System and method for tracking animals

Abstract

The present disclosure relates to an animal monitoring system (100) for an animal cage, the animal monitoring system (100) comprising: a physical platform divided into a grid of cells (106) wherein each cell (106) comprises at least one weighing sensor (108); an animal identification system; a processing unit configured to: identify at least one animal on the physical platform using the animal identification system; continuously obtaining a load signal from the load cell (108); the identified position and / or change in position of the at least one animal on the physical platform is tracked based on the load signal from the weighing sensor (108). The present disclosure also relates to a method of investigating an animal in an animal cage.

Description

TECHNICAL FIELD

[0001] The present invention relates to an animal monitoring system for animal cages. The present invention also relates to a method for animal monitoring. BACKGROUND

[0002] Rodents are of great importance in medical and pharmaceutical research as their metabolism is similar to that of human subjects and because they are susceptible to many conditions and infections that afflict humans. Research areas that use rodents include genetics, developmental biology, cell biology, oncology and immunology. As a result, a large amount of time is spent by medical and pharmaceutical researchers in tracking, identifying, handling rodents and extracting data from them.

[0003] Prior art for handling rodents includes tracking using, for example, microchips (an invasive method) or infrared tracking methods. Research has shown that animals can react to the presence of humans, which can introduce experimental errors and affect scientific results. Often, the lack of reliable digital devices for tracking and managing rodents during animal testing slows scientific progress and negatively impacts experimental data quality. Another consequence of poor animal management is that more animals than needed will be required for a particular experiment.

[0004] Therefore, there is a need for an improved system for tracking rodents. SUMMARY

[0005] It is an object of the present invention to provide a better solution for researchers handling rodents or other animals for scientific experiments. In particular, the present invention relates to an animal monitoring system for an animal cage, wherein the animal monitoring system comprises a physical platform, which platform is divided into a grid of cells, the cells being defined such that each cell comprises at least one load cell. The platform can be adapted to be placed or integrated in a European type 3 animal cage for mice. However, the animal monitoring system can be placed in any type of cage and used for any animal. The animal monitoring system can be a system that is fitted into an existing cage, but can alternatively be integrated into the cage when the cage is built or assembled. The animal monitoring system can further comprise an animal identification system. The animal identification system can for example be a radio frequency identification (RFID) based and / or a camera based identification system. The animal monitoring system further comprises a processing unit configured to: identify at least one animal on the physical platform using the animal identification system; continuously obtain load signals from the load cells; and track a position and / or a position change of the identified at least one animal based on the load signals from the load cells. By using the load signals from the load cells and a supplementary system for identifying individual animals, the load cell system is able to distinguish between animals and track the animals in the cage. In addition to allowing the system to track the animals over time, the load cell data itself can be used for further data processing.

[0006] The researchers will thereby be able to monitor a plurality of animals using such cages equipped with a monitoring system. The system will typically first measure and record the weight of each rodent entering the grid. For example, the weight of each animal can be individually weighed on a scale. Alternatively, the weight of the animals can be weighed on the monitoring system. The reference value can then be continuously compared with new input values for each load cell in case the animals move. The above process is implemented by using computer software that processes the data regarding each cage.

[0007] The load cells can measure the weight of the rodents by various methods, such as springs underneath each load cell, or voltage fluctuations in the load cells due to strain when the animals are placed on the load cells. The load cells should be interpreted broadly to include any suitable weight sensor that can be used in the monitoring system. The load cells will typically convert a force, such as tension, compression, pressure, or torque, into a measurable and standardized electrical signal. It can comprise a force transducer. As the force applied to the load cell increases, the electrical signal will also change proportionally. The load cells can be, but are not limited to, resistance-based load cells, such as piezoelectric load cells and strain gauge load cells. Piezoelectric load cells and strain gauge load cells are built on similar principles. When the shape of a strain gauge changes, its electrical resistance changes. In a piezoelectric load cell, the voltage output is also proportional to the deformation of the load cell.

[0008] The animal monitoring system disclosed can be able to extract changes in activity of the animals over time by continuously measuring the weight on each load cell and converting changes in weight to movement of the animals over the grid. Furthermore, by utilizing the grid of plates (typically thin plates) covering the cells, it is possible to estimate the weight distribution of multiple animals. The thin plates allow the weight of an animal placed on a first load cell to be distributed to other load cells, e.g. adjacent load cells. It is possible to observe not only the difference in load in the cell where the animal is placed, providing a more detailed view of where the animal is located. This can be used, for example, to monitor the activity of the animal when staying on the same load cell, e.g. shivering, turning around or two-legged standing. Furthermore, it is also possible to estimate, for example, the distance moved by the animal within a given time interval or the frequency of movement between different cells based on the data from the load cells. This information can further be used to extract and / or classify expected or unexpected behavior of the animals.

[0009] By calculating the activity of the animals over time, it is possible to extract the resting time of the animals. The movement pattern can also be used to calculate if the animals are moving material over the grid. It can also provide information on how the behavior of the animals is when in contact with other animals compared to when alone. All of the above information can be extracted by continuously measuring the weight of each load cell, identifying each animal and tracking its movement within the cage.

[0010] The monitoring system can be adapted to fit into a EU type 3 animal cage for mice. For example, the monitoring system can have the dimensions: length 375-400 mm, width 215-240 mm. The grid of cells can be oriented diagonally with respect to the generally rectangular shape and / or can have a total difference in orientation of at least 10° with respect to the generally rectangular shape.

[0011] The present invention also relates to a method for investigating a plurality of animals in an animal cage. By scaling up the cage and changing the load capacity and sensitivity of the load cells, the principle of the load cells can be extended to any type of animals that can need to be monitored.

