Publicat în Lucr.St. FMV Timisoara, 2011




1 Faculty of Veterinary Medicine, Timisoara, Calea Aradului 119, 300645 Romania

2 College of Veterinary Surgeons, Timis. Tel.0040766719381.






The principle of Transmittance (T) is proposed for a test to detect mastitis cow side with the aid of a portable electronic device, called Milktester RMM. Transmittance value depends on the concentration, color and homogeneity of a sample, crossed through by a beam of monochromatic electromagnetic radiation. Milktester RMM is o prototype designed to detect and quantify modifications of homogeneity and color which characterizes an abnormal milk, in accordance with the current Regulation (EC) No 853/2004 of the European Parliament. The proposed optical device is able to measure, calculate and display, in a few seconds, the transmittance coefficient of intensity (TCI) and deviation coefficient (DC) which, in their turn, is influenced by at least three kinds of alterations: concentration in different substances, color and turbidity. Every reading for every quarter is taken into account and calculates by comparison to the average value of the other two quarters, belonging to the same animal, considered normal, healthy and taken as reference.


Key words: abnormal milk, milk tester, mastitis detection


One of the most widely debated issues along the elapse of time, both within food industry and in veterinary medicine, was the quality control of milk. From the chemical point of view, the problem was somehow easier to solve, but from the sanitary point of view, things appeared to be more difficult. It is well known that both clinical- and subclinical mastitis cause very important economic losses. Clinical mastitis produces losses in milk production, which must be discarded, and an overuse of medicines. Subclinical mastitis are also costly, because they are more numerous, clinical unapparent and therefore often neglected, their milk seems normal and acceptable for human consumption, but is reduced quantitatively, and has lower shelf-life and suitability for processing (2, 12). Therefore, many efforts were directed in the last half century in order to find increasingly improved methods and techniques of keeping under control subclinical mastitis. Among them, California Mastitis Test (CMT) and related techniques, Electroconductibility (EC) and Somatic Cells Counter (SCC) are now days the most widespread in practice, in spite of the fact that none of them are perfect and none of them could be taken as reference tests, because all of them give a certain percentage of false positive or negative reactions (8). But, the farmers never appreciate a technique which will determine them to discard an amount of milk by mistake diagnosticated as pathologic.

Mammary infection always leads to a lot of different physical and chemical changes in milk secretion, not simultaneously or proportionally installed, and therefore difficult to be diagnosed using a test based on a single modification, like pH, EC, SCC. That is why nowadays, beside the CMT, a number of new tests occurred, able to reveal two or three changes simultaneously, like: SCC-EC; EC-colorimetry; SCC-pH-colorimetry; SCC-pH-thermometry; SCC-EC-colorimetry and others (5, 4, 14, 16). An interesting approach to this subject was also made by Kamphuis et al. (2008) in a farm using automatic milking. Predictive value and correlation between three parameters – EC, color and production – were also studied. A new perspective concerning the mastitis control strategy has emerged in the last two decades, after Commission Directive 89/362/EEC (1989) specified that the milker is the person who must inspect the aspect of the milk and if any physical abnormality is detected, the milk must be considered abnormal and withheld from delivery (15). Let us note that the decision has nothing to do with SCC and, in this case, the status is by definition considered as clinical mastitis. According to this specification, milk can be considered normal and suitable for consummation if it does not exhibit modifications in homogeneity or color, regardless of the number of cells count (11). Therefore, lately, a certain part of researchersattention was somehow moved from the subclinical mastitis to the mild clinical mastitis, which produces abnormal milk. There also exists the opinion that between abnormal milk and SCC there is however a relationship, in the sense that abnormal milk will almost always have over then 200000 cells / mL and normal milk would have almost always less then 100000 cells / mL (11).




