April 14, 2021


Born to play

Keeping It Fresh: New AI-based Strategy Can Assess the Freshness of Beef Samples

Researchers combine spectroscopy and deep finding out in an productive strategy for detecting spoiled meat

Researchers at Gwangju Institute of Science and Know-how, Korea, combine an low-cost spectroscopy strategy with synthetic intelligence to establish a new way of evaluating the freshness of beef samples. Their process is remarkably quicker and a lot more charge-successful than conventional strategies whilst protecting a comparatively superior accuracy, paving the way for mass-generated products to establish spoiled meat both in the market and at property.

Picture credit: Pixabay (Cost-free Pixabay license)

Whilst beef is one particular of the most eaten meals all-around the environment, eating it when it is previous its primary is not only unsavory, but also poses some significant overall health threats. Sad to say, offered approaches to examine for beef freshness have a variety of disadvantages that keep them from getting practical to the community. For illustration, chemical analysis or microbial population evaluations choose also much time and involve the capabilities of a experienced. On the other hand, non-damaging strategies based on around-infrared spectroscopy involve pricey and complex devices. Could synthetic intelligence be the crucial to a a lot more charge-successful way to assess the freshness of beef?

At Gwangju Institute of Science and Know-how (GIST), Korea, a workforce of experts led by Associate Processors Kyoobin Lee and Jae Gwan Kim have developed a new method that combines deep finding out with diffuse reflectance spectroscopy (DRS), a comparatively low-cost optical strategy. “Contrary to other styles of spectroscopy, DRS does not involve advanced calibration instead, it can be utilized to quantify element of the molecular composition of a sample applying just an inexpensive and easily configurable spectrometer,” points out Lee. The conclusions of their review are now posted in Foodstuff Chemistry.

To figure out the freshness of beef samples, they relied on DRS measurements to estimate the proportions of different types of myoglobin in the meat. Myoglobin and its derivatives are the proteins mainly responsible for the coloration of meat and its improvements throughout the decomposition procedure. However, manually converting DRS measurements into myoglobin concentrations to ultimately make a decision upon the freshness of a sample is not a extremely accurate strategy—and this is in which deep finding out comes into perform.

Convolutional neural networks (CNN) are widely utilized synthetic intelligence algorithms that can discover from a pre-labeled dataset, referred to as ‘training set,’ and come across concealed patterns in the data to classify new inputs. To prepare the CNN, the researchers gathered data on seventy eight beef samples throughout their spoilage procedure by consistently measuring their pH (acidity) together with their DRS profiles. Just after manually classifying the DRS data based on the pH values as ‘fresh,’ ‘normal,’ or ‘spoiled,’ they fed the algorithm the labelled DRS dataset and also fused this facts with myoglobin estimations. “By delivering both myoglobin and spectral facts, our educated deep finding out algorithm could correctly classify the freshness of beef samples in a make a difference of seconds in about 92% of scenarios,” highlights Kim.

Other than its accuracy, the strengths of this novel method lie in its pace, small charge, and non-damaging nature. The workforce thinks it may possibly be probable to establish little, portable spectroscopic products so that all people can easily assess the freshness of their beef, even at property. Also, identical spectroscopy and CNN-based approaches could also be prolonged to other products and solutions, this sort of as fish or pork. In the long term, with any luck, it will be a lot easier and a lot more available to establish and steer clear of questionable meat.


Authors: Sungho Shin (1), Youngjoo Lee (2), Sungchul Kim (2), Seungjun Choi (1), Jae Gwan Kim (2) Kyoobin Lee (1)

Title of initial paper:       Immediate and non-damaging spectroscopic process for classifying beef freshness applying a deep spectral community fused with myoglobin facts

Journal: Foodstuff Chemistry

DOI: ten.1016/j.foodchem.2021.129329


  • School of Integrated Know-how, Gwangju Institute of Science and Know-how (GIST)
  • Division of Biomedical Science & Engineering, Gwangju Institute of Science and Know-how (GIST)

About Gwangju Institute of Science and Know-how (GIST)

Gwangju Institute of Science and Know-how (GIST) is a analysis-oriented college located in Gwangju, South Korea. A single of the most prestigious universities in South Korea, it was founded in 1993. The college aims to develop a solid analysis setting to spur enhancements in science and technological know-how and to promote collaboration in between foreign and domestic analysis plans. With its motto, “A Proud Creator of Potential Science and Know-how,” the college has continuously gained one particular of the best college rankings in Korea.

Internet site: https://www.gist.ac.kr/

About the authors

Kyoobin Lee is an Associate Professor and Director of the AI laboratory at GIST. His team is creating AI-based robot vision and deep finding out-based bio-medical analysis approaches. In advance of joining GIST, he obtained a PhD in Mechatronics from KAIST and finished a postdoctoral teaching system at Korea Institute of Science and Know-how (KIST).

Jae Gwan Kim is an Associate Professor at the Division of Biomedical Science and Engineering at GIST since 2011. His current analysis topics include mind stimulation by transcranial ultrasound, anesthesia depth monitoring, and screening the phase of Alzheimer’s disorder by way of mind useful connectivity measurements. In advance of joining GIST, he finished a postdoctoral teaching system at the Beckman Laser Institute and Health-related Clinic at UC Irvine, United states of america. In 2005, he gained a PhD in Biomedical Engineering from a joint system in between the College of Texas at Arlington and the College of Texas Southwestern Health-related Middle at Dallas, United states of america.