WO2014032121A1

WO2014032121A1 - Method and system for classifying a foodstuff

Info

Publication number: WO2014032121A1
Application number: PCT/AU2013/000993
Authority: WO
Inventors: Cameron Jay SCADDING; Rachel Louise GREEN; Christopher David MAY; Roger John Watling
Original assignee: Australian Egg Corporation Limited
Priority date: 2012-09-03
Filing date: 2013-09-03
Publication date: 2014-03-06
Also published as: AU2013204502A1; AU2013204502B2; AU2013308335A1

Abstract

The present invention relates to a method and system for classifying a commodity such as a foodstuff. The method involves analysing a variety of components of a multiple component foodstuff to determine constituents of the components. The values of the constituents are compared with known values that correlate with defined origins, to determine the origin of the foodstuff.

Description

METHOD AND SYSTEM FOR CLASSIFYING A FOODSTUFF

Field of the Invention

The present invention relates to a method and system for determining the lineage of a commodity, and

particularly, but not exclusively, to a method and system, for determining the origin of a foodstuff.

Background of the Invention

In the food industry (especially the produce

industry), the origin of a foodstuff is important. The origin of foodstuffs is often associated with the quality. For example, Queensland (Australia) is renowned for good quality mangos and bananas. Consumers may place

importance on the origin of a foodstuff, which can

influence consumer purchasing habits and market prices.

The means of production or processing of a foodstuff can also influence consumer habits. For example, grass fed beef is often considered by consumers to be better quality than grain fed beef and organic vegetables are often considered to be better than chemically treated vegetables.

The geographical origin and the means of production are often used as marketing tools to promote foodstuffs. These attributes of a foodstuff can be used to

differentiate the product from other similar foodstuffs. Typical examples include "California oranges" or

"Queensland bananas".

Same foodstuffs comprise several components. Eggs comprise 3 components, the shell, the albumen and the yolk. This can make independent origin analysis complicated.

Summary of the Invention In accordance with a^' first aspect, the present

invention provides a method of classifying a

multi-component foodstuff comprising:

receiving a plurality of input values that represent the composition of a foodstuff component, each input value being derived from analysis of one foodstuff component; and

processing the inputs to determine the origin of the foodstuff by comparing the inputs with known values that correlate to defined origins.

"Origin" as utilised in this specification can mean either geographical origin or be indicative of a method of production of the foodstuff. The method as described is particularly useful for classifying the origin of a

multi-component foodstuff such as an egg. An egg

comprises three distinct components, the shell, yolk and albumen.

In an embodiment the method may comprise the

additional step of processing the plurality of input values to derive a single indicator value that is related to the three input values. The three input values are processed using a mathematical function such as a Linear Discriminant Analysis. Other mathematical functions can also be used.

In accordance with a second aspect, the present invention provides a system for classifying a multi- component foodstuff comprising:

an analytics module arranged to receive a plurality of input values, each respective input value based on the analysis of a respective component of a foodstuff having a plurality of components; the analytics module being configured to process the plurality of input values to derive an indicator value; and

the analytics module being arranged to determine the origin of the foodstuff using the indicator value.

In an embodiment the system comprises:

a solution module arranged to receive data from dissolving a component of the foodstuff in an acid;

an ablation module arranged to receive data from ablation of a component of the foodstuff; and

the solution module and ablation module arrange to perform spectrometry analysis on the received data to generate the plurality of input values.

In an embodiment the foodstuff is an egg and wherein one input value relates to the composition of the yolk, a second input relates to the composition of the albumen and a third input value relates to the composition of the shell and wherein the analytics module is arranged to derive an indicator value based on these three input values.

In accordance with a third aspect, the present

invention provides a method of creating a database useful for determining the origin of a foodstuff comprising the steps of:

dividing a foodstuff into a plurality of components; performing a spectrometry analysis on the plurality of components;

generating a plurality of input values from the spectrometry analysis, each respective input value

relating to the elemental composition of a component of the foodstuff; and

correlating the plurality of input values with a known origin.

The present invention is advantageous because it allows accurate determination of the origin of a multi-component foodstuff. Such a method and system can be used to regulate various foodstuff industries. In one example the present invention can be used to test- eggs to determine the origin of the eggs. This testing is

advantageous because it provides regulators with an accurate system and method to test the origin of eggs to ensure correct and accurate labelling of eggs. The present invention can also be used on other

multi-component foods to determine origin and can provide a similar advantage with respect to regulation and

accurate labelling of foodstuff.

