CN114864105A - Animal disease early warning method and system based on social graph network - Google Patents
Animal disease early warning method and system based on social graph network Download PDFInfo
- Publication number
- CN114864105A CN114864105A CN202210577566.5A CN202210577566A CN114864105A CN 114864105 A CN114864105 A CN 114864105A CN 202210577566 A CN202210577566 A CN 202210577566A CN 114864105 A CN114864105 A CN 114864105A
- Authority
- CN
- China
- Prior art keywords
- social graph
- animals
- graph network
- node
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 208000031295 Animal disease Diseases 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 44
- 241001465754 Metazoa Species 0.000 claims abstract description 232
- 230000036541 health Effects 0.000 claims abstract description 86
- 201000010099 disease Diseases 0.000 claims abstract description 46
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 46
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 35
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 230000036760 body temperature Effects 0.000 claims description 27
- 230000003862 health status Effects 0.000 claims description 7
- 208000015181 infectious disease Diseases 0.000 abstract description 13
- 238000010586 diagram Methods 0.000 description 14
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 230000006399 behavior Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 3
- 208000035143 Bacterial infection Diseases 0.000 description 2
- 241000700605 Viruses Species 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 208000022362 bacterial infectious disease Diseases 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 230000036772 blood pressure Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 208000035473 Communicable disease Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 244000052769 pathogen Species 0.000 description 1
- 230000001717 pathogenic effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/70—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in livestock or poultry
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a social graph network-based animal disease early warning method and a social graph network-based animal disease early warning system, wherein the social graph network-based animal disease early warning method comprises the following steps: acquiring health information and position information of a plurality of animals in real time; modeling the health information of a plurality of animals by using a long-short term memory (LSTM) model to obtain the current health state information of the plurality of animals; generating a social graph network of a plurality of animals according to the position information of the plurality of animals; modeling the social graph network by using a graph neural network algorithm to obtain a social graph network model, and initializing the social graph network model by using the current health state information; and acquiring node characteristics of the social graph network model, and performing disease early warning on each animal according to the node characteristics. The technical scheme of the invention can solve the problems that the prior art is difficult to early warn the health condition of animals in time and the infection condition of animal diseases cannot be early warned in advance.
Description
Technical Field
The invention relates to the technical field of neural networks, in particular to a social graph network-based animal disease early warning method and system.
Background
With the development of society, more and more people begin to raise pets. In the process of raising pets, the pets are difficult to avoid infectious diseases due to mutual contact and pathogen attachment. Because the disease condition of the pet is difficult to find or later, the adverse effect on the physical and mental health of the pet is easy to cause, therefore, the disease early warning system which can find the symptom of the disease of the animal in time even in advance has important significance for the early warning of the disease of the animal.
In the prior art, for animal disease early warning, the industry generally adopts a method of monitoring various indicators (such as body temperature, blood oxygen or blood pressure, etc.) of pets. Specific disease monitoring methods include the following: 1. the method based on the rule, when the physiological index is abnormal, namely the physiological index exceeds the normal range, the disease early warning system directly alarms; 2. the sequence-based method monitors the sequence of the physiological index, and the occurrence of sequence abnormality by using methods such as Markov chain and the like; 3. the method based on the model trains sequence models such as RNN (Ringnun neural network) by collecting sequences of physiological indexes, and uses the model to perform abnormity discrimination even advance pre-discrimination and the like through the index sequences.
For animals, the animals are easy to touch other animals in daily activities, and the animals can perform interactive behaviors such as bacterial infection, virus transmission and the like; however, the method usually only considers a single physiological index of the animal's own condition, and only can give an alarm when the pet's own health condition is abnormal, and timely early warning is difficult, so that early warning cannot be truly made on the animal's disease infection behavior.
Disclosure of Invention
The invention provides a method and a system for early warning animal diseases based on a social graph network, and aims to solve the problems that in the prior art, only a single physiological index of the self condition of an animal is considered, only an alarm can be given when the self health condition of a pet is abnormal, and early warning is difficult to be given in time, so that early warning cannot be truly given to the infection behaviors of animal diseases.