Materials and methods


The new test and device for the sanitary control of cow milk was conceived having as support the truism that the four samples of milk provided by the four compartments of a healthy udder, belonging to a healthy cow, must have similar physicochemical and biological characteristics. If one of the samples exhibits some differences that exceed a certain threshold, arbitrarily established, it reveals an abnormal milk and a pathological condition that should be considered either as subclinical or as clinical mastitis, depending on the magnitude of the deviation coefficient (DC) provided by the device. Summarizing, we can say that the new device, named Milktester RMM, serves to increase the macroscopic visual capacity, enabling the simultaneous recording and quantification of changes in concentration, color, and homogeneity of the milk, noticeable or unnoticeable to the direct naked eye examination.

The necessity to develop a sensor able to record and quantify changes of milk’s homogeneity and color, has been raised and forecast by many specialists, especially after Commission Directive 89/362/EEC (1989) established that the milker is the first person empowered to establish, simply by eye inspection, if a sample of milk may be considered normal or abnormal (7, 9, 15). Let’s note that, in this case, neither SCC nor other changes, visually unnoticeable, should be considered as criteria of differentiation between normal and abnormal milk.Thus, the most acute problem remained to develop a method for accurate measurements of homogeneity/consistency and color changes of milk, and then to establish the convenient threshold between normal and abnormal milk, valid for each farm and also to compare data registered for each mammary compartment with the average of the others,   belonging to the same animal, considered normal and taken as reference. In a way, this paper could be taken as an answer to the demand expressed by Hogeveen et al (2001, 2003) concerning the actual necessity of developing a new sensor able to detect and quantify milk color and consistency. Such trials, but with a different approach than ours, have already been undertaken and are cited in recent literature (8).

In order to be welcome in practice, such a device must meet several conditions, like that:

-To provide the most reliable results, preferably numerical quantifiable, closely correlated with those of the main diagnostic tests known and eventually taken as reference; -this is possible if the new device is able to detect simultaneously multiple changes that occur from the very beginning of the onset in a mammary infection;

-To be very quick and efficient; requiring very low-costs per sample;

-To be simple, requiring no particularly skilled labor;

-The tests have to be easily performed in farms with a portable device, or to be adapted to automatic milking. We assumed that, for the moment, all those conditions could be achieved most likely by optical means. It is generally known that a clinical mastitis produces abnormal milk, which is non-homogeneous, containing small gritty stuff or floaters, and changed color frequently to creamy, with changed density, more viscous or watery than normal. The subclinical mastitis produces milk with normal aspect at visual direct examination, but some of them – as it resulted from our subsequent research-, present however the same physical changes as clinical mastitis, but much more subtle, revealed only with special optical equipment. Just for this reason we conceived a device designed to indirectly measure the transmittance of a monochromatic radiation beam passing through a column of milk. The device was designed to process electronically the data obtained and display the results. Our tests showed that transmittance is influenced not only by the composition and concentration of the contained substances but also by color and suspended particles.


Results and discussions


The new device, conceived by Radu Moga Manzat and called Milktester RMM (13), currently found in the prototype stage, is an electronoptical device, designed to perceive and measure changes in composition, concentration, color and turbidity of a sample of milk crossed by a beam of monochromatic radiation. This is possible through indirect measurement of the intensity of the emerging radiations. The device is composed of: a monochromatic radiation source; a sensor; a display window; an accu-battery; a digital multimeter; a processor designed to calculate and compare the measurements and to display the result; a potentiometer; a switch; and a charger. On the cover of the device are located four buttons and four warning LEDs, one for each udder’s compartment, and one that triggers calculation and display instantly the result, as red color LED for compartment’s sample assessed as positive, and the green color LEDs corresponding to the compartments assessed as negative (normal milk). Red is assessed as positive for the sample for which the device has calculated a value that exceeds a certain threshold, arbitrarily established. It is possible for the device to detect sometimes simultaneously two positive samples, on the same animal. As appendixes, the device is accompanied by a partitioned tray for sampling, plastic rectangular vials of 3 mL and plastic pipette of 3 mL. The device is simple to use cow side, having a volume of only 15/12/5 cm and weighing less than 300g. In principle, the health status of each mammary quarter (Q) is assessed by dividing the absolute value (AV) read on screen for each Q, to the reference average value (RAV), specific for each animal. Thus, we obtain the discrimination procent (DP) wich is Q-specific, according to the equation: DP = 100×AV / RAV. Our unpublished investigations have shown that, in case of mastitis in one quarter, its AV may be greater or lesser than that of those unaffected (depending on the nature of the pathogen, the phase and form of infection). Therefore RAV is calculated for each animal according to the following procedure: the highest and the lowest values of the four read on the display for each animal are discarded and then the arithmetic average of the other two values is calculated. Te values of DP in excess of- or smaller than 100 are the deviation coefficient (DC) from RAV and serve at the assessment of each Q health state. The final evaluation is done by including PD obtained in a grid of eligibility, usually arbitrarily set somewhere between 5-15% or more, depending on the user’s requirements.


