Diabetes is one of the most common chronic diseases, affecting around 66 million adults in Europe – a figure projected to rise to more than 72 million by 2050. For people with type 1 diabetes, keeping glucose levels within a safe range requires continuous monitoring and careful insulin management. Yet even with experience and careful treatment, unexpected fluctuations can occur, potentially leading to hypoglycaemia or hyperglycaemia.
Glucose levels are influenced by far more than food and insulin. Physical activity, stress, hormonal changes, sleep quality, circadian rhythms and individual differences in metabolism can all play a role. Anticipating fluctuations could therefore help assess risks and support better-informed decisions about diabetes management.
Continuous glucose monitoring systems already record glucose levels every few minutes, generating vast amounts of data. Researchers at Kaunas University of Technology (KTU) are now exploring whether artificial intelligence (AI) can uncover patterns within these data. Their approach brings together glucose readings with information on insulin delivery, carbohydrate intake and physical activity.
AI Identifies What Matters Most
In their latest study, KTU researchers developed a model that forecasts glucose changes and assesses the risk of hypoglycaemia 30 and 60 minutes ahead. They also investigated how AI could be used to retrospectively evaluate personalised insulin adjustments.
To achieve this, the researchers combined several AI methods within a single framework. While the underlying technology may appear complex, KTU Professor Rytis Maskeliūnas says the basic idea is considerably simpler.
“Conventional forecasting models treat data as a chronological sequence, whereas the new approach represents each point in time as a node in a graph network and connects it to similar points in the past. This allows the model to learn not only from the direct sequence of events, but also from complex relationships between different physiological states,” explains Maskeliūnas.
An important component of the model is what is known as an attention mechanism. It allows the algorithm to identify which earlier data points are most relevant to the outcome.
“It is similar to an experienced doctor assessing a patient’s condition: rather than mechanically reviewing all the available data, they identify the factors that matter most in that particular situation,” explains Maskeliūnas.
This principle also addresses another important challenge surrounding the use of AI in medicine: explainability. Even a highly accurate algorithm has limited value in healthcare if clinicians cannot understand how it arrived at a particular result. The researchers’ approach therefore makes it possible to identify which features and earlier points in time had the greatest influence on the model’s output.
Model Could Support Clinical Decision-Making
The model was tested using international type 1 diabetes datasets and demonstrated high accuracy in both glucose forecasting and hypoglycaemia risk assessment. The researchers also found that performance improved when the algorithm was trained to carry out both tasks simultaneously rather than separately.
“The results showed that the new models achieve very high forecasting accuracy and perform well even with patients whose data the algorithm has not previously encountered,” says KTU PhD student Muhammad Abdullah Sarwar, who contributed to the development of the model.
The personalised insulin adjustments evaluated in the study are not intended to allow patients to alter their treatment independently. Instead, they were investigated as a potential decision-support tool for clinical analysis.
“This is particularly important when developing widely applicable clinical decision-support tools, as such systems need to perform reliably across different hospitals and countries, and with different patient populations,” emphasises Sarwar.
High accuracy, however, does not mean that such algorithms are ready to become part of patients’ everyday lives. “Further clinical studies are needed before the technology can be widely adopted in clinical practice. The models were evaluated using historical patient data, so it is essential to demonstrate that they remain equally accurate under real-world clinical conditions. Our medical partners will be responsible for this next stage,” says Maskeliūnas.
Alongside clinical testing, questions remain around data security, patient privacy and the seamless integration of algorithms into clinicians’ everyday work. The researchers see AI not as a technology that could replace doctors, but as a way to make better use of the vast amounts of data generated by patients and, in the future, provide more personalised support.
“Advanced, explainable AI models will not only allow us to predict glucose changes more accurately, but will also provide a foundation for developing safer, more personalised and more effective decision-support systems that can help reduce the heavy workload faced by clinicians,” says the KTU professor.
The article GAT-BiGRU: explainable multi-task temporal graph learning for glucose forecasting, hypoglycemia risk, and counterfactual insulin adjustment is available here .
Journal of the American Medical Informatics Association
GAT-BiGRU: explainable multi-task temporal graph learning for glucose forecasting, hypoglycemia risk, and counterfactual insulin adjustment
18-Jun-2026