Assessment of the risk factors for vitamin D3 deficiency in chronic hepatitis B patient using the decision tree learning algorithm in Birjand
Freshteh Osmani, Masood Ziaee
Abstract
Vitamin D deficiency as a common problem is related to the severity of chronic liver disease. In this study, we have proposed a novel model based on data mining techniques to analyze the potential risk factors related to vitamin D3 deficiency among chronic Hepatitis B (CHB) patients along with a healthy population of Birjand, an Eastern-located city in Iran. As a case-control study, their serum biochemical characteristics were measured. For constructing the decision-tree, 60% of the data were allocated to a training dataset randomly. To evaluate its performance, the remaining was used. The validation of the model was evaluated by the receiver operating characteristic (ROC) curve. The prevalence of vitaminD3 deficiency was 63.0% and 32.9% among CHB and healthy groups respectively. We concluded that the serum levels of Zn are a predictive variable for vitamin D3 deficiency in healthy subjects. The risk of vitamin D3 deficiency, due to chronic hepatitis B, could be predicted with high accuracy using a decision tree learning algorithm that could be used for antiviral therapy in CHB patients.