DC example of calculation                                                           Table no.1


                   Q          AV      RAV      DP        DC
       Anterior right          400        405        99        1
       Posterior right          455        405      112      12
       Anterior left          390        405        96        4
       Posterior left          410        405      101        1


Our preliminary tests of the device showed that a DC of up to 5% (5 DC) should not be considered because they are inherent and mainly due to inevitable biological differences among Q of the same animal. The more advanced is the infection, the more distant will be DP from 100, either over or under, and the higher will be the DC value. In the example above, only the right posterior Q has exceeded both the minimum compulsory threshold (5%) and even that arbitrarily fixed at 10%. The outcomes obtained by us in a field trial, in two different dairy farms, on 488 quarters are presented in table no.2.






The distribution on thresholds groups                                         Table no.2


     DC interval 0-5 0-10 0-15 15,1-25 25,1-50 50,1-100 >100
No q   reacted 404 450 474 9 2 2 1
% q     reacted 87,79 92,21 97,13 1,84 0,41 0,41 0,20


Argherie (2008), in her Academic Dissertation analyzed the correlation coefficient between the results obtained with Milktester RMM and other five tests. The highest correlation coefficient calculated, using Correl program, was 0.85 between TNCS and %Ne (Table 3) but a good correlation was also obtained between Milktester and %Ne (0.79), between Milktester and TNCS (070) and between Milktester and EC (0.70)


Correlation coefficient between Milktester RMM and other 5 tests

Table no.3


       TEST      C M T      E C M R M M     % Ne   T N C S     B E


       C M T          x      0.46      0.63      0.46      0.39     0.39


       E C      0.46        x      0.70            0.64      0.59     0.40


       M R M M      0.63      0.70        x      0.79      0.70     0.56


         % Ne      0.46      0.64      0.79        x      0.85     0.76


       T N C S      0.39      0.59      0.70      0.85        x     0.70


       B E     0.39      0.40      0.56      0.76      0.70      x


Legend: CMT-California Mastitis Test; EC-electroconductivity; MRMM-Milktester RMM; %Ne- percentage of neutrophils; TNCS- total no. of somatic cells;   BE-bacteriological examination.

Let us note that all measurements and calculations, including the final result, are accomplished electronically, in less than 8-10 seconds/cow, excluding the time of sampling. Collection and examination of samples were dan immediately before milking.

