NIT Rourkela develops AI-driven model to give improved blood sugar predictions

NEW DELHI: The National Institute of Technology (NIT) Rourkela has developed a new artificial intelligence (AI)-powered model to improve blood sugar predictions for people with diabetes. The newly found approach to predict near to accurate blood sugar levels is designed to help patients, and doctors in acquiring better medical results.

The researchers at the technology institute designed the machine-learning model to enhance the accuracy of blood glucose level prediction. It will help individuals and healthcare providers make better and personalised treatment decisions, the institute said.

Recently, the NIT has developed a model based on machine learning to evaluate ground water quality for irrigation purposes.

Mirza Khalid Baig, assistant professor, biotechnology and medical engineering, NIT Rourkela, said: “Our core innovation lies in using multi-head attention layers within a neural basis expansion network, which allows the model to focus on the most relevant data points while ignoring unnecessary noise.”

“By combining precision with efficiency, we aim to provide a practical tool that can be integrated into digital health solutions, helping patients and doctors manage diabetes more effectively,” Baig added. The researchers at the institute used a specialised AI model that learns from past blood sugar trends and predicts future levels more accurately than the existing models, the university added.

Also read IIT Delhi partners with NRLDC to develop advance solutions for restoring power after blackout

NIT Rourkela: Research findings, clinical trials

“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,” it said.

The AI-driven model also enables prediction that matches a person’s unique glucose patterns and as a result, it will provide better accuracy in sugar levels. Hence, making timely and personalised adjustments to future insulin doses, meals, and physical activity.

The model is designed in a way that it can be linked with smartphones and insulin pumps, making it more accessible for everyday diabetes management.

“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,” as per the university’s official statement.

The researchers at NIT are looking forward to test the model through clinical trials at hospitals, in partnership with experienced diabetologists in Odisha.