The present invention further provides a method of - classifying a foodstuff comprising:

processing two or more constituent components from a multi-component foodstuff sample to determine a plurality of composition values for each of the two or more

constituent components,

deriving a supply chain indicator for the foodstuff from composition values determined from each of the constituent components, and

evaluating correlations between the supply chain indicator and a compatible indicator derived for a defined origin to evaluate the lineage of the foodstuff. ,

The present invention further provides a method of classifying a multi-component commodity, comprising:

receiving a plurality of input values that represent the composition of a component of the commodity, each input value being derived from analysis of one commodity component; and

processing the inputs to determine the origin of the commodity by comparing the inputs with known values that correlate to defined origins.

Detailed Description of the Drawings

An embodiment of the present invention will now be described, by way of example only, with reference to the accompanying figures, in which:

Figure 1 is a schematic diagram of a system for classifying a foodstuff in accordance with the present invention;

Figure 2 is a flow diagram showing. one embodiment of a method for classifying a foodstuff in accordance with the present invention;

Figure 3 is a flow diagram of another embodiment of a method for classifying a foodstuff, in particular an egg, in accordance with the present invention.

Description of Preferred/Specific Embodiments This specification discloses a system and method for classifying commodities, particularly foodstuffs,,

particularly the, origin of a foodstuff. The disclosed system and method are particularly useful in classifying the origin of a. multi-component foodstuff.

The term "origin" in the specification refers to the lineage of a foodstuff, including the production regime of used to produce the foodstuff, the geographical location of production, and the processing facilities associated with the foodstuff supply chain (such as packaging

facilities) .

"Foodstuff" in this specification refers to edible products that are used for food. Some foodstuffs can include have multiply constituent components, such as eggs, beef and avocados. Cows comprise skin, flesh and. bone (i.e. three separate components) . Eggs comprise shell, yolk and albumen (three separate components) .

Avocados comprise the seed (pip) , flesh and skin (three separate components) . This specification discloses a system and method for determining the origin of multi-component foodstuffs

(described primarily with reference to eggs). Embodiments of. the disclosed method use a ' supply chain indicator' to evaluate the supply chain origins of a foodstuff. The

^supply chain indicator' is derived from constituent elements of the foodstuff that are extracted from two or more components (such as elements extracted from the shell, yolk and albumen of an egg) . The ^supply chain indicator' is a compositional fingerprint that reflects the lineage of the foodstuff and facilitates

identification of various origins within the supply chain

(such as the production regime, geographical location of ^■■ production . and/or processing facilities in the supply chain) .

Eggs are frequently marketed based on origin. This includes the production regime used to produce the eggs and the geographical location of production and/or

processing. Eggs are generally produced by one of the following three production regimes-:

1. free range;

2. barn laid; or

3. , caged.

Free range eggs are produced by hens that are allowed to roam freely and forage food from the land. Barn laid ^! eggs are produced by hens that are enclosed in a barn and allowed to forage food from the barn floor . Cage eggs are produced by hens kept in cages.

It is often important to distinguish between cage eggs, free range eggs and barn laid eggs in order to regulate the industry and prevent false marking by egg producers. Egg cartons are often marked with the

production regime (particularly "free range" eggs). Generally "free range" eggs are considered superior and typically attract higher prices. It is important for industry regulators to be able to accurately determine the production regime used to produce eggs, to avoid false advertising .

Eggs can also be marketed based on the geographical location of production, including the location of the farm where the eggs were produced and the production facility were the eggs were processed (such as a packaging facility packaged) . Typical geographic identifiers include country or origin ( 'produced in Australia' ) or region of origin ('made in Queensland' ) markings . Fresh produce may be marked with the farm where the produce originated (for example Meggies Farm) or an indication that the produce was farmed locally. It is advantageous regulators and marketers to evaluate the lineage of produce (such as eggs) .

Eggs are an example of a multi-component foodstuff. Multi-component means - a foodstuff that comprises two or more constituent parts or elements. In the case of eggs, the shell, albumen and yolk are three individual

components that make up an egg.

