A Model for Risk Prediction of Fracture in Diabetic Patients With Osteoporosis

Overview

The fracture risk of diabetic patients proves to be higher than those without diabetesdue to thehyperglycemia, usage of diabetes drugs, the changes in insulin levels and excretion, and this risk begins as early as adolescence.Many factors may be related to bone metabolism in patients with diabetes, including demographic data (e.g. age, height, weight, gender), medical history (e.g. smoking, drinking, menopause) and examination (e.g. bone mineral density, blood routine), urine routine).However, most of existing methods are qualitative assessments and do not take the interactions of the physiological factors of humans into consideration. In addition, the fracture risk of diabetic patients with osteoporosis has not been further studied before. In order to investigate the effect of patients' physiological factors on fracture risk, in the paper, we used a hybrid model combining XGBoost with deep neural network to predict the fracture risk of diabetic patients with osteoporosis.

Conditions

Interventions