Method for detecting oestrus behaviour of a milking animal

Field of the invention

The present invention is related to the field of milking animals and in particular to a method for detecting oestrus behaviour of a milking animal as defined in the introductory part of claim 1.

Background of the invention

Artificial insemination is used within the dairy industry in order to propagate desirable characteristics of a particular bull to many females or to overcome breeding problems. Artificial insemination thus allows for many cows, as well as other milking animals, to be impregnated with the sperm of a bull having traits for improved milk production. However, a prerequisite for successful artificial insemination is that the farmer is able to accurately estimate the time of ovulation .

Many Asian countries depend on the water buffalo as their primary cattle species and domestication of the water buffalo is prominent in many of these countries. The fat content of buffalo milk is the highest amongst farm animals and the water buffalo is therefore extensively used as a milking animal. Thus taking the water buffalo as an example, the female water buffaloes show week signs of heat compared to, for example, cows. It is therefore particularly difficult to estimate the time of ovulation. Even the buffalo bull can have difficulties in precisely detecting when the female buffalo ovulates. The high frequency of silent heat in the water buffaloes is a major obstacle for the successful introduction of artificial insemination.

However, the oestrus behaviour may be difficult to detect among other milking animals as well. Cows are often held

tied-up and in such systems the heat detection is more difficult due to the limitation for the cows to express their natural behaviour. Detection of oestrus behaviour in a tied- up environment is therefore difficult.

There are known methods for detecting oestrus behaviour among dairy cows. For example, activity monitoring can be used as a means for determining whether a cow is in heat or not, since she becomes more restless and has an increased activity level when being in heat. An increase in the number of steps taken by the cow can therefore be taken as an indication of heat.

Although the activity monitoring is fairly reliable for detecting oestrus behaviour, it is still possible to miss the oestrus behaviour of the milking animal. The oestrus behaviour can be influenced by different factors, such as the season of the year and temperature during day and night, whereby for example a cow generally shows oestrus behaviour more clearly at cold weather. Another example of a factor influencing the oestrus behaviour is the housing conditions, which can also have an effect on the distribution of heat during a 24 hour period. Further yet, some milking animals, for example water buffaloes, have different levels of activity during the day and the night, which makes it even more difficult to detect when they are in heat.

Undetected heat lengthens the calving interval with several weeks and milking animals may be culled for infertility without due cause. Further, missed heat also entails costs for the farmer.

In view of the above it would be desirable to provide means for determining when a milking animal is in heat.

Summary of the invention

It is an object of the invention to provide a reliable method and means for determining when a milking animal is in heat, while overcoming or at least alleviating the shortcomings of the prior art.

It is another object of the invention to provide a method for a farmer to determine the optimum time for insemination or mounting by a bull in order for the farmer to plan his breeding in an optimized way.

These objects, among others, are achieved by methods for detecting oestrus behaviour of a milking animal and by computer program products as claimed in the appended claims.

In accordance with the invention a method for detecting oestrus behaviour of a milking animal is provided. The milking animal is provided with a sensor means for detecting her activity level. The method comprises the steps of: monitoring an activity level increase of the milking animal by the sensor means; and detecting the oestrus behaviour in dependence on a set threshold activity level increase, wherein the threshold activity level increase is indicative of an oestrus behaviour of the milking animal and set m dependence on time of day. The invention provides a reliable means for determining when a milking animal is in heat. The increased reliability is accomplished by utilizing adjustable activity level increase indicative of oestrus behaviour of the milking animal. The farmer is thereby provided with most efficient means for determining when a milking animal is in heat and the farmer is able to optimize his breeding program. This leads to lower breeding costs for example owing to fewer services per conception. The invention also increases conceptions on the milking animals with silent heats. This method is thus suitable for milking animals having different oestrus behaviour depending on time of day, for example

buffaloes, as their oestrus behaviour differs during day and night .

In accordance with another embodiment of the invention, the activity monitoring is combined with other data about the milking animal. The activity monitoring could for example be combined with one or more of: the frequency of visits to automatic out-of-parlour feeders or other barn devices equipped with readers for recording animal ID (e.g. water bowls or roughage feed mangers) , information from visual monitoring systems, data about time from last calving, latest date of manually observed heat, date of confirmed pregnancy, and data from the measuring of progesterone levels in milk. A further yet increased reliability in detecting time of ovulation of the milking animal is thereby obtained.

In accordance with still another embodiment of the invention, an alarm is started in order to alert a user upon the detection of oestrus behaviour of the milking animal. Such alarm could for example be an audible alarm or a visual alarm.

Further embodiments of the invention are defined m the dependent claims, and will also be evident from the detailed description .

