Bayesian_Networks_for_Risk_Prediction_of_Breast_Cancer_Recurrence/4208183/1 (show fewer)
Dataset