After the last Q recording, the R button (Result) is pressed and the device instantly displays the final result by staining in red the LED corresponding to the Q affected, who’s DC has exceeded the choused threshold. The other three LEDs turn green. The thresholds delineating a normal milk, supplied by a healthy udder from an apparently normal milk supplied by a subclinical mastitis may be fixed at 5%, and the threshold delineating an abnormal milk supplied by a clinical mastitis, are relative values, which could be arbitrarily set between 5-15 % or higher, and modified according with manufacturer’s requirement . It should be noted, however, that a high threshold will cause a high specificity but with lower sensitivity and vice versa (6) DC of an affected quarter is influenced by many factors- such as milk chemical composition, flakes and clots, watery appearance, SCC and color-, some of them upwards, others of them downwards. DC values are the result of simultaneously influence of all these parameters but, of course, some of them acting in opposite ways, could cancel each other out. A DC above the threshold chosen at 5% indicates, depending on the magnitude of DC, either: 1) a subclinical mastitis, with milk apparently normal -at regular visual examination-, but with noticeable changes in homogeneity and color, when is examined by the device or 2) a clinical mastitis, with abnormal milk –as a rule above 10%. In both cases there are an increased number of somatic cells in the milk, which will have a negative effect on production, the purchase price, the shelf-life and quality of both the milk and processed products (2, 12) Therefore, both farmers and processors would need to be economically very interested in reducing to minimum the SCC in cooling tank by temporarily removing the milk from cows with large numbers of SCC – respectively high DC. There are cases when DC> 5 are recorded in two or, rarely, even in three Q at the same animal, but always at different values, because the infection is usually installed sequentially not simultaneously, and therefore also inflammatory processes will be found in different phases. Let us note that Milktester RMM will never confuse mastitis milk with colostrums, milk preceding weaning or milk from postpartum mammary edema, because in all these three situations, unlike in mastitis milk, AV appears changed in all four Q, in the same sense, at very close values. We believe that the next investigations will have to refer to:

-Determination of optimal intervals between tests.

-Comparing the results of the Milktester to those obtained by other methods.

-DC Limits (valid for Milktester RMM) recommended for milk samples classification in one of the categories: 1) normal milk – healthy mamma, 2) abnormal milk – mamma apparently normal (subclinical mastitis), 3) abnormal milk – mamma acute (mild, sever) or chronic inflammation (clinical mastitis).

-The possibility to adapt Milktester RMM in automatic milking systems.

-The possibility to interblend the results obtained on the same sample, simultaneously, by the optical device (Milktester RMM), EC, pH or others.

-The possibility to use the device for other purposes, such as to detect counterfeit milk or to assess the quality of milk from the cooling tank, compared to a standard sample.




  The conception of the device is based on the assumption that the physical, chemical and biological characteristics of normal milk, taken from the four mammary quarters of a healthy cow, usually are similar. An Ssignificant change of some parameters, over a certain threshold in a quarter, by comparison to the others, considered normal and taken as reference, shows an abnormal state of the milk in that quarter. Milktester RMM is designed to detect and instantly display any such difference occurred in a quarter’s milk transmittance, when is crossed through by a monochromatic beam of electromagnetic radiation. The magnitude of the transmittance is influenced by composition, concentration, color and homogeneity of the crossed sample of milk. So, Milktester RMM works as if three different devices –a luxmeter, a colorimeter and a turbidimeter (or nephelometer)- were incorporated into one, to provide a single result for every quarter. The device calculates and displays, also instantly, the final result for the cow. The threshold is arbitrarily chosen, depending of the level of the beneficiary’s exigency regarding desired sensibility and specificity. Like the EC, Milktester RMM is extremely fast, portable and does not consume reactive but, unlike the EC, which is influenced by a single parameter – the concentration of electrolytes – Milktester RMM is influenced by several parameters. The device was conceived to replace the naked eye assessment of the milk, as it is made now by milker – which could be subjective, superficial and random-, by a more accurate and objective assessment, possibly standardized, made cow side periodically by an electronic optical instrument. The next investigations must establish the optimal threshold for differentiation between normal and abnormal milk, milk tester’s efficency and utility in evaluating the quality of mixed milk in containers or cooling tanks and, finally, the chances of adapting the Milktester RMM to the automatic milking sistems on line. Of course, the test and device proposed are susceptible of improvements based on further research.