Referring to Figure 1 an embodiment of a system for classifying a multi-component foodstuff in accordance with the present invention is shown. The system 1 comprises an interface for receiving input data, a processor with three inputs to generate an indicator value, wherein the

indicator value is used to classify the foodstuff. In this example embodiment, the interface and processor are implemented by a server or computing device having an ^■ appropriate user interface..

The server may be implemented by any computing architecture, including a standalone PC, client /server architecture, terminal/mainframe architecture, a laptop or any other appropriate architecture. The computing device is appropriately programmed to implement the invention. In this embodiment, the system may include one or more memory modules that store information relating indicator values with the origin of a foodstuff. The computing device may comprise a plurality of memory modules that store the relationships between indicator values and .

origin for many different foodstuffs, meaning which

indicator value relates to which various origins. These indicator and origin relationships may be stored as a lookup table or graph or in any other suitable form. In the specification "known indicator values" relates to the indicator values that have been related to the origin of a foodstuff.

In another embodiment, the known relationship between an indicator value and the origin of the foodstuff may be stored in a database that may be part of the computing device or separate from the computing device 100. The database can be accessed by the processor in order to compare a derived indicator value with the stored

indicator values relating to the origin of a foodstuff.

In an alternative, the database may not be separately administered and may be administered via an external party.

The memory modules and other modules described may be implemented as hardware or software. The modules could be implemented as a combination of hardware and software.

The modules may be separate from each other and may be arranged to communicate with each other. In other forms the modules may be integrated with each other as a single hardware or software package. In more detail the server 100 comprises suitable components necessary to receive, store and execute

appropriate computer instructions. The components may include a processing unit 102, read only memory (ROM) 104, random accessory memory (RAM) 106 and input/output devices such as disc drives 108, input devices 110 such as

ethernet port, USB port, etc. The system may further comprise a display 112 such as a liquid crystal display a light emitting display or any other suitable display and a communications link 114. The server 100 includes

instructions that may be encoded in the ROM 104, RAM 106 or disc drives 108 and may be executed by the processing unit 102. There may be provided a plurality of

communication links 114 which may variously connect to one or more computing devices such as another server personal computers, terminals, wireless or handheld computing devices. At least one of a plurality of communication links may be connected to an external computing network through a telephone line, Wi-Fi connection or other type of communication link.

The server 100 may include storage devices such as disc drives 108 which may encompass solid state drives, hard disc, optical drives or magnetic tape drives. The server 100 may use a single disc drives or multiple disc drives. The server 100 may also have a suitable operating system 116 which resides on" the disc drive 108 or in the ROM 104 of the server 100

The system has a database 120 residing on a disc or other storage device which is arranged to store at least one record 122 providing a relationship between an

indicator (i.e. indication value) and the origin of a foodstuff. As mentioned earlier, origin in this case can mean the location from where the foodstuff originated or the production regime by which the foodstuff was made. The database 120 is in communication with the sever 100. In another embodiment the database 120 may be implemented as part .of the server 100. The server further comprises an interface 121 provides means by which a user can input information into the computing system. The system may further comprise a display 125 that displays information to the user. In this case the display 125 can be used to illustrate the classification of the foodstuff to the user.

The system may further comprise a plurality of modules arranged to perform a variety of functions. As illustrated in Figure 1, the system 100 may comprise a solution module 400, an ablation module 401 and an

analytics module 402. These modules will be described later in greater detail with respect to the method of determining the origin of the foodstuff. These modules may be implemented as software modules within the

processor or as software instructions stored either on a disc drive, ROM or RAM of the system. In the illustrated embodiment the modules 400-403 are shown as existing in the processing unit 102. In another form, the modules may be separate hardware devices or may be separate servers or computing devices arranged to communicate with each other as well as the processing unit 102. In one form the modules 400-403 may be separate hardware devices arranged to communicate with each other.

In an alternate embodiment, the modules may be implemented in a single hardware devices or a single package. The modules 400-403 may be isolated from each other and in communication with each other even though they are part of a single package or device. In other forms the modules 400-403 may be implemented as software. The modules 400-403 may be implemented as separate software programs . that can interact or interface with each other. In a further alternative embodiment, the modules 400-403 may be implemented as a single software program, the program having the functionality of all the modules 400-403.

A method for classifying foodstuffs will be described with reference to Figures 2 and 3. Figure 2 depicts a general method for classifying a foodstuff. The method shown in Figure 2 can be implemented utilising the system illustrated in Figure 1.

