Diabetes is a chronic metabolic condition. According to International Diabetes Foundation (IDF), 463 million people between the ages of 20 and 79 have diabetes, and 374 million have impaired glucose tolerance. 8.8% of the world population was reported to have diabetes in 2017, and the numbers are projected to rise to 10% by 2045.  

AI finds widespread use in four key areas in diabetes care. They are: 

• Automated retinal screening 
• Clinical decision support 
• Predictive population risk stratification 
• Patient self-management tools 

AI can influence three main domains of diabetes care: patients with diabetes, health care professionals and health care systems. AI has added newer dimensions of self-care for patients with diabetes, introduced rapid and reliable decision-making and flexible follow-ups for healthcare providers and optimized resource utilization in healthcare systems. 

AI-based techniques 

Several Ai-based techniques have been applied in diabetes care. With the advent of AI, the diagnosis of diabetes has evolved beyond a few measurements of blood glucose level and glycosylated hemoglobin. 

Case-based reason (CBR) is a technique used to solve new problems based on learning from similar past encounters. The 4 Diabetes Support System is an example of CBR used in diabetes care. The system aims to automatically detect problems in controlling blood glucose, propose solutions to the detected problems and remember the effective and ineffective solutions for individual patients. CBR is a technique used to individualize insulin therapy for various meal situations in diabetes. 

Numerous machine learning processes have been used to build digital support in diabetes care. These include support vector machine, ANN, naïve Bayes, decision tree, random forest, classification and regression trees and k-nearest neighbor. In addition, machine learning has been applied to create auto-mated screening for blood glucose variability. ML programs can also identify people with diabetes based on genetic and metabolic factors. 

ANN has been created to link and analyze disparate information and build personalized solutions. Neural network methodology has found particular and vast applications in diabetes diagnosis. Intelligent algorithms have been constructed to study the impact of various factors on glycemic indices.  

AI applications  

Deep learning algorithms have been developed to automate the diagnosis of diabetic retinopathy. AI-based screening of the retina is a feasible, accurate and well-accepted method for the detection and monitoring of diabetic retinopathy. High sensitivity and specificity of 92.3% and 93.7%, respectively, have been reported for automated screening of the retina. 

Machine learning-based clinical decision support tools have been developed to predict short and long-term HbA1c response after insulin initiation in patients with type 2 diabetes mellitus. These tools also help to identify clinical variables that can influence a patient's HbA1c response.  

Machine learning has been used to develop an intuitive approach for customizing interventions in medication adherence and predicting the risk of all-cause hospitalization. 

Advanced molecular phenotyping, genomics, epigenetic alterations, and developments of digital biomarkers are new advances in the diagnosis and management of disease cantons. These can be applied to diabetes, where huge data sets are generated owing to the disease's heterogenous nature and chronic course. 

Self-management is the key to the treatment of diabetes. With the advent of AI, patients are empowered to manage their diabetes, generate data for their parameters and be experts for health.