[0012] In summary, the present invention allows researchers to continuously collect data on the weight and behavior of animals, in particular rodents, in an efficient way. With the disclosure above, it is possible to collect more data accurately while spending less time and using fewer animals. At the same time, there is no need for animal marking of each animal and no need for manual weighing of the weight of each animal, which would otherwise be an invasive and time-consuming task, respectively. BRIEF DESCRIPTION OF DRAWINGS

[0013] Various embodiments will be described below with reference to the drawings. The drawings are examples of embodiments and are intended to illustrate some features of the animal monitoring system disclosed and do not limit the method and system of the present disclosure.

[0014] Figure 1A - B shows a schematic top view and a side view of a monitoring system, respectively; Figure 2A - B shows a monitoring system with a grid comprising a plurality of load cells; Figure 3A - B shows an embodiment of a monitoring system with a grid comprising a plurality of load cells and RFID interrogation units; Figure 4 An example of a monitoring system with a diagonally oriented grid pattern is shown; Figure 5 An example of a flowchart of a method of investigating an animal is shown; Figure 6 An example of extracting the movement of an animal over time by continuously measuring the signal given by the load cells is described; Figure 7 An embodiment of a monitoring system equipped with a software interface is shown; Figure 8 A perspective exploded view of a monitoring system is shown; Figure 9 An example of a schematic view of a layer in which a load cell can be placed is shown; Figure 10 An example of a weight distribution element arranged below a top plate is shown; Figure 11 A cross-sectional view of an embodiment of the animal monitoring system of the present disclosure is shown; Figure 12A - B shows two examples of animal monitoring using a grid of modeled units. DETAILED DESCRIPTION

[0015] The present disclosure relates to an animal monitoring system for an animal cage. Figure 1A A conceptual example of such an animal monitoring system 100 is shown. Figure 2A An example of a grid of units 106 is shown, wherein each unit 106 comprises at least one load cell 108. Figure 2BAn example of a side view of a load cell 108 is shown. In the particular embodiment of Fig. 2, the animal monitoring system 100 comprises a physical platform that is divided into a grid of cells 106. In this example, there are 3x3 cells. A "grid of cells" is to be interpreted as including any suitable division of an area into smaller sub-areas, the sub-areas being the "cells". Typically, but not necessarily, the cells will have a square, rectangular or diamond shape. However, it is also possible to have cells with irregular and / or circular shape.

[0016] The load cells can be configured to be coupled with load cell amplifiers and analog-to-digital converters. The skilled person will understand how to implement such components, which can be part of the load cells themselves, or separate components. The load cell 108 in the example of Fig. 2b has a load cell amplifier 121 and an analog-to-digital converter 122. The load cells can be calibrated so that they respond correctly to weight on the animal monitoring system. This can take into account e.g. variations in load cell installation precision and characteristics of the top plate to transmit force to the individual load cells.

[0017] Preferably, the animal monitoring system comprises an animal identification system. This animal identification system can be e.g. a radio frequency identification based and / or a camera based identification system. In a camera based identification system, one or more cameras can be strategically installed within the cage. The cameras can be configured to continuously capture images, or to capture images upon e.g. a certain load cell being triggered. There are many ways of animal identification. For example, computer vision algorithms can be used to extract features from images or videos. Animals can also be marked, e.g. by adding color or other visible markings. Also trained machine learning models, such as convolutional neural networks, can be used to identify animals in the cage.

[0018] Another option is to use an RFID based animal identification system. In such a system, each animal can be marked with a small RFID transponder. The RFID tag contains a unique identification code that can be read by an RFID interrogation unit, which can also be referred to as an RFID reader. Thus, the animal identification system can comprise one or more radio frequency identification interrogation units. The RFID interrogation units can be strategically installed within, around or below the cage. When an animal moves around the cage, the RFID interrogation units will detect whether an RFID tag is present within a certain range. Other similar concept technologies, such as near field communication (NFC), will also be possible to use. In the context of the present invention, NFC can be considered a subset of RFID.

[0019] As previously mentioned, the animal identification system comprises one or more radio frequency identification interrogation units. Preferably, the animal identification system comprises a plurality of radio frequency identification interrogation units distributed over the physical platform. The radio frequency identification interrogation units can be arranged at strategic places within the cage. The RFID readers can be placed below the physical platform. Figure 3A and 3B An example of the distribution of the load cells and the RFID interrogators is shown. In the example there are three RFID interrogators 126. Each unit also has at least one load cell, in the example each unit 106 has four load cells 108. Figure 3B A side view of the animal identification system is shown, showing that both the load cells 108 and the RFID readers 126 can be arranged below the surface on which the animals are located.

[0020] In one embodiment, the activation of a predefined load cell triggers the identification of an animal using the associated predefined radio frequency identification interrogation unit. This means that, starting from the situation that the system does not know the location of one or more animals and thus cannot track any animal, the system can obtain information that serves as a starting point for further tracking and observation using the load cells. For example, if a given RFID reader with a given location in the grid identifies an animal with an RFID tag, this location can be used as a starting point. From this starting point, the load cells can subsequently track the identified animal.

[0021] The placement and use of RFID interrogators to track animals and observe their behavior generally provides a relatively poor resolution. By first using the identification system to obtain a preliminary identification, and then using the load cells for actual tracking and observation, more detailed information can be provided. Preferably, the animal monitoring system comprises more load cells than the number of radio frequency identification interrogation units.

[0022] Regardless of whether the animal identification system is an RFID-based system, a camera-based system, or other system, the animal identification system provides a general location of at least one animal on the physical platform. The load signals from the load cells can provide a more precise location of at least one animal on the physical platform. In Figure 3A In the example, the animal is close to one of the RFID interrogators 126. When the RFID interrogator identifies the animal, tracking using the load cells can be started from one or more of the load cells 108a / 108b.