The classification method illustrated in Figure 2 initiates with the system receiving a plurality of inputs that relate to the composition of a multi-component

foodstuff sample (step 201) . The system, typically

receives a plurality of inputs that are derived from analysis of individual foodstuff component (such as the shell, yolk or albumen of an egg) .

The inputs are compared with similar compositional records obtained from defined supply chain ^origins' (step 202). This is typically achieved by deriving a supply chain indicator for the foodstuff from the individual inputs and comparing the supply chain indicator with compatible indicators derived from defined origins in the foodstuff supply chain. The respective indicators are compared to evaluate the lineage of the foodstuff by compositional correlation (i.e. correlation of the

foodstuff supply chain indicator with foodstuff samples obtained from the defined origins in the supply chain) . The supply chain indicator defines a Composition

fingerprint' that facilitates evaluation of the foodstuff lineage (including the production regime for the foodstuff, the location of production and the supply chain processing facilities) .

The foodstuff is classified based on the derived origins at step 203. The compositional records for supply chain origins (i.e. the compositional data used to define origins in the foodstuff supply chain) are preferably stored in a database ( the supply chain database' ) . The supply chain database is established by testing sample foodstuffs from a plurality of origins and calculating the compositional records from the test samples. The

compositional records are typically compiled into an indicator fingerprint for the respective origin and stored in the supply chain database as a reference for- origin determinations. The records stored in the supply chain database may be periodically updated (by regular testing and evaluation) to ensure that the reference indicators accurately reflect the respective origin.

The compositional input values for a foodstuff component represent the concentration of defined analytes (elements) present in the sample. The compositional input values for the albumen and yolk of an egg can be

determined by dissolving the respective components in acid and peroxide and analysing the resulting solution by inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma atomic emission spectroscopy (ICP-AES) . The compositional input values for the egg shell can be determined by Laser Ablation Inductively Couple Plasma Mass Spectrometry (LA-ICP-MS) . This enables robust assay of the very high calcium matrix of the shell. More information can be provided if required. Between 20 and 100 analytes may be extracted from the foodstuff components and used as compositional inputs to the

classification process. Characterisation of the foodstuff typically improves with higher numbers of analytes as the indicator 'fingerprint' established for the sample

contains more compositional ^' detail .

The supply chain indicator is determined by

mathematically processing the compositional inputs to generate an indicator that reflects the composition of each foodstuff component (such as the shell, yolk and albumen or an egg) . The indicator value can be compared with reference indicator values for defined production regimes, production origins and processing facilities.

The reference indicator value relates to a known origin. The reference indicator values may be stored in a database or other data structure such as a look up table. The processor implementing the me-thod 200 can compare the calculated indicator value with the known indicator value to determine the origin of the foodstuff. The

relationship between an indicator value and an origin is derived by testing foodstuffs from a known origin,

calculating the indicator value for the particular

foodstuff and storing the relationship between the

indicator value and the origin in a database or other data structure such as a look up table. The foodstuff is classified based on the origin of the foodstuff once the origin of the foodstuff is determined.

The foodstuff is classified in terms of its Origin.

As mentioned earlier origin can encompass the production means/method used to produce the foodstuff or the

geographical location of origin of the foodstuff. In one form the method may comprise the additional steps of determining the geographical location of origin of the foodstuff and then processing the input values further to determine the production regime used to produce the foodstuff. The processing step utilises any suitable statistical or mathematical procedure to generate an indicator value. In one example, a plurality of input values are processed to generate an indicator value. The indicator value is derived .from the one or more input values by applying a multivariable statistical analysis to the one or more input values. One example of the multivariable

statistical analysis is linear discriminant analysis.

Principle Component Analysis (PCA), .Hierarchical Cluster Analysis (HCA) and neural networks may also be used to derive a supply chain indicator for the foodstuff. The processor unit 102 is arranged to execute a routine that performs the linear discriminant analysis on the , input values to generate an indicator value.

In order to classify the foodstuff the indicator value is checked against a set of known indicator values that have been correlated to known origins, in order to classify the foodstuff. The set of known indicator values is uploaded to a database or memory of the system, such as the RAM or ROM or disk drive. The set of known indicator · values is constructed prior to classifying foodstuffs with unknown origins. The known relationship between indicator values and foodstuff origins helps the system to classify a foodstuff with unknown origin.