In order to achieve the above object, according to a first aspect of the present invention, the present invention provides a social graph network-based animal disease warning method, including:
acquiring health information and position information of a plurality of animals in real time;
modeling the health information of a plurality of animals by using a long-short term memory (LSTM) model to obtain the current health state information of the plurality of animals;
generating a social graph network of a plurality of animals according to the position information of the plurality of animals;
modeling the social graph network by using a graph neural network algorithm to obtain a social graph network model, and initializing the social graph network model by using the current health state information;
and acquiring node characteristics of the social graph network model, and performing disease early warning on each animal according to the node characteristics.
Preferably, in the method for warning animal diseases, the step of collecting health information and position information of a plurality of animals in real time includes:
setting a unique identification for each of a plurality of animals;
acquiring a body temperature index sequence of each animal in real time by using a body temperature sensor to serve as health information of each animal;
acquiring position information of each animal in real time by using a position sensor;
and establishing corresponding relations between the unique identification and the health information and the position information respectively.
Preferably, in the method for warning of animal diseases, the step of generating a social graph network of a plurality of animals according to the location information of the plurality of animals includes:
providing a plurality of nodes, the plurality of nodes representing a plurality of animals;
identifying contact conditions among the animals according to the position information of the animals;
generating edges among a plurality of nodes according to the contact conditions of a plurality of animals and setting the weight of the edges;
and updating the weight of the nodes, the edges and the edges in real time to obtain the social graph network of the animals.
Preferably, in the method for warning animal diseases, the step of modeling the social graph network by using a graph neural network algorithm and initializing the social graph network model by using the current health state information includes:
inputting the information of the social graph network into a message passing algorithm for modeling to obtain a social graph network model;
the node characteristics of each node in the social graph network model are initialized using the current health status information of the plurality of animals.
Preferably, in the animal disease early warning method, the step of obtaining node characteristics of the social graph network model and performing disease early warning on each animal according to the node characteristics includes:
performing multilayer convolution operation on the social graph network model according to initialized node characteristic information of each node;
acquiring node characteristic information of each node in the last layer of convolution layer of the social graph network model;
and inputting the node characteristic information of each node in the last layer of the convolutional layer into a full-connection layer of a message passing algorithm to obtain the disease early-warning information of each animal.
Preferably, in the method for warning animal diseases, the step of performing a multilayer convolution operation on the social graph network model according to initialized node feature information of each node includes:
using the initialized node characteristic information of each node according to a formula
Sequentially obtaining node characteristic information of each node in each convolution layer of the social graph network model; wherein,
According to a second aspect of the present invention, the present invention provides a social graph network-based animal disease early warning system, comprising:
the information acquisition module is used for acquiring the health information and the position information of a plurality of animals in real time;
the first modeling module is used for modeling the health information of a plurality of animals by using a long-short term memory (LSTM) model to obtain the current health state information of the plurality of animals;
the social graph network generating module is used for generating a social graph network of a plurality of animals according to the position information of the plurality of animals;
the second modeling module is used for modeling the social graph network by using a graph neural network algorithm to obtain a social graph network model and initializing the social graph network model by using the current health state information;
and the disease early warning module is used for acquiring the node characteristics of the social graph network model and carrying out disease early warning on each animal according to the node characteristics.
Preferably, in the animal disease early warning system, the information acquisition module includes:
an identification setting submodule for setting a unique identification for each of the plurality of animals;
the body temperature acquisition submodule is used for acquiring a body temperature index sequence of each animal in real time by using a body temperature sensor to serve as health information of each animal;
the position acquisition submodule is used for acquiring the position information of each animal in real time by using a position sensor;
and the relationship establishing submodule is used for establishing the corresponding relationship between the unique identifier and the health information and the position information respectively.
Preferably, in the animal disease early warning system, the social graph network generating module includes:
a node setting submodule for setting a plurality of nodes, the plurality of nodes representing a plurality of animals;
the contact identification submodule is used for identifying the contact condition among the animals according to the position information of the animals;
the edge generation submodule is used for generating edges among the nodes and setting the weight of the edges according to the contact conditions of the animals;
and the network acquisition submodule is used for updating the weights of the nodes, the edges and the edges in real time so as to acquire the social graph network of the animals.
Preferably, in the animal disease early warning system, the second modeling module includes:
the algorithm input submodule is used for inputting the information of the social graph network into a message passing algorithm to learn and update the graph network model so as to obtain the social graph network model;
and the model initialization submodule is used for initializing the node characteristics of each node in the social graph network model by using the current health state information of the animals.
