Comparison of gestational diabetes prediction with artificial neural network and decision tree models
Background: Gestational diabetes mellitus (GDM) is one of the most common metabolic disorders in pregnancy, which is associated with serious complications. In the event of early diagnosis of this disease, some of the maternal and fetal complications can be prevented. The aim of this study was to ear...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | fas |
| Published: |
Tehran University of Medical Sciences
2019-09-01
|
| Series: | Tehran University Medical Journal |
| Subjects: | |
| Online Access: | http://tumj.tums.ac.ir/article-1-9923-en.html |
| _version_ | 1819008608707280896 |
|---|---|
| author | Mansour Rezaei Negin Fakhri Fateme Rajati Soodeh Shahsavari |
| author_facet | Mansour Rezaei Negin Fakhri Fateme Rajati Soodeh Shahsavari |
| author_sort | Mansour Rezaei |
| collection | DOAJ |
| description | Background: Gestational diabetes mellitus (GDM) is one of the most common metabolic disorders in pregnancy, which is associated with serious complications. In the event of early diagnosis of this disease, some of the maternal and fetal complications can be prevented. The aim of this study was to early predict gestational diabetes mellitus by two statistical models including artificial neural network (ANN) and decision tree and also comparing these models in the diagnosis of GDM.
Methods: In this modeling study, among the cases of pregnant women who were monitored by health care centers of Kermanshah City, Iran, from 2010 to 2012, four hundred cases were selected, therefore the information in these cases was analyzed in this study. Demographic information, mother's maternal pregnancy rating, having diabetes at the beginning of pregnancy, fertility parameters and biochemical test results of mothers was collected from their records. Perceptron ANN and decision tree with CART algorithm models were fitted to the data and those performances were compared. According to the accuracy, sensitivity, specificity criteria and surface under the receiver operating characteristic (ROC) curve (AUC), the superior model was introduced.
Results: Following the fitting of an artificial neural network and decision tree models to data set, the following results were obtained. The accuracy, sensitivity, specificity and area under the ROC curve were calculated for both models. All of these values were more in the neural network model than the decision tree model. The accuracy criterion for these models was 0.83, 0.77, the sensitivity 0.62, 0.56 and specificity 0.95, 0.87, respectively. The surface under the ROC curve in ANN model was significantly higher than decision tree (0.79, 0.74, P=0.03).
Conclusion: In predicting and categorizing the presence and absence of gestational diabetes mellitus, the artificial neural network model had a higher accuracy, sensitivity, specificity, and surface under the receiver operating characteristic curve than the decision tree model. It can be concluded that the perceptron artificial neural network model has better predictions and closer to reality than the decision tree model. |
| first_indexed | 2024-12-21T00:43:11Z |
| format | Article |
| id | doaj.art-9688d046d5a0480d8946b53cd97c1eff |
| institution | Directory Open Access Journal |
| issn | 1683-1764 1735-7322 |
| language | fas |
| last_indexed | 2024-12-21T00:43:11Z |
| publishDate | 2019-09-01 |
| publisher | Tehran University of Medical Sciences |
| record_format | Article |
| series | Tehran University Medical Journal |
| spelling | doaj.art-9688d046d5a0480d8946b53cd97c1eff2022-12-21T19:21:36ZfasTehran University of Medical SciencesTehran University Medical Journal1683-17641735-73222019-09-01776359367Comparison of gestational diabetes prediction with artificial neural network and decision tree modelsMansour Rezaei0Negin Fakhri1Fateme Rajati2Soodeh Shahsavari3 Department of Biostatistics, Social Development and Health Promotion Research Center, School of Public Health, Kermanshah University of Medical Sciences, Kermanshah, Iran. Department of Biostatistics, Student’s Research Committee, School of Public Health, Kermanshah University of Medical Sciences, Kermanshah, Iran. Research Center for Environmental Determinants of Health, Health Institute, Kermanshah University of Medical Sciences, Kermanshah, Iran. Department of Health Information Technology, Faculty of Para Medicine, Kermanshah University of Medical Sciences, Kermanshah, Iran. Background: Gestational diabetes mellitus (GDM) is one of the most common metabolic disorders in pregnancy, which is associated with serious complications. In the event of early diagnosis of this disease, some of the maternal and fetal complications can be prevented. The aim of this study was to early predict gestational diabetes mellitus by two statistical models including artificial neural network (ANN) and decision tree and also comparing these models in the diagnosis of GDM. Methods: In this modeling study, among the cases of pregnant women who were monitored by health care centers of Kermanshah City, Iran, from 2010 to 2012, four hundred cases were selected, therefore the information in these cases was analyzed in this study. Demographic information, mother's maternal pregnancy rating, having diabetes at the beginning of pregnancy, fertility parameters and biochemical test results of mothers was collected from their records. Perceptron ANN and decision tree with CART algorithm models were fitted to the data and those performances were compared. According to the accuracy, sensitivity, specificity criteria and surface under the receiver operating characteristic (ROC) curve (AUC), the superior model was introduced. Results: Following the fitting of an artificial neural network and decision tree models to data set, the following results were obtained. The accuracy, sensitivity, specificity and area under the ROC curve were calculated for both models. All of these values were more in the neural network model than the decision tree model. The accuracy criterion for these models was 0.83, 0.77, the sensitivity 0.62, 0.56 and specificity 0.95, 0.87, respectively. The surface under the ROC curve in ANN model was significantly higher than decision tree (0.79, 0.74, P=0.03). Conclusion: In predicting and categorizing the presence and absence of gestational diabetes mellitus, the artificial neural network model had a higher accuracy, sensitivity, specificity, and surface under the receiver operating characteristic curve than the decision tree model. It can be concluded that the perceptron artificial neural network model has better predictions and closer to reality than the decision tree model.http://tumj.tums.ac.ir/article-1-9923-en.htmlgestational diabetes mellitusartificial neural networkdecision treeaccuracysensitivity |
| spellingShingle | Mansour Rezaei Negin Fakhri Fateme Rajati Soodeh Shahsavari Comparison of gestational diabetes prediction with artificial neural network and decision tree models Tehran University Medical Journal gestational diabetes mellitus artificial neural network decision tree accuracy sensitivity |
| title | Comparison of gestational diabetes prediction with artificial neural network and decision tree models |
| title_full | Comparison of gestational diabetes prediction with artificial neural network and decision tree models |
| title_fullStr | Comparison of gestational diabetes prediction with artificial neural network and decision tree models |
| title_full_unstemmed | Comparison of gestational diabetes prediction with artificial neural network and decision tree models |
| title_short | Comparison of gestational diabetes prediction with artificial neural network and decision tree models |
| title_sort | comparison of gestational diabetes prediction with artificial neural network and decision tree models |
| topic | gestational diabetes mellitus artificial neural network decision tree accuracy sensitivity |
| url | http://tumj.tums.ac.ir/article-1-9923-en.html |
| work_keys_str_mv | AT mansourrezaei comparisonofgestationaldiabetespredictionwithartificialneuralnetworkanddecisiontreemodels AT neginfakhri comparisonofgestationaldiabetespredictionwithartificialneuralnetworkanddecisiontreemodels AT fatemerajati comparisonofgestationaldiabetespredictionwithartificialneuralnetworkanddecisiontreemodels AT soodehshahsavari comparisonofgestationaldiabetespredictionwithartificialneuralnetworkanddecisiontreemodels |