An embodiment of the method for classifying a

multi-component foodstuff will be described in more detail with reference to Figure 3. The method of classifying a foodstuff will be described with reference to an egg. An egg is a multi-component foodstuff. The method described, with reference to Figure 3, can be used to classify the origin of an egg. The method can be used to determine either the geographical location of origin of an egg or the production regime used to produce the eggs or both. The egg is selected, from a set of eggs of unknown origin.

The method 300 comprises the first step 301 of separating each component of. the egg. In one example, the yolk and albumen can be removed from the egg. The yolk and albumen can be removed by any suitable process, for example by simply cracking the egg and separately the three components.

The yolk and albumen are separated at step 302. The yolk and albumen can be separated by any suitable process for example, using a centrifuge or by a separator. A sample of the yolk and albumen is removed for testing. The sample can be any suitable size. In one example the sample of yolk and sample of albumen are each l-2g in weight. The whole egg or the separated yolk and albumen can be freeze dried and finely ground to form' a powder. The finely ground powder ensures homogeneity of the sample. The samples are dried and the moisture loss is calculated. The amount of moisture will be used to generate input values relating to the composition of the yolk, or albumen, or shell or all three.

The yolk and albumen are dissolved in a solvent at step 303. The yolk and albumen are dissolved separate to each other. The yolk and albumen may be dissolved in a suitable solvent. In one example the solvent may be an acid such as nitric acid. As part of this step, each component of the egg is accurately weighed in triplicate into disposable, sterile centrifuge tubes and left to react at room temperature with redistilled concentrated nitric acid. Varying volumes of a base, such as hydrogen peroxide can be added to the samples. The samples are left to reflux at 80 degrees Celsius overnight. The resulting solutions are dried. Any other suitable method of preparing the samples using a solvent such as an acid can be used. The samples of at least the yolk and albumen were prepared for analysis using a spectrometer. This method is termed as "chemical digestion".

At step 304 the digested yolk and albumen is analysed to produce at least one input that relates to at least a partial composition of the yolk or albumen. Ideally at least input relating to the composition of the yolk is generated and at least one input relating to the albumen is generated. The inputs are sent to the processing unit 102 in order to generate at least one indicator value that is used to classify the egg in terms of the origin of the egg. The composition of the yolk and albumen is analysed using mass spectrometry. Other types of spectrometry, such as energy dispersive X-Ray spectrometry, may also be used to determine at least part of the composition of the yolk or albumen. Any suitable instrument can be used to perform the spectrometry procedure, for example an

inductively coupled plasma mass spectrometry instrument. One example is the Agilent 7700 inductively coupled plasma mass spectrometry instrument.

At step 304 the yolk and albumen is- analysed to determine trace metal concentrations in the solutions of digested yolk and albumen. A variety of trace metals can be tested for, some trace metals that can be tested for are Al, B, Be, Si, Ti, Cr, Nd, Em, Zr, V, Cu, Zn and so on. In another form various isotopes of various trace metals can be tested for also.

In an alternate embodiment, concentrations of major elements such as Na, K, Ca, Mg etc can be determined using a suitable spectrophotometry process. This process can be performed using any suitable spectrophotometer such as a Thermo iCAP Inductively coupled Plasma Atomic Emission. Spectrophotometer (ICP-AES) . In this embodiment, the inputs generated relate to the concentrations of major elements.

In an alternative embodiment, the inputs may relate to the concentrations of trace elements and major

elements. In this form the both spectrometry- and

spectrophotometry processes are used.

. The analysis of the albumen and yolk can .be performed at the solution module 400. The solution module is

arranged to receive the spectrometer or spectrophotometer ■data. The solution module 400 processes the raw data to determine at least a part of the composition of the yolk and a part of the composition of the albumen. The

solution module 400 can be programmed or arranged to determine the concentration of any one single element or the concentration of multiple elements, such as trace elements and major elements. The solution module 400 generates one or more input values that relate to the concentration of one or more elements that form the

composition of the yolk and albumen.

The shell also can be dissolved into a solvent and then analysed by a similar process as that used on the yolk and albumen to determine at least part of the

composition of the shell. However it was found by the applicant during validation of the method of classifying a food product, in particular for an egg, that the since the shell had such a large amount of calcium within it, it was not possible to differentiate the other elements that made up the composition of the shell using the chemical

dissolving method. .

