IIIT-Hyderbad’s AI tool to detect malnutrition

The new smartphone-based tool developed by researchers of Raj Reddy Centre for Technology and Society (RCTS) at IIIT-Hyderabad promises to revolutionise the way malnutrition is detected among children.

Hyderabad: Researchers at the International Institute of Information Technology (IIIT) – Hyderabad have come up with a new AI-based tool that promises to revolutionise the way malnutrition is detected among children.

The new smartphone-based tool developed by researchers of Raj Reddy Centre for Technology and Society (RCTS) at IIIT-Hyderabad, represents a significant departure from the traditional practice of manual recording of measurement scales, which are prone to human errors.

---

The researchers have come out with two setups where the measurement can be taken with the web app using the Smartphone. In the first, by placing the smartphone over the child’s head two images are taken i.e., one of weighing scale reading and the other of the region directly in front of the scale (on the floor).

In the second method, with the child standing against a wall with a height chart, a photograph is captured of the child as well as of the weighing scale. Images thus generated are analysed to derive the height and weight measurements, which are in turn compared with standard growth charts to generate malnutrition estimates.

The web app comes with pictograms, guiding on how to position the smartphone and the child for taking photographs. The app also displays appropriate warnings based on malnutrition and such information is sent to parents and those concerned.

The tool has been developed by the RCTS researchers after they were approached by i-Saksham, a Bihar-based NGO that turned to AI to drive its various initiatives focused on women empowerment and Anganwadi centres. The current UI for the web app is being planned in both English and Hindi.

“We are also exploring the feasibility of using 3D anthropometry, i.e, 3D reconstruction of subjects using 2D images,” said Dr. Arjun Rajasekar, an applied researcher at RCTS. This is part of a larger effort by the RCTS to create an AI solution for reducing morbidity and mortality in newborns via automatic anthropometry measurements, he said.