Preferably, in the animal disease early warning system, the disease early warning module includes:
the multilayer convolution submodule is used for carrying out multilayer convolution operation on the social graph network model according to initialized node characteristic information of each node;
the characteristic obtaining submodule is used for obtaining node characteristic information of each node in the last layer of convolution layer of the social graph network model;
and the characteristic input submodule is used for inputting the node characteristic information of each node in the last layer of the convolutional layer into the full-connection layer of the message passing algorithm to obtain the disease early-warning information of each animal.
In summary, according to the animal disease early warning scheme based on the social graph network provided by the technical scheme of the application, health information and position information of a plurality of animals are firstly acquired, so that the social graph network can be modeled by using a graph neural network algorithm according to the position information of the plurality of animals to obtain a social graph network model, and the social graph network can reflect social relations and affinity degrees of the plurality of animals; modeling health information of a plurality of animals by using an LSTM model to obtain current health state information of each animal so as to preliminarily determine whether the animals have diseases, initializing the social graph network model by using the current health state information to obtain initial node characteristics of the social graph network model, wherein the social graph network model is a neural network model and can perform machine learning on the current health state of each animal to determine the disease condition of each animal, so that disease early warning is performed on each animal in the graph according to the node characteristics; through the animal disease early warning scheme based on the social graph network, the physiological indexes and daily activities of animals can be fused, early warning is carried out on the infection conditions of animal diseases, the accuracy of animal disease early warning is improved, and the false report condition of the animal diseases are reduced, so that the problems that in the prior art, timely early warning is difficult to carry out on the health conditions of the animals, and early warning cannot be carried out on the infection conditions of the animal diseases are solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
FIG. 1 is a block diagram of a social graph network model according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a method for animal disease warning based on a social graph network according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of an animal information collection method provided by the embodiment shown in fig. 2;
FIG. 4 is a flowchart illustrating a method for generating a social graph network according to the embodiment shown in FIG. 2;
FIG. 5 is a flowchart illustrating a modeling method of a social graph network according to the embodiment shown in FIG. 2;
FIG. 6 is a flow chart of a disease pre-warning method provided by the embodiment shown in FIG. 2;
fig. 7 is a flowchart illustrating an animal disease warning method based on a social graph network according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of an information collection module provided in the embodiment shown in FIG. 7;
FIG. 9 is a block diagram illustrating a social graph network generating module according to the embodiment shown in FIG. 7;
FIG. 10 is a schematic diagram of a second modeling module provided in the embodiment of FIG. 7;
fig. 11 is a schematic structural diagram of a disease warning module provided in the embodiment shown in fig. 7.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The embodiment of the invention mainly solves the technical problems as follows:
for animals, the animals are easy to touch other animals in daily activities, and the animals can perform interactive behaviors such as bacterial infection, virus transmission and the like; however, the existing scheme for acquiring the animal physiological indexes only considers the single physiological index of the animal self condition, only can give an alarm when the self health condition of the pet is abnormal, and is difficult to early warn in time, so that the early warning on the infection behaviors of animal diseases cannot be truly realized.
In order to solve the above problems, referring to fig. 1, the present invention provides a social graph network-based animal disease early warning scheme, which includes sampling health information (e.g., body temperature characteristics) of an animal to obtain a body temperature sampling sequence of the animal, performing machine learning processing through a sequence model network using the body temperature sampling sequence, obtaining current health status information of each animal through an LSTM network, generating a social graph network of the plurality of animals by collecting location information of the plurality of animals, wherein each node in the social graph network represents an animal, initializing node characteristics of each animal using the current health status information, obtaining a plurality of convolution layers by performing multiple convolutions of a social graph network model, performing binary classification processing using node characteristics of a last convolution layer (layer 3 in the graph), and predicting in advance whether the animal has a disease infection risk, disease forewarning was performed in advance for each animal.
To achieve the above object, please refer to fig. 2, fig. 2 is a flowchart illustrating a method for early warning of animal diseases based on a social graph network according to an embodiment of the present invention, and as shown in fig. 2, the method for early warning of animal diseases based on a social graph network includes:
s110: health information and location information of a plurality of animals is collected in real time. The health information comprises information such as body temperature, blood oxygen or blood pressure of the animal, body temperature indexes are used as the health information of the animal in the following embodiments of the application, and the position information is a coordinate point of the animal acquired in real time, so that the motion track and the social situation of the animal can be determined according to the coordinate position of the animal.