Therefore an input value relating to at least part of the composition of the egg shell was derived using a different process. In particular, an input value relating to the shell has to be determined using a different process. At step 305 a small fragment of the shell is removed and prepared suitably for performing a laser ablation process. Preferably a number of fragments of a single egg shell are removed to ensure accuracy of the process. The shell fragments are mounted on a Perspex disc.

At step 306 the shell fragments are ablated using a laser ablation process. The laser ablation process is performed by any suitable instrument (e.g. a laser with appropriate electrolysis) . The laser ablation process may be combined with a mass spectrometry process in order to derive an input value that relates to at least a part of the composition of the egg shell. In one example the shell fragments are subjected to a Laser -Ablation

Inductively Coupled Plasma Mass Spectrometry process

(LA-ICP-MS) in order to determine the concentration of the trace metals within the^' shell. Suitable instruments include the UPI 213 nm YAG laser system and the Agilent 7500CS (both inductively coupled plasma mass spectrometer that facilitate the determination of. trace metal

concentrations). Some examples of trace metals measured . are Al, B, Be, Si-, Ti, Cr, Nd, Em, Zr, V, Cu, Zn and so on. Various isotopes of these can also be detected using the laser ablation mass spectrometry process.

In another form the concentrations of the major elements such as Na, K, Ca, Mg in the shell may also be determined using a suitable laser ablation and

spectrometry process such as Inductively coupled

Plasma-Atomic Emission Spectrometry.

The laser ablation process can be performed on the interior surface of the shell, the outer surface of the shell or both. In another form the data from the. laser

/ ablation inductively coupled plasma mass spectrometry process can be sent to an ablation module 401. The ablation module 401 is arranged to receive the raw data from the laser ablation mass spectrometry process. The ablation module 401 processes the raw data to determine at least a part of the composition of the shell. The

ablation module 401 can be programmed or arranged to determine the concentration of any one single element- or the concentration of multiple elements, such as trace elements and major -elements. The ablation module 401 generates one or more input values that relate to the concentration of one or more elements that form the composition of the shell.

At step 307 the input values are processed using a suitable function to derive an ^"indicator value. The indicator value is used to classify the foodstuff, in this case classify the egg, in terms of its origin. At step 307 the input values are processed using a Linear

Discriminant Analysis to derive the indicator value.

Linear Discriminant Analysis is performed on laser

ablation data and the chemical digestion data. Linear Discriminant Analysis is advantageous because this

statistical method can be used on a multiple input values to derive ^'at least a single indicator value. Linear

Discriminant Analysis provides a mechanism for visual distinction between various eggs.

. In one form the indicator value is derived by the analytics module 402. The analytics module 402 is

arranged to receive the input values generated from the laser ablation process and the chemical digestion process. In another form the analytics module 402 is arranged to receive input values generated from the ablation module 401 and the solution module 400. The analytics module 402 is further arranged to perform a Linear Discriminant Analysis on the input values to derive at least one indicator value. Preferably an iterative stepwise linear Discriminant analysis is used to derive at least one indicator value. The discriminant analysis is used to determine the variability between the various eggs. The generalised function for multiclass linear discriminant anal sis is:

This type of function can be modified in any suitable way and implemented by the analytics module 402 to derive at least one indicator value. The indicator value is a function of the input values.

In another form the input values can be processed to derive multiple indicator values. It should be, understood the input values can be manipulated in various ways to determine one or more indicator values (discriminant scores) that are used to classify the egg.

This type of statistical analysis is particularly useful for multi-variable data such as that obtained from trace material analysis (i.e. multi-element), as it

reduces and simplifies the entire data set into a single variable (i.e. a single indicator). The set of indicators generated are based on the combinations of predictor.

At step 308 the egg is classified using the indicator value. The egg is classified in terms of the origin of the egg, where origin can mean the geographical location of origin or the production regime used to produce the egg. The egg is classified using a set of known indicator values. The set of known indicator values are indicator values that have been correlated to known production regimes and known geographical origins. The set of known indicator values may be stored as a look up table in a text file or any other suitable file structure. In

another form the set of known indicator values may be stored in a database, such as database 120. The set of known indicator values are created by using eggs having a known origin, meaning a known production regime and from a known geographical location. The eggs with a known origin are subject to the same chemical digestion process to the yolk and albumen and the same laser ablation process for the shell to generate input values. The input values are analysed using the linear discriminant analysis to

generate indicator values. The indicator values are correlated with the origin because the eggs are of known origin. The known correlation is stored in the database 120 or in another file structure.

