Comparing Performances of Logistic Regression, Classification & Regression Trees and Artificial Neural Networks for Predicting Albuminuria in Type 2 Diabetes Mellitus

In this study, performances of classification methods were compared in order to predict the presence of albuminuria in type 2 diabetes mellitus patients. A retrospective analysis was performed in 266 subjects. We compared performances of logistic regression (LR), classification and regression trees (C&RT) and two artificial neural networks algorithms. Predictor variables were gender, urine creatinine, weight, blood urea, serum albumin, age, creatinine clearance, fasting plasma glucose, post-prandial plasma glucose, and HbA1c. For validation set, the best classification accuracy (84.85%), sensitivity (68.0%) and the highest Youden index (0.63) was found in the MLP model but the specificity was 95.12%. Additionally, the specificity of all the models was close to each other. For whole data set the results were found as 84.21%, 53.95%, 0.50 and 96.32% respectively. Consequently, the model had the highest predictive capability to predict the presence of albuminuria was MLP. According to this model, blood urea and serum albumin were the most important variables for predicting the albuminuria. On the basis of these considerations, we suggest that data should be better explored and processed by high performance modeling methods. Researchers should avoid assessment of data by using only one method in future studies focusing on albuminuria in type 2 diabetes mellitus patients or any other clinical condition.

[1]  Tahseen Ahmed Jilani,et al.  Hepatitis-C Classification using Data Mining Techniques , 2011 .

[2]  M.H. Hassoun,et al.  Fundamentals of Artificial Neural Networks , 1996, Proceedings of the IEEE.

[3]  P Zimmet,et al.  International Diabetes Federation: a consensus on Type 2 diabetes prevention , 2007, Diabetic medicine : a journal of the British Diabetic Association.

[4]  Mevlut Ture,et al.  Comparing classification techniques for predicting essential hypertension , 2005, Expert Syst. Appl..

[5]  Roberto Hornero,et al.  Comparison of logistic regression and neural network classifiers in the detection of hard exudates in retinal images , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[6]  Patrick van der Smagt,et al.  Introduction to neural networks , 1995, The Lancet.

[7]  Reza Karimi,et al.  Inconsistency in albuminuria predictors in type 2 diabetes: a comparison between neural network and conditional logistic regression. , 2013, Translational research : the journal of laboratory and clinical medicine.

[8]  I. Kurt,et al.  The Effect of Health Status, Nutrition, and Some Other Factors on Low School Performance Using Induction Technique , 2006 .

[9]  Mevlut Ture,et al.  Comparing performances of logistic regression, classification and regression tree, and neural networks for predicting coronary artery disease , 2008, Expert Syst. Appl..

[10]  Hiroshi Tanaka,et al.  Comparison of Seven Algorithms to Predict Breast Cancer Survival( Contribution to 21 Century Intelligent Technologies and Bioinformatics) , 2008 .

[11]  Engin Güney,et al.  The relationship between mean platelet volume with microalbuminuria and glycemic control in patients with type II diabetes mellitus , 2012, Platelets.

[12]  Xuehui Meng,et al.  Comparison of three data mining models for predicting diabetes or prediabetes by risk factors , 2013, The Kaohsiung journal of medical sciences.

[13]  Aesha Drozdowski,et al.  Standards of medical care in diabetes. , 2004, Diabetes care.

[14]  Adrian G. Bors,et al.  Introduction of the Radial Basis Function (RBF) Networks , 2000 .

[15]  Antonino Staiano,et al.  A multilayer perceptron neural network-based approach for the identification of responsiveness to interferon therapy in multiple sclerosis patients , 2010, Inf. Sci..

[16]  David V. Power,et al.  Standards of Medical Care in Diabetes: Response to position statement of the American Diabetes Association , 2006 .

[17]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[18]  D. Hosmer,et al.  Applied Logistic Regression , 1991 .

[19]  D. E. Busby,et al.  Comparison of Commonly Used Assays for the Detection of Microalbuminuria , 2004, Journal of clinical hypertension.

[20]  Bendi Venkata Ramana,et al.  Liver Classification Using Modified Rotation Forest , 2012 .

[21]  S. Bakker,et al.  Macroalbuminuria and microalbuminuria: do both predict renal and cardiovascular events with similar strength? , 2007, Journal of nephrology.

[22]  Martin Kiefel,et al.  Quasi-Newton Methods: A New Direction , 2012, ICML.

[23]  M. Radmacher,et al.  Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification. , 2003, Journal of the National Cancer Institute.

[24]  Christian Borgelt,et al.  Introduction to Neural Networks , 2016 .

[25]  N. B. Venkateswarlu,et al.  A Critical Study of Selected Classification Algorithms for Liver Disease Diagnosis , 2011 .

[26]  Dursun Delen,et al.  A machine learning-based approach to prognostic analysis of thoracic transplantations , 2010, Artif. Intell. Medicine.

[27]  João Maroco,et al.  Prediction of Dementia Patients: A Comparative Approach Using Parametric Versus Nonparametric Classifiers , 2013 .

[28]  Kemal Polat,et al.  A cascade learning system for classification of diabetes disease: Generalized Discriminant Analysis and Least Square Support Vector Machine , 2008, Expert Syst. Appl..

[29]  D. Zeeuw.  Albuminuria: a target for treatment of type 2 diabetic nephropathy. , 2007 .

[30]  V. Kvasnicka,et al.  Neural and Adaptive Systems: Fundamentals Through Simulations , 2001, IEEE Trans. Neural Networks.

[31]  Sophia Decker,et al.  Logistic Regression A Self Learning Text , 2016 .