Specifically, as a preferred embodiment, as shown in fig. 3, the step of collecting health information and location information of a plurality of animals in real time includes:
s111: setting a unique identification for each of a plurality of animals; in the embodiment of the application, the ID of the constructed animal is used as the unique identification of the animal, and particularly, a chip can be implanted into the animal and integrates the ID.
S112: acquiring a body temperature index sequence of each animal in real time by using a body temperature sensor to serve as health information of each animal; the body temperature data collected in the embodiment of the application are obtained by real-time sampling, so that the serialized body temperature index sequence can be obtained and used as the health information of animals.
S113: acquiring position information of each animal in real time by using a position sensor; the position information can be acquired by using various sensors or by using the implanted chip integrated positioning function, the position coordinates of the animals are acquired in real time, the movement track of the animals is further determined, and the social relationship among a plurality of animals is searched.
S114: and establishing corresponding relations between the unique identification and the health information and the position information respectively. Because the implanted chip is arranged in the animal body, the implanted chip integrates the ID of the animal and can acquire the body temperature index sequence and the position information in real time; therefore, the relevant characteristics of the animals are established into corresponding relations, and the searching is convenient.
After the health information and the position information of a plurality of animals are collected in real time, the animal disease early warning method shown in fig. 2 further comprises the following steps:
s120: and modeling the health information of the animals by using the long-short term memory (LSTM) model to obtain the current health state information of the animals.
In the embodiment of the present application, the sequence model adopts an LSTM model, which is a cyclic model, and can reform health information of the plurality of animals, such as a body temperature index sequence, according to a time memory length, so as to model the health information of each of the plurality of animals, and obtain current health state information, where the current health state information includes a current health state of the animal, such as a current body temperature characteristic, and the current body temperature characteristic is obtained by processing the LSTM model according to a body temperature characteristic at each time in a previous predetermined period, so that the obtained current health state information can predict the health state of the animal more accurately. For example, the body temperature index sequence of the animal is input into an LSTM model for sequence modeling, and the input at each time is recorded as T t And the output is recorded as Y t ，Y t Representing to some extent the current moment health status of the current animal node.
S130: generating a social graph network of a plurality of animals according to the position information of the plurality of animals;
as a preferred embodiment, as shown in fig. 4, the step of generating a social graph network of a plurality of animals according to the location information of the plurality of animals includes:
s131: providing a plurality of nodes, wherein the plurality of nodes represent a plurality of animals; in the embodiment of the present application, the IDs of an animal and its socially contacted animals may be recorded as nodes of a social graph network.
S132: identifying contact conditions among the animals according to the position information of the animals; the contact condition between the animals comprises information such as contact time and contact frequency of the animals, and whether the animals contact or not is judged according to whether the coordinates of the two animals in the position information coincide or not.
S133: and generating edges among the nodes according to the contact conditions of the animals and setting the weight of the edges. The contact between the animals is that an edge is arranged between the animals, and if the contact is not carried out within a preset time, the edge is disconnected, wherein the weight of the edge can be determined by the length of the contact time and the number of times of contact.
S134: and updating the weight of the nodes, the edges and the edges in real time to obtain the social graph network of the animals.
According to the technical scheme, the implantable chip is arranged in the animal, the implantable chip is integrated with a positioning function, the position information of the animal is collected in real time, and the animal social graph network is arranged tender according to the position information. Specifically, a plurality of nodes are arranged, the ID of each animal is used for representing the node, edges between the nodes are arranged according to the contact condition between the animals, and the weights of the edges are arranged, so that the social graph network of the Tongtai is generated. The animal social graph network consists of nodes and edges, namely G ═ V, E, wherein V represents the nodes, and E represents the edges G represents the social graph network; node (V): i.e., a pet, represented by the ID of the pet chip; side (E): the contact between the animals is judged according to the coincidence of the positions of the animals, whether the animals contact each other is judged, whether one edge is established is judged according to the contact, and the contact time and the contact times are used as the weight of the edge;
because the animal moves, the social graph is a dynamic graph, and the social graph changes along with the time, namely, the animal contacts the node and breaks the node when the animal does not contact the node.