The known indicator values or the database of known indicator values are used to classify unknown eggs at step 308. The indicator values generated at step 307 are compared with the known indicator values to classify the egg. The known indicator values can be used^' as part of the linear . discriminant analysis to classify unknown eggs. The indicator values generated at step 307 may be plotted on a scatter plot to show the discrimination. This

scatter plot can be generated by the analytics module 402.^' The analytics module is further arranged to classify the eggs based on the known indicator to origin relationship.

In one form the indicator value is essentially formed from three input values, one input value for the ablation of the shell, one input for the chemical analysis of the yolk and one input for the chemical analysis of the

albumen. The indicator value is generated from these three input values. Each of the. input values is used to determine the indicator value using the linear discriminant analysis.

At step 309 the eggs are classified based on the geographical location from where the egg originated. The eggs can be classified at any suitable resolution for the geographical location. Resolution in terms of

geographical location means how detailed the geographical location is. The egg can be classified by the general state of origin or the egg can be classified in terms of the specific farm the egg was produced at. The resolution of the geographical location can be determined . based on the detail of the database of known correlation between the geographical location and the indicator values.

At step 310 the eggs are classified based on the production regime of. the egg. The egg is classified based on production regime of the egg by checking the indicator value with the known indicator values. The steps 309 and 310 may be sequential to each other and may follow on from each other Alternatively only one or the other step may be performed. It should be understood a foodstuff

(i.e. the egg in this example) can be classified in terms of geographical location of origin and production regime or as either one.

These steps may also be implemented by the analytics module 402. These classifications may be displayed in any suitable form, for example a scatter plot or a table or any other format.

The method and system for classifying . a foodstuff is advantageous because it provides a robust classification method that can be applied to classify unknown foodstuffs. In the particular example of an egg, the method provides farmers and particularly industry regulators a method to correctly classify an egg in terms of its origin. The method allows regulatory bodies to correctly identify what production regime was used to produce the eggs. The method also allows regulatory bodies to correctly identify the geographical location or particular farm eggs came from. This allows regulatory bodies to control the

industry and labelling of eggs. The method allows

regulatory bodies to maintain standards of the industry and ensure correct labelling of the origin of eggs in order to provide accurate information to consumers.

- The method of classifying a foodstuff relies on a plurality of inputs, processing the inputs to derive an indicator value. The method is a robust method because it uses a plurality of inputs. In the case of the egg, the method uses the composition data (trace metal

concentrations)^' of the yolk, albumen and shell to provide' a more accurate method than using the trace metal

concentration of only one part of the egg.

Trace metal concentrations were measured because that provides a more accurate measure of origin of the egg.

This is because free range eggs are likely to have

different trace metals concentrations than eggs produced by barn or cage kept hens. These trace metals appear as part of the egg. Therefore measuring the concentration of trace metals provides one variable to classify the

production type of eggs.

In alternate forms any suitable statistical technique can be used to determine the indicator value.- Any other suitable statistical technique for multiple variables can be used to determine at least one indicator value.

In further alternate forms the method of classifying a foodstuff can be used to classify other foodstuffs such as beef, vegetables, fruits, poultry or any other

foodstuff. The method and system of the present invention also has application in other embodiments to a wider range of commodities than foodstuffs. It may be applied to fibres, for example .

A fibre such as cotton, for example, may have a number of components, particularly in the early stages of processing to produce cotton bale. It may include

components of seed, leaf as well as the cotton fibre.

These multiple components can each be analysed in

accordance with the processes described above in relation to foodstuffs. The result can be a classification of origin of the fibre.

The approach may be applied to other commodities.

It will be appreciated by persons skilled^' in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments ^' without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Claims

1. A method of classifying a multi-component foodstuff comprising:

2. _;A method in accordance with Claim 1 wherein the method comprises the further steps of:

determining an indicator value from the plurality of input values; and

comparing the indicator value with known indicator values that^' are correlated with defined origins, to determine the origin of the foodstuff.

.

3. A method in accordance with Claim 2, wherein the indicator value is derived from a combination of the plurality of input values.