In accordance with a further aspect of the invention computer program products are provided, whereby advantages similar to the above are achieved. In particular, the invention may thereby easily be implemented in already existing milking systems .

Further characteristics of the invention and advantages thereof will be evident from the detailed description of embodiments of the present invention given hereinafter and

the accompanying figures, which are only given by way of illustration .

Brief description of the drawings

Figure 1 illustrates schematically differences in activity level increase of a milking animal during night and day.

Figure 2 is a flow chart illustrating steps of the present invention .

Detailed description of embodiments

Generally there is a predetermined activity level increase set to indicate an oestrus behaviour of a milking animal. In particular, the activity of the milking animal is monitored over time and a certain increase in the activity level is set to indicate her oesterus behaviour. However, as mentioned in the introductory part, there are different factors affecting the oestrus behaviour of milking animals and it is often difficult to detect when they are in heat.

In accordance with the present invention, a milking animal's activity level increase is made adjustable regarding heat detection. That is, the activity level increase that is indicative of a milking animal being in heat is set differently on an individual basis. In particular, the activity level increase is set in dependence on time of day and can differ over a 24-hour period.

For example, a particular first milking animal may show her oestrus behaviour (as measured by an activity increase) most clearly during morning, while a second milking animal may show her oestrus behaviour most obviously between noon and 6 p.m. Further, the first milking animal may have small differences in activity level when in heat compared to when not in heat, while the second milking animal may have great

differences in activity level when in heat and when not in heat. In short, a certain activity level increase of the first milking animal may be indicative for her being in heat, while the same activity level increase for the second milking animal need not be an indication of her being in heat.

More specifically, if the threshold activity level increase is set egual for all milking animals over a 24-hour period, which is common practice today, then either a false oestrus behaviour can be detected for one milking animal or an actual oestrus behaviour can be missed for another milking animal.

In accordance with the invention, the knowledge of each individual animal is used for setting the most appropriate activity level increase indicative for the oestrus behaviour of the particular milking animal. In the following the adjustable and individually set activity level increase is denoted δL _TO D-

In the following the water buffalo is used as an example for illustrating an aspect of the present invention. The water buffaloes are not as active during daytime as during nighttime when it comes to oestrus behaviour, which makes it difficult to detect when they are in heat. In particular, it is difficult to set a suitable activity level increase that should be indicative for the water buffalo coming into heat. A certain daytime activity level increase of the water buffalo may be indicative for her being in heat, while the same activity level increase at nighttime need not be an indication of her being in heat.

An adjustable sensitivity for the water buffalo's activity level increase regarding heat detection is therefore used m accordance with the present invention. That is, the activity level increase δL _TOD that is indicative of a water buffalo in heat is set differently depending on time of day, and

especially differently during day and night. During daytime the activity level increase that is indicative for an oestrus behaviour, δL _ToD , is smaller than during nighttime. A higher sensitivity for such heat related activity level increase is thus required during daytime compared to nighttime.

A certain daytime activity level increase is used as a threshold value for determining, during daytime, when the water buffalo is in heat. If the daytime activity level increase is measured during daytime, then it is determined that she is in heat.

Likewise, a nighttime activity level increase, which is generally different than the daytime activity level increase, is used as a threshold value for determining, during nighttime, when the water buffalo is in heat. If the nighttime activity level increase is measured during nighttime, then it is determined that she is in heat.

The time dependent activity level increases are set in relation to a normal activity level during day and night, respectively. The normal activity level may be set as an averaged activity level of the water buffalo when she is not in heat. As a water buffalo is generally more active during daytime she has a higher activity level during the day even if she is not in heat compared to the activity level during night .

The activity level increase δL _ToD set to indicate oestrus behaviour is thus an increase in activity level compared to an activity level when she is not in heat. The increase in daytime activity level, wherein the increase is related to her being in heat, is generally lower than the increase m nighttime activity level when the water buffalo is in heat. Equivalently, the sensitivity for an activity increase is higher during daytime than during nighttime.

Figure 1 illustrates the above differences in activity level during night and day. It is pointed out that the indicated levels are used only to clarify and explain the differences in activity level during daytime and nighttime and are not actual values or relationships. On the left hand side the nighttime activity level L _N of a water buffalo is illustrated. Nighttime could for example be between the last milking occasion of a day and the first milking occasion of the next day, or between specific times, for example between 21.00 and 06.00. On the right hand side the daytime activity level L _D of a water buffalo is illustrated. Daytime is then the remaining part of a 24-hour period. For both the nighttime and daytime parts of the figure, the normal activity level, L _N and L _D , respectively, as well as the activity level increase indicative for when the water buffalo is in heat, δL _N and δL _D , respectively are illustrated. That is, δL _ToD is set to δL _N during nighttime and to δL _D during daytime .