Conflict of interes statement: “None of the authors of this paper has a financial or personal relationship with other people or organizations that could inappropriately influence or bias the content of the paper”



I am grateful to Mr. Lucian Căprărescu Dipl. Eng, for his technical support, offered with competence and skillfulness, in the building of this equipment.





    1. Argherie Diana Ccytological research in cow’s mastitis. Academic Dissertation PhD. Faculty of Veterinary Medicine. USAMV Timisoara. Romania, 2008
    2. Barbano D. M., Ma Y., Santos M. V. Influence of Rau Milk Quality on Fluid Milk Shelf Life. Journal of Dairy Science. 2006, 89: E15-E19, 1-10
    3. Borecki M., Szmidt M., Korwin Pawlowski M., Beblowska M., Niemice P., Wrzoseck P. A method for testing the quality of milk using optical capillaries. Photonic Letters of Poland. 2009, Vol. 1 (1), .37-39
    4. Duglas J. R., Jason M.H. Online Milk Sensing Issues for Automatic Milking. ASAE/CSAE Annual International Meeting 1-4 August 2004. No. 04-4191, 1-9
    5. Hillerton J.E. Detecting Mastitis Cow-side. National Mastitis Council. Annual Meeting .Proceedings, 2000, 48-53
    6. Hogeveen H., Van der Vorst Yvone., Ouweltjes W., Betsie A., Slaghuis B. A. Automatic miilking and quality: an european perspective. National Mastitis Council Annual Meeting. Proceedings 2001, 152-162
    7. Hogeveen H., Ouweltjes W. Sensors and management support in high technology milking. Journal of Animal Science.2003, 81 Suppl. 1-10
    8. Hovinen Mari Udder health of dairy cows in automatic milking. Academic Dissertation. Faculty of Veterinary Medicine. Helsinki 2009
    9. Kamphuis Claudia., Pietersma D., Van der Tol R., Wiedemann M., Hogeven H. Using Sensor data patterns from an automatic milking system to develop predictive variables for classifying clinical mastitis and abnormal milk. Computers and Electronics in Agriculture. 2008, Vol. 62 no.2, 169-181
    10. Lam T.J.G.M., Olde Riekerink R.G.M., Sampimon O.C., Smith H. Mastitis diagnostics and performance monitoring: a practical approach. Irish Veterinary Journal. 2009, Vol.62 Supplement 34-39
    11. Larry Smith K., Hillerton J.E., Harmon J.R. Guidelines on Normal and Abnormal Raw Milk Based on Somatic Cell Counts and Signs of Clinical Mastitis. National Mastitis Council. Approved by the NMC Board of Directors, February 2001.
    12. Ma Y., Ryan C., Barbano D.M., Galton D.M., Rudan M.A., Boor K.J. Effects of Somatic Cell Count on Quality and Shelf-Life of Pasteurized Fluid Milk. Journal of Dairy Science. 2000, Vol. 83 no.2, 264-274
  • Moga Manzat R. Detectarea mamitelor cow side cu ajutoru unui aparat portabil. (Portable Device for Cow Side Detection of Mastitis). Rev. Rom.Med.Vet.2010, no.4, 77-89


  1. Olde Riekerink R. G. M., Barkema H. V., Veenstra W., Berg F. E., Strihn H., Zadoks R. N. Somatic Cell Count During and Between Milking. Journal of Dairy Science. 2007, 90, 3733-3741
  2. Rasmussen M.D. Definition of normal and abnormal milk at time of milking. Internal report no.169 for Workshop of the EU-project (QLK-2000-31006). Implications of the introductions of automatic milking on dairy farms. November 27, 2002, 102
  3. Viguier Caroline., Arora S., Gilmartin N., Webeck Katherine., Kennedy R. Mastitis detection: current trends and future perspectives. Trends in Biotechnology. 2009, Vol 27 no.8, 486-493

Posted in Profesional-stiintific and tagged , by