Type 2 Diabetes with Artificial Intelligence Machine Learning: Methods and Evaluation

• DOI: 10.1007/s11831-021-09582-x
• Publication Date: 2021-04-15T20:07:15Z
• Authors: Leila Ismail; Huned Materwala; Maryam Tayefi; Phuong Ngo; Achim P. Karduck
• Keywords: 03 medical and health sciences; 0302 clinical medicine; 3. Good health

Abstract Diabetes, one of the top 10 causes death worldwide, is associated with interaction between lifestyle, psychosocial, medical conditions, demographic, and genetic risk factors. Predicting type 2 diabetes important for providing prognosis or diagnosis support to allied health professionals, aiding in development an efficient effective prevention plan. Several works proposed machine-learning algorithms predict diabetes. However, each work uses different datasets evaluation metrics algorithms’ evaluation, making it difficult compare among them. In this paper, we provide a taxonomy factors evaluate 35 machine learning (with without features selection) prediction using unified setup, achieve objective comparison. We use 3 real-life 9 feature selection evaluation. accuracy, F-measure, execution time model building validation under study on diabetic non-diabetic individuals. The performance analysis models elaborated article.