IIT Hyderabad developing smartphone-based sensor to detect milk adulteration
On the eve of Dr. Verghese Kurien's birthday, let's talk about that one food item he revolutionized in India: Milk. Kurien turned the country from being milk-deficient to doubling the milk-output four folds in 30 years. However, evil-minded businessmen were also quick to devise methods to poison the nutritious liquid for profit and today, milk-adulteration is one of the menaces we've to tackle.
Components: Paper, phone camera and algorithm
To help us counter the danger, a group of researchers from Indian Institute of Technology, Hyderabad, is developing a device that would help us detect the contaminants in milk. The device or smartphone-based sensors would use an indicator paper to measure the acidity in milk and after an algorithm detects the color change, the phone camera would transform the data to pH (acidity) ranges.
Purpose of study was to develop a cheap sensor
The team led by professor Shiv Govind Singh of the Department of Electrical Engineering, and Associate Professors Soumya Jana and Siva Rama Krishna Vanjari conducted the study, whose sole purpose was to develop a cheap sensor. There are 'techniques such as chromatography and spectroscopy' that can be used to check milk adulteration, but these are not feasible and also expensive, says Singh.
Idea struck Singh when he failed to find suitable testing-material
"Hence, they do not appeal to the vast majority of milk consumers in the developing world," Singh added. In fact, the idea struck Singh when he himself failed to find a suitable testing material to detect the acidity in milk supplied by local producers.
Using electrospinning, they produced paper-like material comprising nano-sized nylon fibers
So to turn a functional and affordable solution into reality, the team produced a paper-like material made of nano-sized (~10-9 m diameter) fibers of nylon using a process called 'electrospinning'. They loaded it with a combination of three dyes and made the paper halochromic, which means it would change color in response to pH. They then dipped the paper in different samples of milk.
Three machine-learning algorithms were tested out
The color-change data was captured by the phone camera, helped by a prototype smartphone-based algorithm, which was then transformed into pH (acidity) ranges. The researchers claim they have used three machine-learning algorithms to test out their detection efficiencies in classifying the color. After testing various combinations and putting in a lot of effort, the team finally found their near-perfect classification with accuracy of 99.71%.
Work on making device publicly available is underway: Singh
"We need to develop simple devices that the consumer can use to detect milk contamination. It should be possible to make milk adulteration detection failsafe by monitoring all of these parameters at the same time," Singh said, adding work on making the device available to public is underway. The study was published in the November issue of 'Food Analytical Methods' journal.
