IIIT Bangalore student develops easy-to-use-device to classify liquids with sensors

The team's study consisted of five liquids such as vinegar, tea, coffee, a lemon-based drink and an orange-based drink. The model achieved an accuracy rate of 99.70 per cent
"We are looking at creating an app that will convert the recorded data points on how many calories the liquid item has," Yash Gupta said
Yash Gupta is a third-year student in Electronics Communication Engineering (CSE), Integrated Master's in Technology (IMTECH)(Pic: EdexLive Desk)

While cases of malnutrition and obesity are on the rise especially in the younger individuals, students at the International Institute of Information and Technology (IIIT) Bangalore are working on an easy-to-use device that can aid in addressing such conditions with continuous monitoring of food intake.

They developed a novel electronic-nose (e-nose) system, encasing four gas sensor channels which are trained with machine-learning (ML) model, consisting of temporal responses from the four sensory channels to detect the different liquids. These sensors can detect gases like Carbon monoxide (CO), Nitrogen dioxide (NO2), Ethyl alcohol (C2H5OH), and Volatile Organic Compounds (VOC).

Explaining the practical examples of e-nose, Madhav Rao, Associate professor, IIIT Bangalore said, "Post classification of the liquids, the students, in the next step are trying to build a prototype where one can identify the exact concentration of the liquid and distinguish the qualities. The idea was to build a camera-less system to not violate privacy fundamentals to identify liquids under the IoT (Internet of Things) for food products."

"Self-reporting of the food intake on a daily basis is viewed as a tedious process, leading to under-reporting. Which affects the overall evaluation of food intake. We wanted to develop a separate portable platform which is not a wearable solution, and neither a camera-based device," said Yash Gupta, a third-year student in Electronics Communication Engineering (CSE), Integrated Master's in Technology (IMTECH).

The student added that currently, most of the devices in the market cater to solid food items, and not much is available on consumable liquid items. This device is a compact glass container with a gas sensor interfaced with the micro-controller for signal acquisition.

Chemical sensors are immersed inside the liquid to appropriately detect liquids. The team's study consisted of five liquids such as vinegar, tea, coffee, a lemon-based drink and an orange-based drink. The model achieved an accuracy rate of 99.70 per cent.

Gupta also explained that the device can be made cost-effective, and can widely be used by individuals not just to understand the different liquids, but also their quantity.

"We are looking at creating an app that will convert the recorded data points on how many calories the liquid item has," he said.

Related Stories

No stories found.
X
logo
EdexLive
www.edexlive.com