Artificial intelligence in diabetology
Open Access
- 14 July 2021
- journal article
- review article
- Published by Endocrinology Research Centre in Diabetes mellitus
- Vol. 24 (2) , 156-166
Abstract
This review presents the applications of artificial intelligence for the study of the mechanisms of diabetes development and generation of new technologies of its prevention, monitoring and treatment. In recent years, a huge amount of molecular data has been accumulated, revealing the pathogenic mechanisms of diabetes and its complications. Data mining and text mining open up new possibilities for processing this information. Analysis of gene networks makes it possible to identify molecular interactions that are important for the development of diabetes and its complications, as well as to identify new targeted molecules. Based on the big data analysis and machine learning, new platforms have been created for prediction and screening of diabetes, diabetic retinopathy, chronic kidney disease, and cardiovascular disease. Machine learning algorithms are applied for personalized prediction of glucose trends, in the closed-loop insulin delivery systems and decision support systems for lifestyle modification and diabetes treatment. The use of artificial intelligence for the analysis of large databases, registers, and real-world evidence studies seems to be promising. The introduction of artificial intelligence systems is in line with global trends in modern medicine, including the transition to digital and distant technologies, personification of treatment, high-precision forecasting and patient-centered care. There is an urgent need for further research in this field, with an assessment of the clinical effectiveness and economic feasibility.Keywords
This publication has 63 references indexed in Scilit:
- Building and analysis of protein-protein interactions related to diabetes mellitus using support vector machine, biomedical text mining and network analysisComputational Biology and Chemistry, 2016
- A mobile diabetes management and educational system for type-2 diabetics in Saudi Arabia (SAED)mHealth, 2016
- Mathematical Modeling of the Blood Glucose Regulation System in Diabetes Mellitus PatientsAnnals of the Russian academy of medical sciences, 2015
- Use of an holistic approach for effective adoption of User-Centred-Design techniques in diabetes disease management: Experiences in user need elicitation2015
- Deep learningNature, 2015
- The Hemoglobin Glycation Index Identifies Subpopulations With Harms or Benefits From Intensive Treatment in the ACCORD TrialDiabetes Care, 2015
- ANDSystem: an Associative Network Discovery System for automated literature mining in the field of biologyBMC Systems Biology, 2015
- Predicting Disease-Related Subnetworks for Type 1 Diabetes Using a New Network Activity ScoreOMICS: A Journal of Integrative Biology, 2012
- Disease gene identification by random walk on multigraphs merging heterogeneous genomic and phenotype dataBMC Genomics, 2012
- Prioritizing candidate disease genes by network-based boosting of genome-wide association dataGenome Research, 2011