WO2018078582A1

WO2018078582A1 - Method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue and corresponding system arranged to execute said method

Info

Publication number: WO2018078582A1
Application number: PCT/IB2017/056694
Authority: WO
Inventors: Angelo Duilio TRACANNA; Elena Maria BOZZETTA; Marzia PEZZOLATO; Serena MEISTRO; Mario Botta
Original assignee: Istituto Zooprofilattico Sperimentale Del Piemonte, Liguria E Valle D'aosta
Priority date: 2016-10-28
Filing date: 2017-10-27
Publication date: 2018-05-03
Also published as: IT201600109354A1

Abstract

A method is described for distinguishing fresh meat from meat subjected to freezing, comprising an acquisition step (900) of at least one image of the muscle tissue sample, a step (1000) of generating at least one counting matrix of the occurrences of colours of the pixels that make up the image, a step (1100) of quantization of the image pixel colours, a step (1200) of converting the acquired image into a reduced colour image, a step (1300) of recognition of vacuoles in the reduced colour image, a comparison step (1400), in which the number of vacuoles (10) recognised is compared with a threshold, determining that the sample of muscle tissue has been frozen if the comparison step determines that the number of vacuoles (10) recognised is greater than the threshold, and determining that the sample of muscle tissue has not been frozen if the comparison step determines that the number of vacuoles (10) recognised is smaller than the threshold.

Description

Method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue and corresponding system arranged to execute said method

Technical field

The present invention relates in general to the field of analysis methods of samples of meat for distinguishing fresh meat from meat subjected to freezing. In particular, the invention relates to a method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples and to a corresponding system for executing said method.

Background art

In the field of analytical methods applicable to distinguish fresh meat from meat previously subjected to freezing, processes are known which require a plurality of complex steps that need to be performed manually by highly qualified laboratory technicians. Moreover, the above steps can be carried out only in suitable environments, such as scientific laboratories provided with sophisticated laboratory equipment.

In particular, as regards the muscle tissue of fish, at the histopathology laboratory of IZ- SPLVA, the histological examination for the differentiation of fresh fish from that defrosted has been developed, standardized, validated and accredited. This method allows to accurately and precisely differentiate the preservation state of fish and is applicable both to bony fish and cartilaginous fish, as its validity is not affected by the fish species analyzed. Histological examination is based on the microscopic observation of a small amount of fish muscle tissue, which is previously subjected to fixing, processing, inclusion, microtome cutting and colouring with hematoxylin-eosin.

The analysis aims to identify microscopic alterations from freezing in the muscle meat of the fish under examination. Freezing in fact causes the formation of ice crystals and as a result, of so-called vacuoles: the latter are microstructural alterations in the form of empty spaces or cavities with clear edges, located at intracellular level. The presence of one or more vacuoles is index of previous sample freezing.

Disadvantageously, histological examination cannot be executed by users other than expert laboratory technicians and in environments other than appropriately equipped laboratories, such as residential environments or supermarkets. This limitation therefore affects the possibility of extending the application of the histological examination.

Moreover, even when the above procedure is performed by expert laboratory staff, a long and complicated series of steps is required for obtaining an analysis result.

Moreover, the fact that the microscopic observation of the histological preparations is necessarily carried out by trained and skilled operators does not exclude the fact that the results may sometimes be compromised by any human error.

A further drawback is the fact that the various steps of the procedure described above cannot at present be automated and performed by electronic systems, but have to be carried out manually by qualified laboratory staff. The histological examination at present, albeit highly standardized, does not allow to overcome the following issues that may occur during the procedure:

- problems of colorimetric differences that can afflict the different tissue samples taken;

- issues of non-uniformity and standardization of illumination with which tissue samples are photographed, which cause tissue and background brightness and colouring levels significantly different between various images of the sample;

- issues of presence of tissue brakes or stains in the samples which may be misinterpreted as vacuoles, but which in fact are not, especially when the samples are stored in "poor" storage condition (autolytic phenomena of the sample that can interfere negatively in the detection of vacuoles). The impossibility to automate, known procedures involves definitely long times for the completion of a sample analysis process, a limited diffusion of the processes in external uses from specialized laboratories and increased costs of conducting such analysis.

