A machine learning model for quantifying the effect of lifestyle interventions for patients with type 2 diabetes mellitus. Issue 1 (January 2021)

• Record Type: Journal Article
• Authors: Bi, Suhuan; Ding, Xiangqian; Yu, Shusong; Guo, Baoqi; Mu, Liangliang; Wang, Bin
• Is Part Of: Journal of physics. Volume 1732:Issue 1(2021)
• Publication Date: 2021-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1732/1/012006
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Type 2 diabetes is the most common type of diabetes. The cornerstone of type 2 diabetes treatment is healthy lifestyle. This paper proposes a machine learning model for quantifying the effect of lifestyle interventions for patients. In the proposed incremental intervention model, the original physical indicators and the lifestyle interventions were taken as input vectors separately and transformed through different nonlinear functions. We evaluated our method with the dataset of 12, 318 patients from a national funding project and compared with MLP and SVR. The experimental results ( R 2 =0.85, RMSE = 0.51, MAE=0.35) indicated that the model outperformed those prediction models. Besides, the machine learning based method is cost-effective and time-saving. The proposed method provides new insights into prevention and treatment of chronic diseases.