4. A method in accordance with claim 2 or claim 3-, wherein the step of determining the indicator value, comprises the step of deriving the indicator value from the input values using a multivariable statistical

procedure using the -indicator and value for origin

determination comparisons.

5. A method in accordance with any one of the preceding claims, wherein input values comprise a trace element composition of a corresponding component of the foodstuff.

6. A method in accordance with claim 5, wherein input values comprise a trace metal composition of a

corresponding component of the foodstuff.

7. A method in accordance with claim 3 or claim 4, wherein the multivariable statistical procedure is a linear discriminate analysis that is performed on the plurality of input values to derive the indicator value.

8. A method in accordance with any one of the preceding claims, comprising the step of performing a spectrometric analysis on a foodstuff component to generate! one or more of the plurality of input values.

9. A method in accordance with claim 7, comprising the steps of :

dissolving a component of the foodstuff in an acid; ablating a component of the foodstuff; and

performing the spectrometry analysis on the dissolved component of the foodstuff and the ablated portion of the foodstuff to generate the one or more input values.

10. A method in accordance with any one of the preceding claims, wherein at least three input values are generated, each input value relating to one component of the

foodstuff.

11. A method in accordance with any one of the preceding claims, wherein the foodstuff is an egg, the egg

comprising three components, the shell, the albumen and the yolk.

12. A method in accordance with claim 11, wherein one input value relates to the composition of the yolk, another input value relates to the composition of the albumen and a third input value relates to the composition_, of the shell. ·

13. A system for classifying a multi-component foodstuff comprising:

an analytics module arranged to receive a plurality of input values, -each respective input value based on the analysis of a respective component of a foodstuff having a plurality of components;

the analytics module being configured to process the plurality of input values to derive an indicator value; and

14. A system in accordance with Claim 13, wherein the analytics module is configured to determine the origin of the foodstuff by, comparing the indicator value with a set of known indicator values that have been correlated to a known origin.

15. A system in accordance with any one of claims 13 to 14, comprising:

a database, the database arranged to store the set of known indicator values, wherein the known indicator values are correlated to a known origin; and

the analytics module arranged to compare the indicator value with the known indicator values from the database in order to classify the foodstuff.

16. A system in accordance with any one of Claims 13 to 15, wherein the analytics module is arranged to perform a multivariable statistical procedure on the plurality of input values to derive the indicator value.

17. A system in accordance with any one of Claims 13 to Claim 16, wherein the one or more input values relate to trace element composition of the foodstuff.

18. A system in accordance with any one of Claims 13 to

17, wherein the analytics module is arranged to perform a linear discriminate analysis on the plurality of input values to derive the indicator value.

19. A system in accordance with any one of Claims 13 to

18, wherein the foodstuff is an egg and wherein one input value relates to the composition of the yolk, a second input value relates to the composition of the albumen and a third input value relates to the composition of the shell, and wherein the analytics module is arranged to derive an indicator value based on these three input values. ^' _}

20. A method of creating a database useful for determining the origin of a foodstuff comprising the steps of:

relating to the elemental composition of a component of the foodstuff; and

correlating the plurality of input values with a known origin .

21. A method in accordance with Claim 2,0, wherein the foodstuff is an egg, the method comprising the steps of: generating three input values, a first input value relating to the elemental composition of the yolk, a second input value relating to the elemental composition of the albumen, a third input value relating the elemental composition of the shell; and

correlating each respective input value with a known origin.

22. A method in accordance with Claim 21, wherein the first and second input values are derived by chemical digestion and then performing a spectrometry analysis on the chemically digested data, the third input value is generated by laser ablation of the shell followed by performing a spectrometry analysis. ¹

23. A method in accordance with claims 20, 21 or 22, wherein the one or more input values are processed to determine a single indicator value, the indicator value being correlated with a known origin.

24. A method of classifying a foodstuff comprising:

constituent components,

¹ deriving a supply chain indicator for the foodstuff from composition values determined from each of the constituent components, and

evaluating correlations between the supply chain indicator and_,;a compatible indicator derived for a defined origin to evaluate the lineage of the foodstuff.

25. A method of classifying a multi-component commodity, comprising :

26. A computer program, comprising instructions for controlling a computer to implement a method in accordance with any one of claims 1 to 12, or claims 20 to 23 or claim 24 or claim 25.

27. A computer readable medium, providing a computer program in accordance with claim 26.

28. A data signal, comprising a computer program in accordance with claim 27.