• CSIRO says its new computer model could be used to reduce the time and effort required to reformulate healthier foods.
    CSIRO says its new computer model could be used to reduce the time and effort required to reformulate healthier foods.
Close×

Consumer demand for healthier foods – that is, foods with reduced salt, sugar and fat – is a challenging requirement to meet without compromising the sensory enjoyment of food.

It is well known that if the flavour is changed to produce a low fat, salt and sugar variant, then the product will most likely not sell well, if at all.

Food researchers and industry need a better understanding of how to reduce unhealthy components in food products without compromising flavour, and also to help incorporate nutrients and fibre from fruit and vegetables.

This requires a more detailed understanding of food structure effects on the breakdown of food during chewing, the release of flavours from the food and the sensory experience of consumption.

The science of chewing

To understand how a change to a food formulation will affect the perceived flavour, it helps to know exactly how the food is broken down, where the food fragments move to in the mouth and how flavour compounds (including taste and aroma components delivered to the taste buds and the nose respectively) are released from the food.

It is practically impossible to measure the positions of food fragments and flavour chemicals as they move around the mouth during chewing, but computer simulation using a “virtual mouth” is providing exactly these details.

Using a technique called smooth particle hydrodynamics, computer modellers and food scientists at CSIRO have developed the world’s first 3D dynamic virtual mouth that predicts how a food breaks down and how flavour compounds are released, given only the initial food structure as a model input.

The model has been described in two recent publications: European Food Research and Technology and the Journal of Texture Studies. This model includes 3D representations of the anatomical features of the inside of the mouth – namely the teeth, gums, tongue, cheeks and palates and how they move during chewing.

It also shows the food structure and how it breaks apart when cut and crushed by the teeth; the saliva flow within the mouth; and the release and transport of tastes such as salt or sugar from the food into the saliva and then to the taste buds.

We can measure the 3D positions and sizes of all food fragments, saliva volumes and taste concentrations at any time during the chewing cycle. We will directly relate these outputs to sensory measures such as saltiness, creaminess and graininess as we develop the model further.

Reformulation and new ideas

We anticipate that our model can be used to reduce the time and effort required to reformulate healthier foods. Many formulation variants can be added to the chewing model and flavour experiences predicted without needing to manufacture the product or perform sensory panel experiments.

Targeted changes to food structure can be made in these virtual experiments and discarded or shortlisted without needing to prototype new formulation methods.

However, we can also progress fundamental scientific understanding of how reduction of salt, sugar or fat affects the complete chewing process using the model. This new understanding is much more informative for the design process than would be possible without the computer model.

As well as assisting the design of healthier food, the model could be used to create new food designs and flavour experiences. The virtual food masticated by the model mouth can be any shape and with no restrictions on locations of flavour compounds and structure.

Without the time required to prototype food products in the laboratory or test kitchen, the virtual food design can be assessed and modified by small or large degrees towards exciting new consumer eating sensations.

Although we expect the food manufacturing industry will be where the major opportunities lie for this technology, dental and therapeutic professions may all be able to benefit from it in the future.

If you are interested in discussing this technology, please contact CSIRO or drop into our booth at foodpro 2014, where visitors will be able to interact with our models.

Computer model: chewing caramel-filled chocolate



The images show stills from a computer simulation of the mouth chewing caramel-filled chocolate. In this example, as the teeth crush the food, the chocolate fractures at the corner and releases the caramel. You can watch a short video simulation here.

The chocolate coating collapses further and the tongue moves to reposition the food between the teeth for the next chewing cycle. The caramel then pours out of the chocolate into the vestibule area.

We can calculate the transport of flavour compounds as the food elements make contact with saliva (not shown).

Variations to thickness of chocolate, chocolate texture (stiffness and brittleness), caramel viscosity and sugar, salt and fat concentrations and locations can all be modified simply to test the effects on how the food breaks down and is distributed through the mouth and how flavour is released.

About the author  
Dr Simon Harrison is a biomechanical engineer and computer modeller at CSIRO computational informatics in Clayton, Victoria. He can be contacted on 03 9545 8450 or simon.harrison@csiro.au.

Packaging News

Earlier this month PKN published an article that voiced industry concern over the speculation that Qenos might be closing its polymer manufacturing plants in Australia. Qenos has remained silent on the matter. The MD of Qenos customer Impact International, Aleks Lajovic, wants some answers.

Amcor, a global leader in developing and producing responsible packaging solutions, is the proud recipient of eight FPA Awards for innovative and sustainable contributions to the industry.

An impressive number of quality entries into the inaugural Hive Awards created a highly competitive line-up for judges to consider. The packaging category was one of the most strongly contested.