Summary of the invention

The present invention therefore aims to provide a satisfactory solution to the problems described above, while avoiding the drawbacks of prior art.

According to the invention, such an object is achieved by a method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of histological preparations of muscle tissue samples and a corresponding system arranged to execute said method, having the features set out in the independent claims.

Particular embodiments are the subject of the dependent claims, whose content is to be understood as an integral or integrating part of the present description.

In brief, the present invention consists in a method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples, which is based on the principle of qualitative analysis of histological samples of muscle tissue, obtained using automated analysis steps executable with the aid of a computer.

The principle is based on the fact that tissue samples of meat subjected to freezing have optically empty spaces with clear edges, called "vacuoles", at intracellular level. This phenomenon occurs probably as a result of the expansion of water, caused by hydrogen bonds, up to the formation of crystals.

In particular, the method object of the invention is able to detect the presence of vacuoles, identify them and perform a count of those present in a tissue sample.

In brief, the method of the invention comprises the following steps:

an acquisition step of at least one image of a muscle tissue sample, wherein the im- age comprises a plurality of pixels each of which having a respective colour;

a generation step of at least one counting matrix of the pixel colour occurrences that compose the image, wherein each element of the matrix is defined by a predetermined set of colorimetric coordinates in a colour space and it has a numeric value corresponding to the number of occurrences of the colour represented by the colorimetric coordinates in the acquired image;

a quantization step of the image pixel colours, which comprises a selection of a predetermined number of reference colours in the colour space on the basis of the number of occurrences of colours of the image in the at least one counting matrix of the colour occurrences;

a conversion step of the acquired image into a reduced colour image comprising a number of conversion colours corresponding to the predetermined number of reference colours, wherein the image acquired is converted into the reduced colour image having a first conversion colour and a second conversion colour;

a recognition step, in the reduced colour image, of areas corresponding to vacuoles in the muscle tissue of the meat, based on the identification of areas including aggregations of adjacent pixels having a predetermined conversion colour and belonging to a predetermined shape category, wherein the number of pixels having said predetermined conversion colour in the aggregations of pixels is greater than a predetermined threshold percentage of the total number of pixels of the identified area;

a comparison step, wherein the number of vacuoles recognized in the recognition step of areas corresponding to vacuoles is compared with a predetermined numerosity threshold of vacuoles;

determine that the muscle tissue sample has been frozen if from the comparison step it is determined that the number of recognized vacuoles is greater than the predetermined numerosity threshold of vacuoles;

determine that the muscle tissue sample has not been frozen if from the comparison step it is determined that the number of recognized vacuoles is smaller than the predetermined numerosity threshold of vacuoles.

The invention also provides a system arranged to execute the method described above. Brief description of the drawings

Further features and advantages of the invention will appear more clearly from the following detailed description of an embodiment thereof, given by way of non-limiting example with reference to the accompanying drawing, in which:

- figure 1 shows an image of a histological sample of muscle tissue of frozen fish, in which the vacuoles are highlighted by circles;

- figure 2 shows a flow chart of the method of the invention;

- figure 3 shows a recognition step of areas corresponding to vacuoles in the muscle tissue of said meat carried out on an exemplary reduced colour image; and

- figures 4A and 4B show two embodiments of a system for executing the method according to the invention.

Detailed description

Before explaining a plurality of embodiments of the invention in detail, it should be noted that the invention is not limited in its application to the construction details and to the configuration of the components presented in the following description or shown in the drawings. The invention can take other embodiments and be implemented or practically carried out in different ways. It should also be understood that the phraseology and terminology are for descriptive purpose and are not to be construed as limiting. The use of "include" and "comprise" and variations thereof are intended as including the elements cited thereafter and their equivalents, as well as additional elements and equivalents thereof.

With initial reference to figure 1 , it shows an image of a histological sample of muscle tissue frozen fish which is usually taken to be subjected to analysis. For illustration purpose, circles are represented at the points where the vacuoles are.

This image represents the starting point of the method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples according to the invention. Preferably, the tissue sample has a length of 1.5 cm, a width of 1 cm and a thickness of 0.5, and the image acquired of such a sample has a resolution higher than 640 x 480 pixels.

