NIT Rourkela develops AI-driven model for improved diabetes management

new delhi: researchers from the national institute of technology (nit), rourkela, have developed a new ai-driven approach to enhance blood sugar level predictions, aiming to improve diabetes management for patients and healthcare providers. the machine-learning model, designed to provide more accurate blood glucose level prediction, could assist in personalised treatment decisions. the findings of the research have been published in the ieee journal of biomedical and health informatics, with the team now panning clinical trial at hospitals. according to mirza khalid baig, assistant professor, biotechnology and medical engineering, diabetes is a major health concern in india, with cases projected to reach 124.9 million by 2045. "effective diabetes management relies on regular glucose monitoring to prevent dangerous spikes (hyperglycemia) and drops (hypoglycemia) in blood sugar levels," he said. biag said that challenges such as lack of specialists, unequal access to healthcare, low medication adherence, and poor self-care, make diabetes management difficult. "these challenges make it harder for patients to keep their blood sugar levels under control, increasing the risk of serious health problems. new digital health technologies, especially those that use artificial intelligence (ai), offer a way to improve diabetes care and reduce costs," he said. research scholar and co-author deepjyoti kalita said while machine learning (ml) has been used in many areas of diabetes research, including basic studies and predictive tools, ai learning models, especially predictive ai models, have a few drawbacks. many of these models work like a black box, meaning their predictions are difficult to interpret, which makes it hard for doctors and patients to fully trust them, he said. "furthermore, traditional models, such as statistical forecasting methods or basic neural networks, often fail to recognise long-term glucose fluctuations and require complex fine-tuning," kalita said. the researchers at nit rourkela worked on improving glucose forecasting using deep learning techniques. their approach incorporated a specialised ai model that learns from past blood sugar trends to predict future levels more accurately than existing methods. unlike traditional forecasting models, which often struggle with long-term trends and require manual adjustments, this model processes glucose data automatically, identifying key patterns and making precise predictions. "our core innovation lies in using multi-head attention layers within a neural basis expansion network, allowing the model to focus on the most relevant data points while ignoring unnecessary noise," baig said. "this results in better performance without the need for large amounts of training data or extensive computing power," he added. he said that by combining precision with efficiency, they aim to provide a practical tool that can be integrated into digital health solutions, helping patients and doctors manage diabetes more effectively. the researchers claimed that in the long run, this ai-driven approach has the potential to enhance diabetes care through various applications. "it could be integrated into smart insulin pumps to automate insulin delivery, incorporated into mobile health apps for real-time glucose tracking, or used in clinical settings to support doctors in making personalised treatment plans," he said.