Main Article Content
This research aimed to develop the model to predict vision threatening diabetic retinopathy (VTDR) condition in diabetic patients so that preventive medical treatments could be started at early stage for prolonging patients’ good quality of life. Two popular predicting techniques for categorical dependent variable were used in comparison. These were logistic regression and probabilistic neural network. Data for the last six months of DR patients at Nakhonpathom hospital were collected based on 18 variables from the literature. Data cleansing were implemented. Eventually, there was a total of 355 data used with 124 patients without VTDR condition and 231 patients with VTDR condition. These data were also separated into two groups. The first group (in sample test) with 90% of the total data was used for developing the models. The second group (out of sample test) with 10% of data was used for checking the models’ accuracy. The research results showed the overall accuracy of logistic regression model was 69.01% while the overall accuracy of probabilistic neural network model was 96.90%. Both logistic regression and probabilistic neural network models were more accurate in predicting patients with VTDR condition than without VTDR condition. The study showed significant risk factors related to VTDR condition were sex, age, cholesterol, hematocrit, creatinine, HDL-cholesterol, and body mass index.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Wanyue Li, Yanan Song, Kang Chen, Jun Ying, Zhong Zheng, Shen Qiao,et al.Predictive model and risk analysis for diabetic retinopathy using machine learning: A retrospective cohort study in China. BMJ Open 2021; 11: e050989.
Ashish Bora, Siva Balasubramanian, Boris Babenko, Sunny Virmani, Subhashini Venugopalan, Akinori Mitani,et al. Predicting the risk of developing diabetic retinopathy using deep learning. The Lancet Digit Health 2021; 3: e10-19.
Hair et al (2006). Multivariate Data Analysis. Sixth Edition. Prentice Hall.
Specht, D. F. Probabilistic neural networks. Neural Networks. 1990; 3: 109-18
Berno E, Brambilla L, Canaparo R, Casale F, Costa M, Della Pepa C, et al. Application of probabilistic neural networks to population pharmacokineties. Proceedings of the International Joint Conference on Neural Networks 2003: 2637-42.
Fernandez-Gamez M.A, Cisneros-Ruiz A.J., Callejon-Gil A. Apply a Probabilistic Neural Network to Hotel Bankruptcy Prediction. Tourism & Management Studies. 2016; 12(1): 40-52.
Huang Chenn-Jung, Liao Wei-Chen. Application of Probabilistic Neural Networks to the Class Prediction of Leukemia and Embryonal Tumor of Central Nervous System. Neural Processing Letters. 2004; 19(3): 211-26.
17. Li, Q. B., Li, X., Zhang, G. J., Xu, Y. Z., Wu, J. G., Sun, X. J. Application of Probabilistic Neural Networks Method to Gastric Endoscope Samples Diagnosis Based on FTIR Spectroscopy. Spectroscopy and Spectral Analysis 2009; 29: 1553-7