With reference to the flow chart in figure 2, there is described in detail a first embodiment of the method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples of said meat, object of the invention.

The method comprises an acquisition step 900 of at least one image of said muscle tissue sample, said image comprising a plurality of pixels each of which having a respective colour.

Moreover, it is comprised a generation step 1000 of at least one counting matrix of the pixel colour occurrences that compose said image, wherein each element of the matrix is defined by a predetermined set of colorimetric coordinates in a colour space and it has a numeric value corresponding to the number of occurrences of the colour represented by said colorimetric coordinates in the acquired image.

Preferably, in the generation step 1000, two separate matrices are generated, respectively a first counting matrix of the light colour occurrences, ClearArray[R][G][B], intended to count the occurrences of the colours having a Cartesian distance from white lower than the Cartesian distance from black, and a second counting matrix of the dark colour occurrences, DarkArray[R][G][B], intended to count the occurrences of colours having a Cartesian distance from black lower than the Cartesian distance from white, each of which is a three- dimensional matrix wherein each dimension represents a colorimetric coordinate in the RGB colour space.

Each colorimetric coordinate has a value of between 0 and 255 and indicates the quantity of red colour, the quantity of green colour and the quantity of blue colour which serve to define a colour of an analyzed pixel.

It should be noted that, in this embodiment, the index value is referred to the space of RGB colours, but other spaces of reference colours may be used, such as CMYK or HSV, having different numerical indexes, with suitable modification of the algorithm to keep into account the coordinate system used in calculating the distances.

Step 1000 of generating the two counting matrices of the pixel colour occurrences that make up said image further comprises the steps of assigning to each element of each matrix an initial value equal to zero

ClearArray[R][G][B]=0;

DarkArray[R][G][B]=0;

and identifying the colour of each of the pixels of the image through three indexes in the RGB colour space.

For each pixel, when the distance of said colorimetric coordinates from black is greater than the distance of said colorimetric coordinates from white, the value of the element of the counting matrix of the light colour occurrences defined by the identifying colorimetric coordinates of colour of the pixel is increased by one unit

ClearArray[R][G][B]= ClearArray[R][G][B] + 1 or when the distance of said colorimetric coordinates from white is greater than the distance of said colorimetric coordinates from black, the value of the element of the counting matrix of the dark colour occurrences defined by the identifying colorimetric coordinates of colour of the pixel is increased by one unit

DarkArray[Rl[G][B]= DarkArray[R][Gl[B] + 1.

The method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples of said meat, according to the invention, further comprises a quantization step 1 100 of the colours of pixels of the image, comprising a selection of a predetermined number of reference colours in^{^}the colour space on the basis of the number of occurrences of colours of the image in said at least one counting matrix of the colour occurrences. For example, quantization is a quantization by distance.

Preferably, in the quantization step 1 100, a reference dark colour is selected, representative of the cellular tissue, corresponding to the colour defined by the colorimetric coordinates of the element of the counting matrix of the dark colour occurrences having the highest value, and a reference light colour is selected, representative of the interstitial tissue, corresponding to the colour defined by the colorimetric coordinates of the element of the counting matrix of the light colour occurrences having the highest value.

The method according to the invention further comprises a conversion step 1200 of said acquired image into a reduced colour image comprising a number of conversion colours corresponding to said predetermined number of reference colours.

In this currently preferred embodiment, the reference colours are two and the first conversion colour is black and the second conversion colour is white.

In particular, the first black conversion colour is assigned to each pixel of the reduced colour image corresponding to a respective pixel of the acquired image having a colour with a Cartesian distance from the dark reference colour lower than the Cartesian distance from the light reference colour and said second white conversion colour is assigned to each pixel of the reduced colour image corresponding to a respective pixel of the acquired image having a colour with a Cartesian distance from the dark reference colour greater than the Cartesian distance from the light reference colour.

The method according to the invention further comprises a recognition step 1300, in the reduced colour image, of areas corresponding to vacuoles in the muscle tissue of said meat. The recognition step 1300 is based on the identification of areas including aggregations of adjacent pixels having a predetermined conversion colour and belonging to a predetermined shape category, wherein the number of pixels having said predetermined conversion colour in said aggregations of pixels is greater than a predetermined threshold percentage of the total number of pixels of said identified area.