[0023] In one embodiment, the processing unit is configured to perform identification of the animal using an animal identification system upon activation of a pre-defined load cell by the animal. The animal identification system does not necessarily have to be activated. For example, if the RFID reader has a known pre-defined location, the activation of a nearby load cell can be used to trigger a reading of the RFID reader. Once the animal is identified, the system can start tracking the position and / or position changes of at least one identified animal from the pre-defined load cell.

[0024] The physical platform can comprise a top plate covering the grid of cells. When a weight is placed on the top plate, it will deform. This will allow small weight changes to be recorded, so that animals, such as rodents, can be observed. As mentioned above, the load cells can be implemented in a variety of ways. In addition to the load cells, a plurality of weight distribution elements, preferably coil springs, can be provided underneath the top plate. The coil springs are preferably distributed over the entire monitoring system and can have a variety of purposes. For example, the top plate can introduce a generally smoother behavior in terms of how the top plate deforms when a weight is applied on it. In addition, it can also reduce the requirements on the installation precision of the load cells. The skilled person will understand that the system needs to be calibrated to take into account that the weight distribution elements can absorb part of the weight on the top plate. Figure 10 An example of weight distribution elements 125 provided underneath the top plate 111 is shown. In this example, the weight distribution elements 125 are arranged between the top plate 111 and the intermediate layer 118 for holding the load cells 108. In this example, the animal monitoring system further comprises a plurality of RFID interrogation units 126.

[0025] The animal monitoring system can further comprise a processing unit configured to obtain individual reference weights for a plurality of animals. This can be done, for example, prior to placing the animals on the monitoring system. Alternatively, it can also be done as an initial step after placing the animals on the monitoring system. Based on the load signals from the load cells, the processing unit can track the position of each of the plurality of animals on the physical platform. In the context of the present invention, "continuously obtaining the load signals" does not necessarily mean that the load signals are obtained all the time. As understood by the skilled person, the signals can be sampled at intervals, for example with a period of less than 5 seconds, preferably less than 1 second, more preferably less than 0.5 seconds, while still being considered as continuously obtained. In order to improve the precision of the monitoring system, a plurality of signals of the load cells can be measured within a given time period, for example 0.2 seconds, and the plurality of information can then be averaged. This process can provide a smaller error in the measured weight value of the animal. Figure 6An example of the process of an animal moving over the grid is shown, where a mouse 107 moves from a first cell 106a of the grid to a second cell 106b and then continues to a third cell 106c. As the weight changes in the three cells 106a, 106b and 106c, a trajectory 109 can be extracted.

[0026] The term“reference weight” can refer to the initial weight of each animal recorded upon entering the platform. A reference load cell can be designed on each monitoring system, where each animal is measured on the reference load cell before entering the grid. Alternatively, each animal can be weighed individually on a scale and its weight can be recorded before it can enter the monitoring platform.

[0027] As previously mentioned, the animal monitoring system disclosed herein can be adapted to be placed or integrated into a Type 3 animal cage for mice according to the European Union. More generally, the system can be adapted to track small rodents, such as rodents having a weight of less than 50 grams, including mice. The system can be adapted to track mice having a weight in the range of 10-40 grams. It will be appreciated by the skilled person that the physical and mechanical properties of the system must be adapted according to the specific circumstances and requirements. The plurality of animals is not limited to any particular number of animals, but can be, for example, 2-10 animals, or more particularly, 3-5 animals.

[0028] In one embodiment, the monitoring system can have a top plate adapted to distribute a portion of the weight exerted on a first cell of the grid of cells to one or more adjacent cells. For example, if the plate is thin enough and / or made of a flexible and / or elastic material, it can allow the weight of the animal to be distributed over the grid. If the animal is positioned in such a way that multiple load cells observe the weight on the monitoring system, this can give a more detailed position of the animal compared to if only one load cell records that the animal is positioned on one of the cells of the grid. The more detailed position can be used for the purpose of more accurately tracking the distance moved by the animal. Distributing the weight to multiple load cells can also provide information about specific activities of the animal, such as shaking, spinning, or bipedal standing, etc. For example, if the animal is shaking, it will trigger vibrations, which the processing unit can capture as a change in weight distribution. Thus, by using a machine learning model, the activities of the animal can be distinguished and pinpointed. The following section will provide more details about these activities. The processing unit can be configured to extract the weight of the at least one animal or to extract the position or position change of the at least one animal by combining the load signals from the first cell and the one or more adjacent cells.

[0029] In Figure 3AIn the example of Fig. 1, it can be seen that if an animal is positioned between a number of load cells 108a / 108b, several of these load cells 108a / 108b can provide a load signal that can be used to gain more detailed insight in the position or moving position of the animal.

[0030] The animal monitoring system can have a processing unit configured to extract the weight distribution based on the load signals from the first cell and one or more adjacent cells. The measurement of the weight distribution can then be used to assess various activities and behavioral patterns of the animal, as will be explained in more detail in the sections below.

