Newswise – A team of scientists led by Nanyang Technological University, Singapore (NTU Singapore) has invented an artificial olfactory system that mimics the nose of mammals to accurately assess the freshness of meat.
The “electronic nose” (e-nose) comprises a “barcode” that changes color over time in reaction to the gases produced by the meat as it spoils, and a barcode “reader” in the form of a smartphone app powered by artificial intelligence (AI). The e-nose has been trained to recognize and predict meat freshness from a large library of barcode colors.
When tested on commercially packaged chicken, fish and beef samples that were allowed to age, the team found that the AI algorithm of the deep convolutional neural network that powers the electronic nose predicted the freshness of the meats with precision. 98.5%. By way of comparison, the research team assessed the prediction accuracy of an algorithm commonly used to measure the response of sensors such as the barcode used in this electronic nose. This type of analysis showed an overall accuracy of 61.7%.
The e-nose, described in an article published in the scientific journal Advanced material in October, it could help reduce food waste by confirming to consumers whether the meat is fit for consumption, more accurately than a “ Best Before ” label could, NTU Singapore’s research team said. who collaborated with scientists from Jiangnan University, Monash University, Australia.
Lead co-author Professor Chen Xiaodong, Director of the Innovative Center for Flexible Devices at NTU, said: “Our proof-of-concept artificial olfactory system, which we have tested in real-life scenarios, can be easily integrated. in packaging materials and yields results in short lead times without the cumbersome wiring used for electrical signal collection in some recently developed electronic noses.
“These barcodes help consumers save money by ensuring they don’t discard products that are still fit for consumption, which also helps the environment. The biodegradable and non-toxic nature of the barcodes also means they could be safely applied. in all parts of the food chain to ensure the freshness of food “.
A patent has been filed for this method of real-time monitoring of food freshness, and the team is now working with a Singapore agribusiness company to extend this concept to other types of perishable products.
A nose for freshness
The e-nose developed by NTU scientists and their collaborators comprises two elements: a colored “barcode” that reacts with the gases produced by decaying meat; and a barcode “reader” that uses artificial intelligence to interpret the color scheme on the barcode. To make the e-nose portable, scientists integrated it into a smartphone app that can produce results in 30 seconds.
The electronic nose mimics the functioning of a mammal’s nose. When the gases produced by decaying meat bind to the receptors in the mammalian nose, the signals are generated and transmitted to the brain. The brain then collects these responses and organizes them into patterns, allowing the mammal to identify the smell present as the meat ages and rots.
In the e-nose, the 20 bars in the barcode act as receptors. Each bar is made up of chitosan (a natural sugar) embedded in a cellulose derivative and loaded with a different type of dye. These dyes react with the gases emitted by decomposing meat and change color in response to different types and concentrations of gases, resulting in a unique combination of colors that acts as a “scent fingerprint” for the state of any meat.
For example, the first bar of the barcode contains a yellow dye that is weakly acidic. When exposed to nitrogen-containing compounds produced by decaying meat (called bioamines), this yellow dye turns blue when the dye reacts with these compounds. The intensity of the color changes as the concentration of bioamines increases as the meat decays further.
For this study, scientists first developed a classification system (fresh, less fresh or spoiled) using an international standard that determines the freshness of the meat. This is done by extracting and measuring the amount of ammonia and two other bioamines present in fish packages wrapped in widely used transparent PVC (polyvinyl chloride) packaging film and stored at 4 ° C (39 ° Fahrenheit) for five days at different intervals.
They simultaneously monitored the freshness of these fish packs with barcodes glued to the inside of the PVC film without touching the fish. The images of these barcodes were taken at different intervals over five days.
> E-nose achieves an overall accuracy of 98.5%
A type of artificial intelligence algorithm known as deep convolutional neural networks was then trained with images of different bar codes to identify patterns in the scent imprint that match each freshness category.
To assess the accuracy of their electronic nose prediction, the NTU scientists then monitored the freshness of commercially packaged chicken, fish and beef with barcodes glued to the packaging film and stored at 25 ° C (77 ° Fahrenheit). . Over 4,000 barcode images of six meat packs were taken at different time intervals for 48 hours without opening the different meat packs.
The research team first trained their system to locate patterns among scent fingerprints captured in 3,475 barcode images, before testing the system’s accuracy on the remaining images.
The results revealed an overall accuracy of 98.5%: 100% accuracy in identifying spoiled meats and 96-99% accuracy for fresh and less fresh meats.
As a comparison, the research team randomly selected 20 barcode images from each freshness category to evaluate the accuracy of the Euclidean distance analysis prediction, a method commonly used to measure the response of sensors such as the code. bar used in this electronic nose. This analysis showed an overall accuracy of 61.7%.
Professor Chen, chair of the chair of Materials Science and Engineering at NTU, said: “Although electronic noses have been the subject of extensive research, there are still bottlenecks to their commercialization due to the current prototype problems with detection. and accurate odor identification system that has both a robust sensor setup and a data analysis method that can accurately predict perfume fingerprints, which is what our electronic nose offers.
“Its non-destructive, automated, real-time monitoring capability could also be used to recognize the types of gases that other types of perishable foods emit as they become less fresh, providing a new, widely applicable platform for quality control of food. foods, which is what we are working on for now. “
About Nanyang Technological University, Singapore
A research-intensive public university, Nanyang Technological University, Singapore (NTU Singapore) has 33,000 undergraduate and postgraduate students in engineering, business, science, humanities, arts and social sciences and graduate colleges. It also has a medical school, the Lee Kong Chian School of Medicine, which was established together with Imperial College London.
NTU is also home to world-class autonomous institutes – the National Institute of Education, S Rajaratnam School of International Studies, Singapore’s Earth Observatory, and Singapore Center for Environmental Life Sciences Engineering – and various major research centers such as Nanyang Environment & Water Research Institute (NEWRI) and Energy Research Institute @ NTU (ERI @ N).
Ranked among the best universities in the world by QS, NTU has also been named the best young university in the world for the past seven years. The University’s main campus is often listed in the top 15 most beautiful university campuses in the world and has 57 Green Mark certified (LEED equivalent) building projects, of which 95% are Green Mark Platinum certified. In addition to its main campus, NTU also has a campus in the Singapore Health District.
As part of the NTU Smart Campus vision, the University harnesses the power of digital technology and technological solutions to support better learning and life experiences, the discovery of new knowledge and the sustainability of resources.
For more information, visit http: // www.