Artificial Intelligence Applications in Type 2 Diabetes Mellitus Care: Focus on Machine Learning Methods.

Area of Science:

• Healthcare Informatics
• Medical Artificial Intelligence
• Diabetes Mellitus Research

Background:

• Type 2 diabetes mellitus incidence is rising globally.
• Artificial intelligence (AI) offers potential for improved diabetes diagnosis, treatment, and outcome prediction.
• Identifying current AI applications in type 2 diabetes mellitus care is crucial.

Purpose of the Study:

• To systematically identify and categorize artificial intelligence (AI) applications utilized in type 2 diabetes mellitus care.
• To review the landscape of AI tools and methodologies applied to diabetes management.

Main Methods:

• A systematic review was conducted in 2018, searching PubMed, Web of Science, and Embase databases.
• Articles were selected based on PRISMA guidelines, with 31 studies meeting inclusion criteria.
• Data extraction and summarization focused on AI applications in type 2 diabetes mellitus.

Main Results:

• AI applications in type 2 diabetes mellitus primarily focus on screening and diagnosis.
• Machine learning (ML) methods constitute the majority (71%) of AI techniques reviewed.
• Support vector machine (21%) and naive Bayesian (19%) algorithms are the most frequently employed ML methods.

Conclusions:

• Machine learning, especially support vector machine and naive Bayesian algorithms, shows promise for diabetes-related outcome classification.
• Further research should explore and test various AI algorithms to identify optimal approaches for diabetes care.
• AI holds significant potential to enhance the management and outcomes of type 2 diabetes mellitus.