[0031] In one embodiment, the processing unit can be further configured to calculate a modeled grid of cells having more cells than the grid of cells of the physical platform based on the extracted weight distribution. This can be achieved by using the extracted weight distribution data. For example, by measuring the load signals of the first load cell in which the animal is placed and its adjacent load cells, additional virtual load signals of intermediate load cells can be calculated by interpolation. This process can be repeated for all load cells and a calculated grid of cells can be visualized, which has a higher resolution than the physical grid. The modeled grid of cells provides the functionality of visualizing the position of the animal in the platform with higher accuracy compared to using the original grid of cells. Creating a modeled grid of cells can involve a series of steps in which one or more items, possibly a number of items of different weights, are systematically placed at different positions of the grid of cells and the load signals from the load cells are observed. Preferably, the one or more items are placed directly on the load cells and at positions between the load cells at which weight is exerted on multiple load cells. Based on the measurements, a modeled grid of cells having more cells than the grid of cells of the physical platform can be extracted. Figure 12A and 12B An example is shown in Fig. 2, which illustrates how a modeled grid of cells can work. In this example, two rodents 107a and 107b are shown in two snapshots of their movement in the monitoring platform 100. When the weight distribution of the rodents is distributed over the platform, four virtual cells 115 and 116 receiving significant weight are highlighted. The granularity of the virtual cells 115 and 116 is higher than the actual 3x3 cells in the grid.

[0032] In one embodiment of the animal monitoring system, the top plate can be flexible. This feature can enhance the functionality of the system, as it can improve the accuracy of the weight distribution, effectively increasing the resolution of the modeled grid of cells. To achieve a higher flexibility, the top plate can be thinner than 5 mm, preferably thinner than 1 mm. The top plate can be made of any suitable material, such as plastic (e.g. polystyrene or polycarbonate) or metal (e.g. aluminum).

[0033] Measuring animal weight and activity The animal monitoring system may have a processing unit further configured to extract the weight changes of a specific animal over time, as well as the activity changes of a specific animal over time. This can be achieved by continuously acquiring signals from all weighing sensors on the platform, thereby simultaneously monitoring the weight across the entire platform. If an animal moves along multiple weighing sensors, its weight changes can be monitored, and the activity of each animal can be extracted.

[0034] Following the principles described above, it is also possible to measure the distance covered by an animal within a certain time interval and compare that distance at different time points; or to measure the frequency of movement between units and compare that frequency at different time points. These operations can be performed because the dimensions of each weighing sensor are known, so the time required to travel from point A to point B can be accurately measured as the animal moves along some weighing sensors. From this, the distance covered by the animal within a certain time interval can be calculated. Data over long time periods (e.g., weeks) can also be collected and stored, and the processing unit can be configured to plot changes in distance, speed, and weight of multiple animals over periods of days or weeks. The above examples can be seen as a simplified general illustration of how movement can be calculated and extracted. Load data from the weighing sensors can provide a more detailed view of movement.

[0035] Animal surveillance systems can also calculate the location of multiple animals by: associating each animal with an identification number and / or an individual reference weight; obtaining a baseline load signal from a weighing sensor; comparing the continuously acquired load signals to the baseline load signal for all cells; and finally, identifying a specific animal in a specific cell by finding the load difference of that specific animal that matches the reference weight of that specific animal. As described above, by utilizing the ability to provide signals from weighing sensors continuously monitoring a grid, each animal present on the grid can be identified and monitored. For example, if two animals are moving toward each other, their initial weighing sensors will identify them by measuring their weight signals. As they approach each other, a weighing sensor located between them will capture the signal and identify their activity. When they reach the same weighing sensor, the total weight can be measured, and since it will be the exact sum of their two weights, the system can conclude that the two animals are placed on that weighing sensor. Using the above principles, multiple animals can also be identified when they are moving on the grid. Furthermore, animal identification is more accurate if the weight distribution of each animal is considered. The weight distribution can provide additional information that can be used as a cross-check to verify the location of each animal.

[0036] In the animal monitoring system disclosed herein, a baseline of the load signal from the weighing sensor will typically be present when the animal monitoring is initialized. This baseline can have the effect of a number of reference levels of the weighing sensor, which can be given for a setup in which a number of items are present in the cage. The items can include bedding (e.g. wood shavings) or toys (e.g. wheels or tunnels) for the animals. The processing unit can be configured to take into account and / or infer the position of an animal located on, for example, an item. For example, the system can use data indicating the position at which an item is located. If the weight of the item increases by the weight of one or more animals, it can be inferred that the animal is located on the item. In one embodiment of the monitoring system disclosed herein, the processing unit is further configured to calculate the amount of material moved from at least one cell to at least one other cell. It can be calculated whether at least one of the animals is moving material from one cell to another. This can be achieved, for example, when an animal gradually moves material (e.g. wood shavings) from one weighing sensor to another. Since the processing unit is capable of continuously monitoring the weight of each weighing sensor, it can calculate small weight changes, for example, wood shavings moved by an animal from one weighing sensor to another. The processing unit can store these small weight changes due to the movement of material along the cell units to recalibrate the baseline weight of each weighing sensor, thereby preventing inaccurate weight measurements of other animals tracked by the weighing sensors. Furthermore, it can also be possible to distinguish between weight changes due to material being moved and animals moving in the platform, since animals tend to move more dynamically and quickly, whereas material will be distributed slowly in the platform. In one embodiment of the animal monitoring system disclosed herein, the baseline of the load signal can be updated. For example, if the animals in the cage move bedding between the cells, the baseline of the load signal can be updated. The weight of the animals is still known and can be used to track the animals, as the difference in load as the animals move between different cells is observed.

[0037] As mentioned above, the position of a plurality of animals located within the same cell can be calculated by obtaining the total load of the cell and the weight distribution; and the plurality of animals is identified as being located within the cell if the combined weight of the plurality of animals matches the total load of the cell and adjacent cells. It can also be cross-checked by measuring the surrounding weighing sensors before and after time and verifying that a given animal appears on the initial weighing sensor.

