Quantifying fat in animal tissue

Scientific contact: Guylaine Collewet

A major field of research in the creation of a modern and sustainable agricultural industry is understanding the relationship between genotype and phenotype. In the context of phenotyping, the MRIFood team has developed non-destructive characterization methods using MRI that can be carried out at high speed. A “whole body” MRI scan involves a 60cm diameter tunnel, making it possible to perform high-speed measurements by simultaneously measuring several samples. Moreover, samples do not need to be prepared before undergoing the MRI scans, which saves significant time.

We have designed a high-speed method of quantifying lipids in this context. It is based on the ability of MRI to separate the signal given by water from that given by lipids. This method has been applied to fish. We have demonstrated that it is possible to quantify lipids in less than a minute for each steak-sized sample.

The high-speed quantification of lipids can also be applied to other food matrices (meat products, some vegetables, dairy products).

The level of fat in pork adipose tissue can also be an indicator of quality in pork meat. The consistency of adipose tissue depends primarily on the proportion of lipids in a crystalline or solid state. Our research, conducted in partnership with INRA in Nantes, has demonstrated that the level of solids measured by NMR spectrometry is closely linked to the chemical composition of the lipids.
The measurement only takes a few seconds and can be performed directly on adipose tissue. Over time, this could become the most practical and relevant way of classifying this tissue. The level of solids can vary significantly from one carcass to another and according to the anatomical location of tissue on a single carcass. A study carried out in partnership with UNIPORC-OUEST and funded by OFIVAL will use magnetic resonance imaging to define the best measurement site to represent the overall quality of each carcass.

As part of a study carried out by SYSAAF and funded by OFIVAL, we have shown that MRI can be used to predict the meat yield for chickens. This reference technique could potentially replace the tedious dissections currently needed to calibrate the body composition estimation devices used, for example, to define lean meat content.

MRI image of pork loin © Irstea