As can be realized by studying figure 1, it would be difficult to detect oestrus behaviour of the buffalo if the activity level increases, δL _D and δL _N , indicating the oestrus behaviour were identical during daytime and nighttime. If δL _D was set equal to δL _N at a level somewhere between δL _N and δL _N , then either a false oestrus behaviour could be detected during night or an actual oestrus behaviour could be missed during daytime.

In accordance with the invention, the increase in activity level used for determining oestrus behaviour can therefore be adjustable in dependence on the time of day. In general, δL _N > δL _D but there may be individual differences among the water buffaloes. The activity level increase may therefore be set on an individual basis suitable for the specific buffalo in question. The times during which the different activity level

increases are to be used may also differ between the water buffaloes and could also be taken into account and be set accordingly.

In the above description, although describing the differences in activity level increases during day and night for water buffaloes, there are other milking animals amongst which the oestrus behaviour differs depending on time of day. A study performed by the applicant of the present invention shows that cows housed in a tied-up system have only a small increase in activity level when in heat. In particular, the limited space available for the milking animal affects her activity level, both her "normal" behaviour and her high and low levels. Therefore, for a tied-up milking animal the ratio between normal and high activity level decreases compared to milking animals in loose housing. Events in the stall during daytime, such as the activation of a feed manger, can therefore render the detection of oestrus behaviour of the milking animal difficult. That is, the activity level increase among the milking animals caused by events in the stall may wrongly be interpreted as oestrus behaviour. The environmental influence on the cow' s behaviour is the lowest at night, and her oestrus behaviour may therefore, like for the buffalos, be detected more easily at night hours.

There are different sensor means available for monitoring the activity level of the milking animal. A pedometer counts the number of steps taken by the animal and such a pedometer could be used in the present invention as a sensor device for determining the activity level of the milking animal. Another example of a sensor device that could be used is an accelerometer, for example mounted to the neck or leg of the milking animal. An accelerometer is similar to a pedometer, but instead of counting steps, it records activity counts per time unit (e.g. per minute) and indicates the intensity of

physical activity and picks up small movements that a pedometer would miss. Any sensor means that is able to register the milking animal's movement pattern could be used in the present invention.

The sensor device should preferably be arranged to monitor the activity level continuously, although a sensor device intermittently registering activity levels could be used. The sensor device, for example an activity meter, could be arranged to store activity data in an internal memory. The information registered by the sensor devices could then be forwarded to one or more information receiving units, preferably at least on an hourly basis and more preferably even more frequently. The frequency with which the information is transferred can be different during the day and during the night or different depending on other times of a 24-hour period. The forwarding of the information from the sensor devices could be made by wireless transmission or by wired transmission. Alternatively, the farmer could make manual readings of the information directly from the sensor device.

Data collected from the activity monitoring could be combined and correlated with other data about the milking animal. For example, the activity monitoring could be combined with the frequency of visits to automatic out-of-parlour or other barn devices equipped with readers for recording the ID of the animal, such as water bowls or roughage feed mangers. Examples of other data that could be used for increasing the reliability of detection of oestrus behaviour of the milking animal include: information from visual monitoring systems, e.g. cameras, data about time from last calving, latest date of manually observed heat, date of confirmed pregnancy, data from the measuring of progesterone levels in milk etc. As yet another example, riding of other milking animals can be a

sign of heat but all riders are not necessarily in heat. Therefore, the most positive sign of heat for a particular animal may be when she frequently rides on or is ridden by other milking animals. Such behaviour can be visually detected or by means of heat mount detectors and such data can be combined with the time of day dependent threshold activity level increase δL _ToD .

In an embodiment of the invention, the activity level is monitored only during night hours. This can in some cases provide a more reliable detection of oestrus behaviour, for example in the case mentioned earlier, for milking animals held in a tied-up environment.

In accordance with the invention, events occurring in the stall and affecting the activity level of the milking animal can be taken into account by disregarding activity levels registered during particular time periods. For example, the particular hours during which the milking animals are fed may be disregarded. The events in the stall may affect her behaviour and also her activity level. This omission of activity increase registrations during certain hours accounts for activity increases that are not caused by oestrus behaviour, but by events in the stall.

The data could be combined in any suitable way and treated by appropriate software and algorithms as basis for obtaining a most efficient tool for alerting the milking animal herd manager or producer when his/her milking animals are in heat and thereby the optimum time of breeding.