[0038] Extracting animal behavior The animal monitoring system can further be configured to extract information about the behavior of the animals. In particular, the processing unit can characterize the behavior and / or the condition of at least one of the animals based on changes in the activity and / or the weight of the animals over time and make a judgment whether the plurality of animals is in a resting state by the movement of the plurality of animals over time. For example, if an animal is found to be isolated in the cell and to move much less than the other animals, the processing unit can conclude that it behaves abnormally and thus store information about its condition. On the other hand, an identified animal that moves a lot between the load cells can also be identified in this way. Information about normal or abnormal behavior can be used to determine whether certain animals show symptoms of neurodegenerative diseases, such as epilepsy or Parkinson's disease.

[0039] As described in the previous section, the weight distribution of the animals in the grid of the cell can be extracted. Thus, it can be distinguished whether an animal is standing on two or four legs, as the weight distribution will be different and thus can be tracked by the load cells and the processing unit. For example, if an animal is standing on four legs, the weight will be more evenly distributed over adjacent load cells; whereas if an animal is standing on two legs, the pressure on the load cells it is standing on will be greater, and thus the weight distribution will be concentrated under the animal. From this, the processing unit can conclude whether an animal is standing on two or four legs. Using interpolation between adjacent cells, it is also possible to obtain a higher spatial resolution than the spatial resolution provided by the different positions in each cell in the grid alone.

[0040] The weight distribution can further be used to extract small movements or shivering of the animals in the monitoring platform. For example, if an animal is grooming its fur or fighting with another animal, the processing unit can track small oscillations in the weight distribution and can distinguish the particular activity with a model.

[0041] The animal monitoring system can be configured to calculate whether an animal is standing on its hind legs, resting against the cage wall, or hanging in the cage. When an animal is partially or completely resting on a different part of the cage from the floor (e.g. the cage wall), a difference in the signals from the load cells can be measured. If the top plate does not experience any bending, the adjacent load cells will also not experience an increase in weight, and thus the processing unit can conclude that the only option left is that the animal is hanging in the grid. Alternatively, if there is a decrease from the adjacent cell but not zero, it can be concluded that the animal is resting against the cage wall. Furthermore, if the animal is standing on two legs, this can also be verified by monitoring its movements, which typically decrease when standing on two legs compared to when the animal is standing on four legs. By continuously taking input from all load cells, the activity of multiple animals in the cage can be monitored. A machine learning model can also be implemented, which can be trained with a large amount of data to more accurately distinguish whether an animal is standing on two or four legs.

[0042] The monitoring system can have a processing unit that is further configured to compare the animal activity when there is a human and / or animal present in the room outside the animal cage. If there is a human or animal activity in the proximity of the cage, special infrared sensors or cameras can be coupled to the processing unit to alert the processing unit of such an event. Thus, the activity of the animals in the cage before and after experiencing the outside presence can be compared. Examples of such activity can be the frequency of movement, whether they are in a group or not and whether they are gathered close or far from the human or animal outside the cage.

[0043] Further, the processing unit can be further configured to calculate whether an animal leaves a group of animals that are in a resting state and instead is in a resting state that is more than a predefined threshold time away from the group. For example, if the signal from a particular weighing sensor corresponds to three animals for a long amount of time without any movement being monitored, it can be concluded that the animals are in a resting state. At the same time, if the signal from the weighing sensor corresponds to one animal for a long amount of time, it can be concluded that the animal is isolated in a resting state. The "long amount of time" can be a configurable parameter threshold in the processing unit and corresponds to a typical resting time for the type of animals under study. Different time thresholds can be included in the processing unit, for example, these thresholds can dictate that the animal has less than normal activity or more than normal activity compared to the general animal population. Other uses of the thresholds can be applied when the animal weight increases or decreases, if the animal runs laps for a long amount of time, if the animal avoids gathering to sleep or if the animal moves back and forth along a particular path of the grid. Further, this process can also be successfully used for animals that have been dosed and the researcher wants to monitor changes in their activity over time.

[0044] The animal monitoring system according to the above can have a processing unit that is further configured to identify rapid repetitive movements. This can be calculated by continuously taking the signal from the weighing sensors and comparing the movement of a particular animal to previously stored movement patterns. If the activity of an animal exceeds a given threshold in the processing unit, an alarm can be triggered for that animal, which informs the researcher that the animal is experiencing rapid repetitive movements. The alarm can also be used for any other use, such as weight loss / gain, inactivity / high activity or fighting between animals. The alarm can function in such a way that the user receives information about when the activity started, the duration and, if applicable, which animals are involved.

[0045] Furthermore, the animal monitoring system according to the above can have a processing unit which is further configured to identify that the animal moves only along cells which are adjacent to a cage wall. Since it can be configured to know which cells are within the cage boundary and which are not, this can be calculated. Thus, if the animal moves only along cells which are adjacent to a cage wall, the signals from such cells will be sent to the processing unit as the animal moves, and the processing unit can conclude that the animal is moving along such cells.

[0046] The load signals from the load cells can also be used to detect specific biomarkers and disease symptoms, such as gait disturbances caused by diseases of the locomotion system, or muscle contractions of a specific frequency which are typically associated with seizures.

[0047] The load signals from the load cells can also be used to detect specific behaviors, such as aggression, avoidance, exploration, foraging or parturition.

[0048] If an abnormal behavior or an expected event, such as parturition, occurs, a notification or alarm can be triggered in order to make the researcher or animal technician aware of the status of the animal in the cage even if located at another location than where the animal is located.

[0049] Finally, the animal monitoring system can have a processing unit which is further configured to identify fast circling movements of the animal. Since the grid of modeled cells can be stored in a matrix form in rows and columns, it can be calculated when for example four modeled load cells which are adjacent to a corner are triggered. In this case, if the signals from the load cells are associated with the same weight, it can be concluded that the same animal is moving through these load cells, and thus it is making a circling movement. The time interval between the modeled load cell signals can be used to determine whether these circling animal movements are fast.

