Volume 7, Issue 3 (12-2020)
jhbmi 2020, 7(3): 304-317 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Abedini A, Jabarpour E, Keshtkar A. Predicting the Risk of Osteoporosis Using Decision Tree and Neural Network. jhbmi 2020; 7 (3) :304-317
URL: http://jhbmi.ir/article-1-453-en.html
URL: http://jhbmi.ir/article-1-453-en.html
Ph.D. Candidate in Industrial Management, Faculty of Management, University of Tehran, Tehran, Iran
Abstract: (2919 Views)
Introduction: Osteoporosis is one of the major causes of disability and death in elderly people. The objective of this study was to determine the factors affecting the incidence of osteoporosis and provide a predictive model to accelerate diagnosis and reduce costs.
Method: In this fundamental descriptive study, a new model was proposed to identify the factors affecting osteoporosis. Data related to 4083 women were investigated with Clementine12, the data mining tool, to discover knowledge. Using data mining algorithms, including decision tree and artificial neural network, some rules were extracted that can be used as a model to predict the condition of patients and finally, the accuracy of the proposed models were compared.
Results: This study examined several models on a number of different characteristics and compared the results in terms of accuracy to find the best predictive model. The classification accuracy of the MLP neural network model was 92.14% which was higher than that of the other algorithms used in this study. According to the identification of factors affecting osteoporosis, the risk of developing this disease can be predicted for a new sample.
Conclusion: Healthcare organizations are always gathering a lot of information while this data is not used properly. This study showed that the hidden patterns and relationships in this data can be discovered and used to improve the quality of diagnostic and treatment services.
Method: In this fundamental descriptive study, a new model was proposed to identify the factors affecting osteoporosis. Data related to 4083 women were investigated with Clementine12, the data mining tool, to discover knowledge. Using data mining algorithms, including decision tree and artificial neural network, some rules were extracted that can be used as a model to predict the condition of patients and finally, the accuracy of the proposed models were compared.
Results: This study examined several models on a number of different characteristics and compared the results in terms of accuracy to find the best predictive model. The classification accuracy of the MLP neural network model was 92.14% which was higher than that of the other algorithms used in this study. According to the identification of factors affecting osteoporosis, the risk of developing this disease can be predicted for a new sample.
Conclusion: Healthcare organizations are always gathering a lot of information while this data is not used properly. This study showed that the hidden patterns and relationships in this data can be discovered and used to improve the quality of diagnostic and treatment services.
Audio File [M4A 980 KB] (101 Download)
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