Vacuoles can be defined as optically empty spaces, with clear edges, located at intracellular level.

Keeping the exemplary assumption described above, i.e. that the image is recreated using white and black, white spots substantially regularly shaped are searched in the image for the identification of the vacuoles. This identification step uses two process colours, defined as "first process colour" and "second process colour". These two colours can be any colour, as long as they are still distinguishable from each other and different from the colours selected for the previous conversion step.

The recognition step 1300 can start from any one of the pixels that make up the reduced colour image and extent iteratively to all the other pixels. It is preferable to start the analysis from one of the pixels forming the angles of the image. For example, in the embodiment described herein, the analysis started from the higher leftmost pixel.

It is verified if the pixel under analysis is white.

If it is white, it is considered as seed pixel and starting from it, an identification step of an area including an aggregation of white adjacent pixels belonging to a predetermined shape category is carried out. The aggregation of white adjacent pixels is identified as follows. An iterative colour verification is carried out analysing all the pixels immediately adjacent to the seed pixel, then analysing the pixels adjacent to the preceding white verified pixels and so on.

The term adjacent preferably indicates a laterally adjacent, i.e. above, below, left and right.

The iterative colour verification ends when there are no more white pixels adjacent to the white verified pixels, without considering the pixels already analysed.

The colour of each pixel verified to be white is changed in a "first process colour".

The area including the aggregation of pixels verified as white and converted into the first process colour is automatically circumscribed by a reference area 30.

Figure 3 shows an exemplary schematisation of a recognition step of a vacuole area in which a reference area 30 is defined. In this currently preferred embodiment, the reference area 30 has a quadrilateral polygonal shape circumscribed to areas 32 in which there are the pixels of the first process colour.

Once the reference area 30 has been traced, in which there is supposed to be a vacuole 10, a verification of the vacuole is executed, which can give a positive result, in which case it is confirmed the presence of a vacuole 10, or a negative result, in which case it is not confirmed the presence of a vacuole 10.

The vacuole verification function is used to verify that a regularly shaped white spot detected in the image is actually a vacuole 10, thus eliminating false positives that could compromise the sample analysis and distort the result of the process.

If from the verification it is not confirmed the presence of a vacuole, all pixels having the first process colour in the reference area are coloured with a second process colour.

The second process colour is assigned to ensure that the pixels of the first process colour already analyzed and detected as false vacuoles are not re-analyzed as potential vacuoles during subsequent pixel analysis.

The vacuole verification returns a positive result if the following conditions occur:

- the reference area 30 circumscribing the area 32 including the aggregation of verified white pixels and converted into the first process colour is surrounded by black pixels, i. e. it is not adjacent to interstitial tissue areas;

- the reference area 30 is a flat image with a fixed geometric relation between two dimensions thereof, preferably a flat quadrilateral image having the major side not longer than twice the minor side;

- the reference area 30 does not contain more than 5% of black pixels compared to the number of pixels having the first process colour within the perimeter thereof.

If the verification confirms the presence of a vacuole, the value of a counter adapted to count the vacuoles recognized by the verification is increased.

At the end of the identification step of an area including an aggregation of adjacent white pixels starting from the current seed pixel, the analysis of pixels continues by searching a subsequent white pixel to be used as seed pixel for a subsequent identification step of another area including an aggregation of adjacent white pixels. Starting from the new seed pixel, the analysis described above is carried out. The search for subsequent white pixels to be used as seed pixels continues for each pixel of the acquired image according to a predetermined scanning direction of the image.

In conclusion, the number of vacuoles 10 recognised and counted is compared with a predetermined numerosity threshold of vacuoles in a comparison step 1400.

It is determined that the muscle tissue sample has been frozen if the number of recognized vacuoles is greater than the numerosity threshold of vacuoles, and vice versa, that the sample of muscle tissue has not been frozen if the number of recognized vacuoles is smaller than said numerosity threshold of vacuoles.

In the currently preferred embodiment, the predetermined threshold is equal to five.

The invention further comprises a system 60 for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples.

System 60 comprises processing means 70 adapted to execute all the steps of the method of the invention described above.