[0050] Platform specifications The animal monitoring system can be installed in a physical platform having a substantially flat top surface, and wherein the load cells are disposed below the substantially flat top surface. The monitoring system can be manufactured in a specific way to comprise three different layers: a base 119 of the monitoring system, an intermediate layer 118 for holding the load cells 108, and a top plate 111, as shown in Figure 8 The top plate 111 has an insertion portion 120 for accommodating the load cells 108.

[0051] The intermediate layer 118 is also shown in Figure 9 The intermediate layer 118 has an insertion portion 120 in which the load cells 108 are disposed. Figure 9The shown view is from the bottom side of the middle layer 118. In this embodiment, each load cell has an elongated shape, with the part that observes the weight arranged in the insert 120, and another attachment part extending and connecting to the attachment hole 105. The middle layer 118 has a number of further general attachment holes 124, which can be used to connect e.g. cables and other components to the middle layer 118. Only two of the 12 inserts 120 have a load cell in the figure - the general configuration would be that all inserts house a load cell 108. When using load cells with an elongated shape, such as the shown load cell, the distance between the load cells can be relatively long, if the load cells are placed side by side, i.e. with the short sides abutting each other. If the load cells 108 are instead arranged in an overlapping pattern, preferably also slightly rotated, such as shown in Figure 9 the load cells 108 can have a distance to each other in length and width dimensions that are similar to each other. Figure 9

[0052] The physical platform can be dimensioned to fit into a EU type 3 animal cage for mice. For example, the length of the physical platform can be 300 mm to 1000 mm, and the width can be 200 mm to 1000 mm. The monitoring system can in principle have any suitable number of units in any grid configuration M x N A typical configuration can comprise 3x3 units, or 4x3 units, or 4x4 units, or 5x5 units, etc. The distance between the load cells can be adjusted according to the area of use and the animals. For example, the distance between the load cells can be between 30 mm and 200 mm, or between 50 mm and 150 mm.

[0053] Furthermore, the monitoring system 100 can have a generally rectangular shape, with rounded corners 103, such as shown in Figure 1A The physical platform can be dimensioned to fit into a EU type 3 animal cage for mice. The type 3 cage can be made of e.g. polycarbonate, polysulfone or polypropylene. To meet the compatibility between the animal monitoring system and the type 3 cage, a tapered side can be designed on the system, which can result in a fine fit of the system in the middle of the type 3 cage. Figure 11 An example of a cross section of the tapered side 112 of the platform is shown.

[0054] As shown in Figure 1A and 1B The length 102 of the physical platform can be 375 mm to 400 mm, the width 101 can be 215 mm to 240 mm, and the height 104 can be 20 mm to 60 mm. The grid of units can be oriented diagonally with respect to the generally rectangular shape, or rotated with respect to the general orientation of the generally rectangular shape of the physical platform. It can have a general orientation difference of at least 10° with respect to the generally rectangular shape. Figure 4 ​Examples of grids with diagonally oriented cells 106 are shown. The animal monitoring system can be designed such that the physical platform has an outer shape comprising a top face, a bottom face and a plurality of, e.g. four, side faces, wherein the outer shape tapers from the top face to the bottom face.

[0055] System, software interface, display The animal monitoring system can be equipped with a display for displaying the position and any other extracted or measured data. Figure 7 Examples of such displays 106 are shown. Furthermore, as Figure 7 shown, the system can support an input interface 124 for obtaining individual weights and identification numbers of the plurality of animals; and a processing unit 123, 124. The display can be installed on each cage, on which a researcher can navigate over all stored data; or it can be a computer software, which can be accessed from a central unit, which can store and analyze data from the individual cages. The animal monitoring system can further comprise an animal identification system 127.

[0056] The animal monitoring system can further comprise peripheral components, such as one or more memories, which can be used to store instructions executable by any of the processors. The animal monitoring system can further comprise internal and external network interfaces, input and / or output ports, keyboards or mice, etc. It will be appreciated by those skilled in the art that the processing unit can also be a single processor in a multi-core / multiprocessor system. Both computing hardware accelerators and central processors can be connected to a data communication infrastructure.

[0057] The animal monitoring system can comprise a memory, such as a random access memory (RAM) and / or a read-only memory (ROM), or any suitable type of memory. The animal monitoring system can further comprise a communication interface that allows transfer of software and / or data between the system and external devices. The software and / or data transferred via the communication interface can take the form of any suitable electrical, optical or RF signal. The communication interface can include, for example, a cable interface or a wireless interface.

[0058] Method for animal monitoring The present invention also relates to a method for investigating a plurality of animals in an animal cage, the method comprising the steps of: providing a physical platform, the physical platform being divided into a grid of cells, wherein each cell comprises at least one load cell; identifying at least one animal on the physical platform using an animal identification system; continuously obtaining a load signal from the load cell; and tracking a position and / or a change in position of the identified at least one animal on the physical platform based on the load signal from the load cell.

[0059] Figure 5 A flowchart illustrating an embodiment of a method 200 of investigating animals in an animal cage is shown. The method 200 comprises the steps of: providing a physical platform divided into a grid of cells, wherein each cell comprises at least one load cell (201); identifying at least one animal on the physical platform using an animal identification system (202); continuously obtaining load signals from the load cells (203); and tracking a position and / or a change in position of the identified at least one animal on the physical platform based on the load signals from the load cells (204).

[0060] For example, the method can be applied to any kind of animals under investigation, wherein their position, behavior, activity and weight are crucial for the experiment. Since the load cells can be designed with different sizes and sensitivities, the animal monitoring platform can be accordingly designed to have larger / smaller sizes and higher / lower sensitivities on the load cells, depending on the size and weight of the animals under investigation.

