Detection and management of disease outbreaks in livestock using health graph networks
Sensor data captured over time and associated with a plurality of livestock animals can be analyzed to determine health conditions associated with a plurality of livestock animals and to determine interactions of the livestock animals. The interactions specify at least distances between the livestock animals, duration of the distances, and frequency of the interactions. A health graph network is constructed, which includes nodes and edges, a node in the nodes representing a livestock animal and specifying at least a health condition of the represented livestock animal, an edge connecting at least two of the nodes and representing an interaction between at least two animals represented by said at least two of the nodes. Based on the health graph network, a potential outbreak among a subgroup of the livestock animals can be predicted.
1. A computer-implemented method comprising:
analyzing sensor data captured over time associated with a plurality of livestock animals;
determining a health condition associated with a plurality of livestock animals based on the analyzing;
determining interactions of the livestock animals based on the analyzing, the interactions specifying at least a distance between the livestock animals, a duration of the distance, and a frequency of the interactions;
building a health graph network including at least nodes and edges, a node in the graph network representing a livestock animal and specifying at least a health condition of the represented livestock animal, an edge in the graph network connecting at least two of the nodes and representing an interaction between at least two animals represented by said at least two of the nodes;
based on the health graph network, predicting a potential outbreak among a subgroup of the livestock animals,
wherein a connectivity between the at least two of the nodes is determined based on at least a proximity detected of the at least two animals represented by said at least two of the nodes, wherein a proximity detection ignores animals facing away from each other as being in proximity.
2. The method of claim 1 , further comprising recommending an amelioration action to prevent the potential outbreak from spreading.
3. The method of claim 1 , wherein the health condition is determined at least based on detection a deviation of an animal's behavior from the normal healthy behavior over time.
4. The method of claim 1 , further comprising back tracing the edges in the health graph network over time to identify animals that have come into contact with one or more animals involved in the potential outbreak.
5. The method of claim 1 , further comprising predicting a movement path of at least one livestock animal in a time scale.
6. The method of claim 5 , further comprising visualizing the predicted movement path using augmented reality technology.
7. The method of claim 6 , further comprising triggering by sending a trigger event to augmented reality glasses to display the current status of the animals responsive to a deviated value exceeding a predefined threshold value.
8. A computer-implemented system comprising:
a hardware processor;
a memory device coupled to the hardware processor;
the hardware processor operable to at least:
analyze sensor data captured over time associated with a plurality of livestock animals;
determine a health condition associated with the plurality of livestock animals based on the analyzing;
determine interactions of the livestock animals based on the analyzing, the interactions specifying at least a distance between the livestock animals, a duration of the distance, and a frequency of the interactions;
build a health graph network including at least nodes and edges, a node in the network graph representing a livestock animal and specifying at least a health condition of the represented livestock animal, an edge in the network graph connecting at least two of the nodes and representing an interaction between at least two animals represented by said at least two of the nodes;
based on the health graph network, predict a potential outbreak among a subgroup of the livestock animals,
wherein a connectivity between the at least two of the nodes is determined based on at least a proximity detected of the at least two animals represented by said at least two of the nodes, wherein a proximity detection ignores animals facing away from each other as being in proximity.
9. The system of claim 8 , wherein the hardware processor is further operable to recommend an amelioration action to prevent the potential outbreak from spreading.
10. The system of claim 8 , wherein the health condition is determined at least based on detection a deviation of an animal's behavior over time.
11. The system of claim 8 , wherein the hardware processor is further operable to back trace the edges in the health graph network over time to identify animals that have come into contact with one or more animals involved in the potential outbreak.
12. The system of claim 8 , wherein the hardware processor is further operable to predict a movement path of at least one livestock animal in a time scale.
13. The system of claim 12 , wherein the hardware processor is further operable to cause visualizing of the predicted movement path using augmented reality technology.
14. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a device to cause the device to:
analyze sensor data captured over time associated with a plurality of livestock animals;
determine a health condition associated with a plurality of livestock animals based on the analyzing;
determine interactions of the livestock animals based on the analyzing, the interactions specifying at least a distance between the livestock animals, a duration of the distance, and a frequency of the interactions;
build a health graph network including at least nodes and edges, a node in the nodes representing a livestock animal and specifying at least a health condition of the represented livestock animal, an edge connecting at least two of the nodes and representing an interaction between at least two animals represented by said at least two of the nodes;
based on the health graph network, predict a potential outbreak among a subgroup of the livestock animals,
wherein a connectivity between the at least two of the nodes is determined based on at least a proximity detected of the at least two animals represented by said at least two of the nodes, wherein a proximity detection ignores animals facing away from each other as being in proximity.
15. The computer program product of claim 14 , wherein the device is further caused to recommend an amelioration action to prevent the potential outbreak from spreading.
16. The computer program product of claim 14 , wherein the health condition is determined at least based on detection a deviation of an animal's behavior over time.
17. The computer program product of claim 14 , wherein the device is further caused to back trace the edges in the health graph network over time to identify animals that have come into contact with one or more animals involved in the potential outbreak.
18. The computer program product of claim 14 , wherein the device is further caused to predict a movement path of at least one livestock animal in a time scale.
19. The computer program product of claim 18 , wherein the device is further caused to cause visualizing of the predicted movement path using augmented reality technology.
20. The computer program product of claim 19 , wherein the device is further caused to display connectivity among the livestock animals overlaid on a real view of a farm with the livestock animals.