Since its inception as an academic discipline in 1956, Artificial Intelligence (AI) has enabled significant breakthroughs in the fields of science, healthcare, and transport. While the term ‘AI’ may conjure up images of self-driving cars or sentient cyborgs, it’s also transforming diabetes care and the way we identify, treat, and monitor disease. Here, I provide three examples of the impact that AI is having on diabetes care across the following key areas: 1) patient enablement and support, 2) data-driven approaches to disease prediction; and 3) clinician support.
Patient enablement and support
First up is the blossoming field of AI-driven continuous glucose monitoring devices. These devices are revolutionising the management of type-1 diabetes by providing automatic and real-time data of the rate of change and concentrations of blood glucose. Specifically, these devices leverage the vast amount of patient data available to provide personalised recommendations for treatment administration, greatly minimising the risks associated with exogenous insulin administration. Going one step further, in 2021 it was announced that 1,000 patients will participate in the ‘artificial pancreas’ pilot programme that applies ‘closed loop technology’ to continually monitor blood glucose and automatically adjust the amount of insulin administered.
Data-driven approaches to disease detection
Continuing the theme of leveraging vast amounts of patient data, AI is changing the way we explore large electronic databases to predict patient outcomes and identify those at a high-risk of associated complications. For example, (1) demonstrate a machine-learning model that was trained using 700 features collected from over one million patients across multiple data sources. This model was able to predict, with high accuracy, the three-year risk of complications including cardiovascular and hyper/hypoglycaemic events in people with diabetes. Such insights have the potential to improve clinical outcomes by promoting the delivery of early and personalised care for patients with diabetes.
Lastly, I provide an example of how AI is supporting clinicians to monitor and detect one of the most common complications of diabetes: diabetic retinopathy. Trained using data gleaned from millions of retinal images, AI tools such as EyeArt and RetinaLyze can detect and grade diabetic retinopathy with high accuracy. Such systems have the potential to reduce the workload and burden placed on healthcare professionals, improve detection rates, and reduce time to referral to specialist eye services. The result is patients can access vital healthcare earlier, reducing loss of vision and highlight diseases strongly associated with diabetic retinopathy, such as stroke and cardiovascular disease.
With the accelerated adoption of AI in UK healthcare, such as the launch of NHSX that aims to improve NHS productivity with digital technology, experts have stressed the importance of tackling emerging issues such as algorithmic bias and lack of transparency in AI-driven models. As more healthcare decisions are being passed from humans to AI-driven models, it’s critical for governmental and regulatory bodies to ensure ethical and fair AI for all.
For more information of the ‘artificial pancreas’ pilot programme, see here: https://www.england.nhs.uk/2021/06/patients-with-type-1-diabetes-to-get-artificial-pancreas-on-the-nhs/
Ravaut, M., Sadeghi, H., Leung, K.K., Volkovs, M., Kornas, K., Harish, V., Watson, T., Lewis, G.F., Weisman, A., Poutanen, T. and Rosella, L., 2021. Predicting adverse outcomes due to diabetes complications with machine learning using administrative health data. NPJ digital medicine, 4(1), pp.1-12.