[0061] Further, according to the method of investigating a plurality of animals in an animal cage described above, it can further comprise a step of sending an alert if abnormal activity is detected. For example, if an animal moves rapidly between the load cells, or an animal does not move at all for a time duration longer than the resting time of the animal, the load cells can measure the weight on the corresponding cells, and the processing unit can calculate the movement or lack of movement of the animal. Thus, if such abnormal activity is detected, an alert can be sent to the system.

[0062] The present invention also relates to a computer program having instructions which, when executed by a computing device or computing system, cause the computing device or computing system to perform any embodiment of the method of investigating a plurality of animals in an animal cage disclosed herein. The computer program can be stored on any suitable type of storage medium, such as a non-transitory storage medium.

[0063] In a further embodiment, the animal monitoring system can be used for off-cage applications. Thus, according to a further embodiment, the present invention relates to an animal monitoring system comprising: a physical platform divided into a grid of cells, wherein each cell comprises at least one load cell; a processing unit configured to: obtain individual reference weights of a plurality of animals; continuously obtain load signals from the load cells; track a position and / or a change in position of each of the plurality of animals on the physical platform based on the individual reference weights of the plurality of animals and the load signals from the load cells.

[0064] In a further embodiment, the animal monitoring system comprises: a physical platform, the platform being divided into a grid of cells, wherein each cell comprises at least one load cell; an animal identification system; a processing unit, the processing unit being configured to: identify at least one animal on the physical platform using the animal identification system; extract information about the position and movement of the plurality of animals on the physical platform based on the load signals from the load cells, and extract biomarkers of the plurality of animals on the physical platform.

[0065] Further details 1. An animal monitoring system for an animal cage, the animal monitoring system comprising: a physical platform, the platform being divided into a grid of cells, wherein each cell comprises at least one load cell; a processing unit, the processing unit being configured to: obtain individual reference weights of a plurality of animals; continuously obtain load signals from the load cells; track the position and / or position changes of each of the plurality of animals on the physical platform based on the individual reference weights of the plurality of animals and the load signals from the load cells.

[0066] 2. An animal monitoring system for an animal cage, the animal monitoring system comprising: a physical platform, the physical platform being divided into a grid of cells, wherein each cell comprises at least one load cell; an animal identification system; a processing unit, the processing unit being configured to: identify at least one animal on the physical platform using the animal identification system; continuously obtain load signals from the load cells; track the position and / or position changes of the identified at least one animal on the physical platform based on the load signals from the load cells.

[0067] 3. The animal monitoring system according to item 1 or 2, wherein the physical platform comprises a top plate covering the grid of cells.

[0068] 4. The animal monitoring system according to item 3, wherein the top plate is adapted to distribute a portion of a weight exerted on a first cell in the grid of cells to one or more adjacent cells.

[0069] 5. The animal monitoring system according to item 4, wherein the processing unit is configured to extract a weight distribution based on the load signals from the first unit and the one or more adjacent units.

[0070] 6. The animal monitoring system according to item 5, wherein the processing unit is configured to calculate a grid of modelled cells having more cells than the grid of cells of the physical platform unit based on the extracted weight distribution.

[0071] 7. The animal monitoring system according to any one of items 3 to 6, wherein the top plate is a flexible plate.

[0072] 8. The animal monitoring system according to any one of items 3 to 7, wherein the top plate has a thickness of less than 5 mm, preferably less than 1 mm, which is placed onto the grid of cells.

[0073] 9. The animal monitoring system according to any one of items 3 to 8, further comprising a weight distribution element, preferably a coil spring, which is arranged below the top plate.

[0074] 10. The animal monitoring system according to any one of the preceding items, wherein each of the at least one load cell comprises a load cell amplifier and an analog-to-digital converter.

[0075] 11. The animal monitoring system according to any one of the preceding items, wherein the processing unit is further configured to extract a weight change of a specific animal over time.

[0076] 12. The animal monitoring system according to any one of the preceding items, wherein the processing unit is further configured to extract an activity change of a specific animal over time.

[0077] 13. The animal monitoring system according to item 12, wherein the activity change comprises a measure of distance moved within a time interval compared at different points in time, or a frequency of movement between cells compared at different points in time.

[0078] 14. The animal monitoring system according to any one of the preceding items, wherein the positions of the plurality of animals are calculated by: associating each animal with an identification number and an individual reference weight; obtaining a baseline load signal from the load cells; for all cells, comparing the continuously obtained load signals with the baseline load signals; identifying a specific animal by looking for a load difference of a specific animal matching the reference weight of the specific animal in a specific cell.

[0079] 15. The animal monitoring system according to item 14, wherein the position of several animals located in the same cell is calculated by obtaining the total load of the cell; and identifying the several animals as being located in the cell if the total weight of the several animals matches the total load of the cell.

[0080] 16. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to characterize the behavior and / or condition of at least one of the animals based on changes in activity and / or weight over time.

[0081] 17. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to calculate whether several animals are in a group resting state.

[0082] 18. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to calculate whether at least one of the animals is gradually moving material from one cell to another cell.

[0083] 19. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to calculate the amount of material that has been moved from at least one cell to at least one other cell.

[0084] 20. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to calculate whether the animals are standing on their hind legs, resting against the cage wall, or hanging in the cage.

[0085] 21. The animal monitoring system according to items 5 and 20, wherein calculating whether the animals are standing on their hind legs is based on weight distribution.

[0086] 22. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to compare the animal activity when other humans and / or animals are present in the room outside the animal cage.

[0087] 23. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to calculate whether one animal is away from a plurality of animals in a group resting state, and instead is in a resting state that is more than a predefined threshold away from the group.

