Does the organic egg really originate from an organically raised laying hen? This question can be answered using nuclear magnetic resonance spectroscopy (NMR spectroscopy). For this, egg yolk samples are analysed, whose spectra provide a lot of information like a fingerprint. A comparison with characteristic patterns per farming method from a database with reference spectra provides information about the actual farming method - and all this with a measurement accuracy of almost 100 percent.

A team of researchers at the DIL Deutsches Institut für Lebensmitteltechnik e. V. in Quakenbrück has developed an innovative method for this purpose: Using NMR spectroscopy, it will be possible to clearly recognise whether an egg really comes from organically reared laying hens with just one measurement. The researchers analysed around 4,500 egg samples for this purpose. The Federal Ministry of Agriculture funded the project through the Federal Organic Farming Programme (Bundesprogramm Ökologischer Landbau, BÖL).

Reference database established with AI

To verify the farming method, the researchers analysed egg yolk extract samples obtained from the eggs using 1H-NMR spectroscopy. The spectra generated in this way represent a highly specific pattern of the egg sample, a kind of fingerprint with a high information value. Using multivariate data analyses, machine learning, and AI, the researchers identified characteristic patterns for each husbandry method and created a database and an authenticity model from these reference spectra. By comparing the 1H NMR spectrum of unknown egg yolk samples with the model from reference spectra, they were able to prove the actual husbandry form.


Almost 100 percent measurement accuracy - also possible to determine breed

The calculated model for the classification of eggs from conventional and organic farming achieves an accuracy of 99.9 percent, while the eggs analysed can be assigned to the four types of farming with an accuracy of 97.1 percent. In addition, the researchers were able to determine the breed of laying hen (Lohmann Selected Leghorn, Dekalb, Lohmann Brown, Sandys) with a model accuracy of 98.4 percent.

Detection method for trade and processing

According to the researchers, the results offer great potential in food control, for example in cases of suspicion or in random sample analyses for the retail and processing sectors. Companies can commission corresponding analyses and thus reliably check whether the stamp code on the eggshell correctly indicates the farming method of the laying hens. This strengthens confidence in the authenticity of organic eggs and thus consumer demand for organic eggs.

Partner wanted for continuation

In order to establish the method, the models must be supplemented with further authentic samples and remain up to date. The models will only remain meaningful if additional influencing factors, some of which are not yet known, such as other breeds and feedstuffs, are considered. The DIL invites companies, research institutions, and associations to collaborate to further develop this pioneering analysis and its potential applications.