NIT Rourkela Develops AI-Driven Model for Improved Blood Sugar Predictions

Unlike conventional forecasting models, which often require manual adjustments and struggle with long-term trends, this new model identifies critical patterns automatically, as stated by the team of researchers.

Researchers at the National Institute of Technology (NIT) Rourkela have developed a new AI-driven approach to enhance blood sugar predictions for individuals with diabetes. 

This machine-learning model improves the accuracy of blood glucose level forecasting, enabling better personalized treatment decisions for patients and healthcare providers.

The team, led by Assistant Professor Prof. Mirza Khalid Baig, utilized deep learning techniques to refine glucose forecasting. Their specialized AI model analyzes past blood sugar trends, offering more precise predictions than traditional methods. 

“According to the results of ICMR INDIAB study released in 2023, the overall prevalence of diabetes in our country is 11.4% and that of prediabetes is 15.3%. Hence, it is crucial that we develop new solutions to tackle this problem. 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. This results in better performance without the need for large amounts of training data or extensive computing power. 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," said Prof Mirza Khalid Baig, Assistant Professor, Biotechnology and Medical Engineering, NIT Rourkela.

Unlike conventional forecasting models, which often require manual adjustments and struggle with long-term trends, this new model identifies critical patterns automatically, as stated by the team of researchers.

Further, the NIT Rourkela researchers claim that this AI model outperforms existing techniques, providing reliable blood sugar predictions tailored to individual glucose patterns.

Additionally, the model is optimized for use on devices such as smartphones and insulin pumps, increasing accessibility for daily diabetes management. 

The researchers further believe this AI approach could revolutionize diabetes care, with applications in smart insulin pumps for automated delivery, mobile health apps for real-time glucose tracking, and clinical settings to aid doctors in crafting personalized treatment plans.

The findings of this study have been published in the IEEE Journal of Biomedical and Health Informatics. 

Moving forward, the research team plans to conduct extensive clinical trials in collaboration with leading diabetologists in Odisha, including Dr Jayanta Kumar Panda and his team.

Stay tuned for more such updates on Digital Health News.