In a first embodiment illustrated in figure 4A, the acquisition of the image at the processing means includes reading numerical data representative of an image previously acquired by a remote system, such as an image database 72. The processing means 70 include, for example, a screen adapted to report whether the tissue sample is recognized to be a sample that has been subjected to freezing and possibly adapted to provide more information, such as the exact number of vacuoles detected.

System 60, in an integrated embodiment shown in figure 4B, further comprises an area 61 for housing a sample of muscle tissue adapted to accommodate the muscle tissue sample under analysis, image acquisition means 63 adapted to collect an image of the muscle tissue sample, processing means (not shown) adapted to execute all the steps of the method of the invention and signaling means (indicated with 62, 64 in association with system 60) adapted to report whether the tissue sample is recognized to be a sample that has been subjected or not freezing.

The signaling means 62 in association with system 60 may for example be two Leds 62, respectively a green one to indicate that the sample has not been subjected to freezing, and a red one to indicate that the sample has undergone a freezing process, and/or a screen 64 adapted to provide further information, such as the exact number of vacuoles detected.

The advantage achieved by the invention consists in the possibility of carrying out analysis for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples even in places other than highly equipped scientific laboratories.

Advantageously, the method of the invention can be easily automated as its steps overcome the issues of automation of known processes. As a result, automation reduces analysis time, promotes the spread of the method even to uses external to specialised laboratories and reduces the cost of conducting analyses.

Several aspects and embodiments of a method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples according to the invention and a corresponding system arranged to execute said method have been described. It is understood that each embodiment may be combined with any other embodiment. The invention, moreover, is not limited to the described embodiments, but may be varied within the scope defined by the appended claims.

Claims

1. A method for distinguishing fresh meat from meat subjected to freezing by means of analysis of images of muscle tissue samples of said meat, comprising:

an acquisition step (900) of at least one image of said muscle tissue sample, said image comprising a plurality of pixels each of which having a respective colour;

a generation step (1000) of at least one counting matrix of the pixel colour occurrences that compose said image, wherein each element of the matrix is defined by a predetermined set of colorimetric coordinates in a colour space and it has a numeric value corresponding to the number of occurrences of the colour represented by said colorimetric coordinates in the acquired image;

a quantization step (1 100) of the image pixel colours, comprising a selection of a predetermined number of reference colours in the colour space on the basis of the number of occuiTences of colours of the image in said at least one counting matrix of the colour occurrences;

a conversion step (1200) of said acquired image into a reduced colour image comprising a number of conversion colours corresponding to said predetermined number of reference colours;

a recognition step (1300), in the reduced colour image, of areas corresponding to vacuoles in the muscle tissue of said meat, based on the identification of areas including aggregations of adjacent pixels having a predetermined conversion colour and belonging to a predetermined shape category, wherein the number of pixels having said predetermined conversion colour in said aggregations of pixels is greater than a predetermined threshold percentage of the total number of pixels of said identified area;

a comparison step (1400), wherein the number of vacuoles (10) recognized in the recognition step (1300) of areas corresponding to vacuoles is compared with a predetermined numerosity threshold of vacuoles;

determine that the muscle tissue sample has been frozen if from the comparison step it is determined that the number of recognized vacuoles (10) is greater than said predetermined numerosity threshold of vacuoles; and determine that the muscle tissue sample has not been frozen if from the comparison step it is determined that the number of recognized vacuoles (10) is lower than said predetermined numerosity threshold of vacuoles.

2. A method according to claim 1 , wherein the generation step (1000) of at least one counting matrix of the pixel colour occurrences that compose said image comprises the steps of:

generating two separate matrix, respectively a first counting matrix of the light colour occurrences, intended to count the occurrences of the colours having a Cartesian distance from white lower than the Cartesian distance from black, and a second counting matrix of the dark colour occurrences, intended to count the occurrences of colours having a Cartesian distance from black lower than the Cartesian distance from white, each of which is a three-dimensional matrix wherein each dimension represents a colorimetric coordinate in the RGB colour space.