[0088] 24. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to calculate whether one animal has significantly less activity than the rest of the animals.

[0089] 25. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to identify rapid repetitive movements.

[0090] 26. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to identify that the animals are moving only along the cells that are adjacent to the cage wall.

[0091] 27. The animal monitoring system according to any of the preceding items, wherein the processing unit is further configured to identify a fast circling movement of the animal.

[0092] 28. The animal monitoring system according to any of the preceding items, wherein the physical platform has a generally rectangular shape with rounded corners.

[0093] 29. The animal monitoring system according to any of the preceding items, wherein the physical platform is dimensioned to fit into a type III animal cage for mice.

[0094] 30. The animal monitoring system according to any of the preceding items, wherein the physical platform has a length of 375 mm to 400 mm and a width of 215 mm to 240 mm.

[0095] 31. The animal monitoring system according to any of the preceding items, wherein the grid of cells is diagonally oriented with respect to the generally rectangular shape and / or has a total orientation difference of at least 10° with respect to the generally rectangular shape.

[0096] 32. The animal monitoring system according to any of the preceding items, wherein the physical platform has an outer shape comprising a top face, a bottom face and a plurality of (e.g. four) side faces, wherein the outer shape tapers from the top face to the bottom face.

[0097] 33. The animal monitoring system according to any of the preceding items, wherein the load cells have a sensitivity of 0.1 g or more.

[0098] 34. The animal monitoring system according to any of the preceding items, further comprising a display for displaying the position and / or any other extracted or measured data.

[0099] 35. The animal monitoring system according to any of the preceding items, further comprising an input interface for obtaining individual weights and identification numbers of the plurality of animals.

[0100] 36. A method of investigating a plurality of animals in an animal cage, comprising the steps of: providing a physical platform, the platform being divided into a grid of cells, wherein each cell comprises at least one load cell; obtaining a baseline load signal from the load cells; obtaining individual reference weights of the plurality of animals; continuously obtaining load signals from the load cells; tracking a position and / or a change in position of each of the plurality of animals on the physical platform based on the individual reference weights of the plurality of animals and the load signals from the load cells.

[0101] 37. The method of investigating a plurality of animals in an animal cage of item 36, further comprising the step of sending an alert if abnormal activity is detected.

Claims

1. An animal monitoring system for an animal cage, the animal monitoring system comprising: a physical platform divided into a grid of cells, wherein each cell comprises at least one load cell; an animal identification system; a processing unit configured to: identify at least one animal on the physical platform using the animal identification system; continuously obtain load signals from the load cells; track a position and / or a position change of the identified at least one animal on the physical platform based on the load signals from the load cells.

2. The animal monitoring system of claim 1, wherein, the animal identification system comprises one or more radio frequency identification interrogation units.

3. The animal monitoring system of claim 2, wherein, the animal identification system comprises a plurality of radio frequency identification interrogation units distributed over the physical platform.

4. An animal monitoring system according to any preceding claim, wherein, a predefined activation of a load cell triggers an identification of an animal using an associated predefined radio frequency identification interrogation unit.

5. The animal monitoring system according to any one of the preceding claims, comprising more load cells than radio frequency identification interrogation units.

6. The animal monitoring system of claim 1, wherein, the animal identification system is a camera-based animal identification system.

7. An animal monitoring system according to any preceding claim, wherein, identifying at least one animal on the physical platform using the animal identification system provides a coarse position of the at least one animal on the physical platform.

8. The animal monitoring system of claim 7, wherein, the load signals from the load cells provide a more accurate position of the at least one animal on the physical platform.

9. An animal monitoring system according to any preceding claim, wherein, the processing unit is configured to perform an identification of an animal using the animal identification system when a predefined load cell is activated by an animal.

10. The animal monitoring system of claim 9, wherein, the processing unit is configured to track a position and / or a position change of the identified at least one animal starting from a predefined load cell.

11. An animal monitoring system according to any preceding claim, wherein, the physical platform comprises a top plate covering the grid of cells, wherein the top plate is adapted to distribute a portion of a weight exerted on a first cell of the grid of cells to one or more neighboring cells.

12. The animal monitoring system of claim 11, wherein, the processing unit is configured to extract a weight of the at least one animal or to extract a position or a position change of the at least one animal by combining load signals from the first cell and the one or more neighboring cells.

13. The animal monitoring system of any one of claims 11-12, wherein, the processing unit is configured to extract a weight distribution based on load signals from the first cell and the one or more neighboring cells.

14. The animal monitoring system of any one of claims 1311-13, wherein, the processing unit is configured to calculate a modeled grid of cells based on the extracted weight distribution, the modeled grid of cells having more cells than the grid of cells of the physical platform.

15. The animal monitoring system according to any one of claims 11-14, further comprising a weight distribution element, preferably a coil spring, arranged below the top plate.

16. An animal monitoring system according to any preceding claim, wherein, the processing unit is further configured to extract a weight change of a specific animal over time.

17. An animal monitoring system according to any preceding claim, wherein, the position of several animals located in the same cell is calculated by obtaining a total load of the cell.

18. An animal monitoring system according to any preceding claim, wherein, the processing unit is further configured to characterize a behavior and / or a condition of at least one of the animals based on an activity change and / or a weight change over time.

19. A method of investigating animals in an animal cage, comprising the steps of: providing a physical platform divided into a grid of cells, wherein each cell comprises at least one load cell; identifying at least one animal on the physical platform using an animal identification system; continuously obtaining a load signal from the load cell; tracking a position and / or a change in position of the identified at least one animal on the physical platform based on the load signal obtained from the load cell.

20. A method of investigating an animal in an animal cage, the method using an animal monitoring system according to any one of claims 1-18.

Images