Detection of milk adulteration using coffee ring effect and convolutional neural network

Area of Science:

• Food Science
• Analytical Chemistry
• Biophysics

Background:

• Milk adulteration is a global concern affecting food safety and consumer trust.
• Current detection methods can be costly and time-consuming.
• A simple, cost-effective method for milk adulteration detection is needed.

Purpose of the Study:

• To develop and validate a low-cost method for identifying water and synthetic milk adulteration in cow's milk.
• To analyze coffee ring patterns formed by milk droplets.
• To classify different types of adulterants using image analysis and machine learning.

Main Methods:

• Cow's milk samples were adulterated with tap water (TW), distilled water (DW), mineral water (MW), and synthetic milk.
• Droplets were cast on glass slides to form coffee ring patterns.
• Ring area, particle area, and particle diameter were measured.
• An Alexnet model was employed for classification of adulterants.

Main Results:

• The ratio of total particle area to total ring area followed an exponential model with average particle diameter, differentiating water types.
• Distilled water and mineral water adulterated milk showed distinct clustering compared to tap water.
• Synthetic milk adulteration resulted in unique patterns like concentric rings and oil globules.
• The Alexnet model achieved high accuracy (96.7% binary, 95.8% tertiary classification).

Conclusions:

• Coffee ring patterns offer a viable, low-cost method for detecting milk adulteration.
• The method can effectively distinguish synthetic milk from pure milk.
• It can also differentiate between types of water used for adulteration.