3. A method according to claim 2, wherein each colorimetric coordinate has a value comprised between 0 and 255, and the generation step (1000) of said two counting matrix of the pixel colour occurrences that compose said image further comprises the steps of: assigning to each element of each matrix an initial value equal to zero;

identifying the colour of each of the pixels of the image through three indexes in the RGB colour space;

for each pixel, when the distance of said colorimetric coordinates from black is greater than the distance of said colorimetric coordinates from white, increasing by one unit the value of the element of the counting matrix of the light colour occurrences defined by the identifying colorimetric coordinates of colour of the pixel;

for each pixel, when the distance of said colorimetric coordinates from white is greater than the distance of said colorimetric coordinates from black, increasing by one unit the value of the element of the counting matrix of the dark colour occurrences defined by the identifying colorimetric coordinates of colour of the pixel.

4. A method according to claim 3, wherein the quantization step (1 100) of pixel colours of the image comprises the steps of: selecting a reference dark colour, representative of the cellular tissue, corresponding to the colour defined by the colorimetric coordinates of the element of the counting matrix of the dark colours occurrences having the highest value;

selecting a reference light colour, representative of the interstitial tissue, corresponding to the colour defined by the colorimetric coordinates of the element of the counting matrix of the light colour occurrences having the highest value.

5. A method according to claim 4, wherein a first conversion colour is black and a second conversion colour is white, and the conversion step (1200) of said acquired image into a reduced colour image comprises the steps of:

assigning said first black conversion colour to each pixel of the reduced colour image corresponding to a respective pixel of the acquired image having a colour with a Cartesian distance from the dark reference colour lower than the Cartesian distance from the light reference colour;

assigning said second white conversion colour to each pixel of the reduced colour image corresponding to a respective pixel of the acquired image having a colour with a Cartesian distance from the dark reference colour greater than the Cartesian distance from the light reference colour.

6. A method according to any one of the preceding claims, wherein the area belonging to a predetermined shape category is an area circumscribed by a predetermined plane figure having a predetermined geometric relationship between two of its dimensions.

7. A method according to claim 6, wherein the plane figure is a rectangle having a ratio between the short side and the long side greater than 0.5.

8. A method according to any one of the preceding claims, wherein the identification of areas including aggregations of adjacent pixels of a predetermined conversion colour in the recognition step (1300) comprises the steps of:

a) selecting a pixel of the colour reduced image;

b) verifying that the pixel under analysis is white;

c) if the pixel under analysis is white, consider the pixel under analysis as a seed pixel;

d) verifying whether the colour of all pixels immediately adjacent to the seed pixel is white;

e) analysing the pixels adjacent to the preceding pixels verified to be white, and verifying whether the colour of these pixels is white;

f) iterating step e);

g) changing the colour of each pixel verified to be white with a first process colour; h) if there is no longer any white pixel adjacent to pixels verified to be white according to steps c), d), e), not considering pixels already analysed, circumscribe a reference area (30) including the aggregation of pixels verified to be white and converted into the first process colour;

i) executing a verification of vacuole which can give a positive result, in which case it is confirmed the presence of a vacuole (10), or a negative result, in which case it is not confirmed the presence of a vacuole (10);

1) if from the verification it is not confirmed the presence of a vacuole, colour with the second process colour all pixels having the first process colour in the area of reference (30);

m) continuing the analysis of the pixels of the image, searching a successive white pixel to be used as a new seed pixel, and starting from said new seed pixel, iterating steps d) to 1).

9. A method for distinguishing fresh meat from meat subjected to freezing by means of the analysis of images of muscle tissue samples according to claim 8, wherein the nu- merosity threshold of vacuoles of comparison step (1400) is equal to five.

10. A system (60) for distinguishing fresh meat from meat subjected to freezing by means of the analysis of images of muscle tissue samples comprising processing means (70) adapted to execute all the steps of the method according to any one of preceding claims.

1 1. System (60) according to claim 10, wherein the system (60) comprises:

- a housing area (61 ) for a muscle tissue sample adapted to accommodate the mus- cle tissue sample under analysis;

- image acquisition means (63) adapted to acquire an image of the muscle tissue sample;

- signalling means (62, 64) adapted to signal if the tissue sample is recognized to be a sample which has been subjected or not to freezing.

12. A system (60) according to claim 1 1 , wherein the signalling means (62) are two LEDs, respectively one green to indicate that the sample has not been subjected to freezing and one red to indicate that the sample has been subjected to a freezing process, and/or a screen (64) adapted to provide further information.