Further, the individually set threshold activity level increase δL _ToD indicating oestrus behaviour can be updated after a detected and confirmed heat period, thereby further increasing the accuracy of the set threshold activity level increase δL _ToD . That is, the data used for determining when

the particular animal is in heat can be evaluated and the threshold activity level increase δL _ToD can be adjusted in dependence thereon. For example, if such evaluation shows that her detected activity level increase is much above the set threshold activity level increase δL _TOD above which it is determined that she is in heat, then the set threshold activity level δL _ToD indicative for the oestrus behaviour could be increased. Likewise, if her detected activity level increase is close to the set threshold activity level increase δL _ToD indicative of oestrus behaviour, then the set threshold activity level δL _ToD could be lowered.

Figure 2 illustrates steps of the method in accordance with the present invention. Further steps, not shown in the figure, could be included. In the method 10 for detecting oestrus behaviour of a milking animal suitable sensor means is used for detecting the activity level of the milking animal. In step 11, the activity level of the milking animal is monitored by the sensor means. Thereafter, in step 12, the oestrus behaviour is detected in dependence on a threshold activity level increase δL _ToD . The set threshold activity level increase δL _A is indicative of an oestrus behaviour of the milking animal. The threshold activity level increase δL _ToD is set in dependence on data for the particular milking animal and in particular in dependence on time of day. That is, when her detected activity level increase reaches the set threshold activity level increase δL _TOD , it is established that she is in heat.

As mentioned, the method 10 can include further steps. For example, the step of monitoring the activity level of the milking animal could comprise updating the activity level at least once an hour. The method 10 preferably comprises the additional step of forwarding, in a suitable manner, information from the sensor means to an information-receiving

unit. The information-receiving unit could comprise a processing unit, such as a computer. The data obtained from the sensor means can then be processed in the computer. The data could, as mentioned earlier, be combined with other data about the milking animal to thereby increase the reliability of the detection. Still another example of an additional step includes updating the threshold activity level increase δL _ToD in dependence on new data about the milking animal. Yet another example of an additional step includes alerting a farmer about the oestrus behaviour upon the detection thereof. Such alarm could be an audible alarm or a visual alarm in the form of a light device or the like.

In another embodiment of the method in accordance with the invention for detecting oestrus behaviour of a milking animal, the milking animal is identified. The identification can be made m any known manner, for example manually, or by having the milking animal wear an identification tag or other identification means, such as a transponder unit. The identification should then be included in the information transferred to the receiving unit.

The activity level increase δL _ToD indicative of an oestrus behaviour may be set so as to have a higher sensitivity for an activity increase during daytime than during nighttime. This is particularly suitable for buffaloes, as their oestrus behaviour differs during day and night, and for cows being held in a tied-up environment.

Thus, rather than setting the threshold activity level increase on an individual basis, the threshold values are set for a group of animals, or a particular species of milking animals (e.g. water buffaloes or camels).

It is noted than more than two threshold activity level increases can be set, for example one activity level increase

for 06.01 am to 2.00 p.m., a second activity level increase for 2.01 p.m. to 10.00 p.m. and a third activity level increase for 10.01 p.m. to 06.00 a.m. Four or even further threshold activity level increases can alternatively be set.

The invention is also related to a computer program product that is loadable into an internal memory of a computer. The computer is used for monitoring and/or processing the data obtained from the activity sensor means carried by the milking animals. The computer can be arranged to receive data from the sensor means directly or via antennas or other receiving units. The computer program product comprises software code portions for carrying out the method described above, when the computer program product is run on the computer. That is, the computer program product comprises software code portions for setting the threshold activity values individually for each animal based on suitable data for the particular milking animal, for monitoring the activity level and for detecting the oestrus behaviour of the milking animal. The software code portions can for example perform the task of comparing the normal activity level to activity levels received from the sensor device. It can then be determined when the threshold activity level increase δL _ToD indicative of the oestrus behaviour is reached. The invention is thereby easily implemented in already existing water buffalo milking systems.

The computer program product may alternatively be stored on a computer readable storage medium, for example a compact disc. The computer readable storage medium comprises computer readable program code means for causing the computer to carry out the method described above.

All milking animals having an oestrus behaviour that differs in activity level in dependence on time of day, which in turn

may depend on different factors, may benefit from the present invention. In the above description, the water buffalo is used as an example of cattle, for which the present invention is advantageous. However, the invention is applicable also for other members of the family of Bovidae, for example other buffaloes, such as the African buffaloes, or for bison. Further examples of milking animals include cows, goats, sheep and camels. The invention can be implemented in connection with yet other, not herein mentioned milking animals.