New Delhi: The National Institute of Technology (NIT) Rourkela has developed an advanced AI-driven model to improve blood sugar predictions for people with diabetes. The innovation aims to assist individuals and healthcare providers in making better, more personalized treatment decisions.

How the AI Model Works

Unlike traditional forecasting methods that require manual adjustments and struggle with long-term glucose trends, this AI model automatically processes data to identify key patterns. By learning from past blood sugar fluctuations, it provides precise predictions, enabling timely adjustments in insulin doses, diet, and physical activity.

Outperforming Existing Techniques

The model has demonstrated superior accuracy compared to existing techniques, offering reliable predictions tailored to an individual’s unique glucose patterns. Additionally, it has been optimized for use on smartphones and insulin pumps, making it more accessible for everyday diabetes management.

Innovative Technology Behind the Model

According to Professor Mirza Khalid Baig, assistant professor of biotechnology and medical engineering at NIT Rourkela, the model leverages multi-head attention layers within a neural basis expansion network. This approach allows the AI to focus on the most relevant data points while ignoring unnecessary noise, enhancing performance without requiring large datasets or extensive computing power.

Potential Applications in Healthcare

The AI model can be integrated into:

1. Smart insulin pumps for automated insulin delivery.

2. Mobile health apps for real-time glucose tracking.

3. Clinical settings to assist doctors in creating personalized treatment plans.

Clinical Trials and Government Support

Currently, the research team is planning extensive clinical trials in hospitals, collaborating with senior diabetologists in Odisha. The project has received support from the Department of Science and Technology (DST), the Department of Biotechnology (DBT), and NIT Rourkela.

 

(with IANS inputs)