AI methods with the latest technologies, including medical devices, mobile computing, and sensor technologies, have the potential to enable the creation and delivery of better management services to deal with chronic diseases like diabetes mellitus.
One of the initial uses of AI in diabetes care involves providing personalised recommendations about therapy adjustments according to glucose levels monitored in real time with wearable continuous glucose monitoring (CGM) and has been demonstrated to have the potential to improve HbA1c levels and reduce the time spent in hypoglycaemia. A large number of studies using different AI approaches have tried to automate insulin infusion rates based on continuous glucose monitoring (CGM) data and also to suggest insulin bolus dose with special impact on prediction of critical glycaemic events such as prediction of impending hypoglycaemia or hyperglycaemia.
Prediction of diabetes based on genetic as well as clinical data, and algorithms have been used to ascertain risk of occurrence of diabetes based on electronic health record data. This can alert physicians towards the possibility of diagnosis of diabetes being missed. Diagnosing diabetes without a blood sample could become a reality thanks to a machine learning algorithm to predict whether people had type 2 diabetes, prediabetes, or no diabetes. This algorithm, with 97 per cent accuracy utilises the basis of measurements of the heart’s electrical activity, determined from an electrocardiogram (ECG).
AI plays an important role in prediction of complications, prediction of risk of retinopathy, nephropathy, neuropathy or cardiovascular event by using baseline clinical and biochemical data. Adoption of these technologies can significantly increase detection and early treatment of diabetic complications. AI contributed to great advances in the field of automated diagnosis and grading of diabetic retinopathy based on fundus photographs which is in practice now. The major limitations were a lack of data sets and severity in misclassification.
The Whole Body Digital Twin is a predictive model that provides individualised nutrition, sleep, activity, and breathing guidance to patients and their healthcare providers, with the potential to help reverse diabetes and metabolic diseases. The technology was built from thousands of data points collected daily via non-invasive wearable sensors, providing a personalised representation of each individual’s unique metabolism.
Limitations have encountered in clinical practice due to suboptimal infrastructure, missing data in patient file, training facility and cost. It needs improvement in many aspects to be functional on a clinical setting.
With the threat of data theft and breach of privacy, due diligence must be given to ethical and legal aspects to protect the patient. It is acknowledged that AI can facilitate the decision making process but not entirely replace a physician’s role. With able governing laws, systems to protect safety, minimise bias and improve transparency, AI and precision medicine could help control the burden of disease.
