In this study, researchers from Universitat de Girona in Girona, Spain, reviewed research in AI techniques to assist in the management of diabetes.

“Over the last decade, the entire paradigm of diabetes management has been transformed due to the integration of new technologies such as continuous glucose monitoring devices and the development of the artificial pancreas, along with the exploitation of data acquired by applying these novel tools,” wrote first author Ivan Contreas, PhD, and colleagues. “By means of complex and refined methods, AI has been shown to provide useful management tools to deal with these incremental repositories of data. Thus, AI has played a key role in the recognition of these systems as routine therapeutic aids for patients with diabetes.”

The study analyzed 1,849 published articles in PubMed from 2018 to 2018—141 of which were selected for review. Researchers found evidence to support the development of research in producing AI-powered tools for the prediction and prevention of diabetic complications. Additionally, researchers noted the benefit in using AI to improve diabetes patients’ quality of life.

Other findings included:

• Prediction and prevention are being revitalized by AI, while “safety and failure detection” have only been reviewed in 6 percent of studies.
• Investigations of the application of AI to early detection of critical issues like exercise, meals and infusion set failures is lacking.
• AI could improve the safety of both AP systems and open-loop tools has the potential to dramatically improve performance.
• 22 percent of studies focused on closed-loop systems and has become the most productive area for AI applications.
• Researchers should take advantage of the latest improvements in AI to combine it with the development of an artificial pancreas.
• Forty-one studies examined blood glucose, 27 of which explored the development of models to predict blood glucose and 14 on the detection of possible blood glucose events.
• Several studies reported accurate prediction and detection tools that had potential as management resources for current and future therapies.

“Our findings show the increasing importance of AI methods for diabetes management,” concluded Contreas and colleagues. “We think these methods will encourage further research into the use of AI methods to extract knowledge from diabetic data. “In general, the most striking advances in the application of AI techniques come from data-driven methods that learn from large datasets. The ability to collect information from individual diabetic patients has led to a shift in diabetes management systems; accordingly, systems that lack access to valuable data will face substantial hurdles. Management protocols provided to diabetic patients should be tailored to address their needs at various points during their illness.”