Association study on bone metabolism in type 2 diabetes by using machine learning

Abstract

Type 2 diabetes mellitus is often accompanied by serious complications, including bone metabolic diseases, liver diseases, and kidney diseases, which are affected by the course of disease, sex, age and individual differences and cannot be a unified treatment paradigm. Therefore, for the in-depth analysis of clinical data, looking for the correlation of type 2 diabetes complication data has important guiding significance for the treatment of type 2 diabetes and its complications. In this paper, multiple linear regression models were established based on the clinical data of type 2 diabetes patients in Anhui Province. Our results suggest that the main factors affecting bone complications of type 2 diabetes include body shape indexes, creatinine, uric acid, triglycerides and blood pressure. Interestingly, the bone mineral density of lumbar vertebrae in patients with type 2 diabetes was increased, suggesting that there was a risk of lumbar hyperosteogeny.