After the social graph network of the plurality of animals is generated, the method for warning the animal disease shown in fig. 2 further includes the following steps:
s140: and modeling the social graph network by using a graph neural network algorithm to obtain a social graph network model, and initializing the social graph network model by using the current health state information. In the embodiment of the application, a message passing algorithm is adopted in the graph neural network algorithm, the obtained social graph network is modeled by using the algorithm, and specifically, the node information is updated by adopting the social graph network information at the moment t, so that the dynamic social state between the animals is reflected. The current health state information is used as an initial characteristic of the nodes in the social graph network model. The body temperature characteristics of each animal in the social graph network can be accurately obtained through continuous convolution of the social graph network model.
Specifically, as shown in fig. 5, the method for modeling the social graph network by using the graph neural network algorithm includes the following steps:
s141: inputting the information of the social graph network into a message passing algorithm for learning and updating to obtain the social graph network model; the message passing algorithm is a mainstream graph neural network algorithm, can reflect the state of a social graph network in real time, namely the social situation of each animal, inputs the information of the social graph network, such as the weight of nodes, edges and edges, into the message passing algorithm for modeling, and can obtain the real-time social situation of the animal.
S142: initializing node characteristics of each node in the social graph network model using current health status information of a plurality of animals. The current health state information is used as the initial node characteristics of each node in the social graph network model, and the health state of each animal in the social graph can be obtained in real time along with the convolution calculation of the social graph network model, so that the disease infection condition of the animal can be timely and accurately warned.
After the social graph network model is obtained, the method for warning animal diseases shown in fig. 2 further includes:
s150: and acquiring node characteristics of the social graph network model, and performing disease early warning on each animal according to the node characteristics. In the embodiment of the application, the node characteristics reflect the health state information of the animals, and the health state information is updated along with the contact between the animals, so that the neural network algorithm is subjected to secondary classification by using the social graph network model according to the contact condition and the health state of each animal, and the disease early warning information of each animal is obtained.
Specifically, as a preferred embodiment, as shown in fig. 6, the step of obtaining node characteristics of the social graph network model and performing disease warning on each animal according to the node characteristics includes:
s151: and performing multilayer convolution operation on the social graph network model according to the initialized node characteristic information of each node.
As a preferred embodiment, the step of performing the multilayer convolution operation on the social graph network model according to the initialized node feature information of each node specifically includes:
using the initialized node characteristic information of each node according to a formula
Sequentially obtaining node characteristic information of each node in each convolution layer of the social graph network model; wherein,
Referring specifically to FIG. 1, each node in the Layer0 is characterized as
I.e. the output of the LSTM model of the i-node at time t, where Y is chosen 0 I represents the i-th node of the Layer0 convolution Layer, and corresponds to circles 1 to 5 in Layer 0.
According to the formula, each node in Layer1 is characterized by
J is all neighbor nodes of the node i at the moment t, and the total number of the neighbors is n; similarly, the feature of each node of Layer2 is represented as
Wherein j is all neighbor nodes of the node i at the time t, and the total number of the neighbors is n.
Layer3 and so on to obtain all the convolutional layers.
S152: and acquiring the node characteristic information of each node in the last layer of convolution layer of the social graph network model.
S153: and inputting the node characteristic information of each node in the last layer of the convolutional layer into a full-connection layer of a message passing algorithm to obtain the disease early-warning information of each animal.
Assuming that the last convolutional Layer is Layer3, the node characteristic information of each node in the convolutional Layer, namely the health state, is input into the fully-connected Layer of the graph neural network, so that the disease early-warning information of each animal is obtained to early warn the disease infection of the animal in advance.
In summary, according to the animal disease early warning method based on the social graph network provided by the embodiment of the application, health information and position information of a plurality of animals are collected, so that the social graph network can be modeled by using a graph neural network algorithm according to the position information of the plurality of animals to obtain a social graph network model, and the social graph network can reflect social relations and affinity degrees of the plurality of animals; modeling health information of a plurality of animals by using an LSTM model to obtain current health state information of each animal so as to preliminarily determine whether the animals have diseases, initializing the social graph network model by using the current health state information to obtain initial node characteristics of the social graph network model, wherein the social graph network model is a neural network model and can perform machine learning on the current health state of each animal to determine the disease condition of each animal, so that disease early warning is performed on each animal in the graph according to the node characteristics; through the animal disease early warning scheme based on the social graph network, the physiological indexes and daily activities of animals can be fused, early warning is carried out on the infection conditions of animal diseases, the accuracy of animal disease early warning is improved, and the false report condition of the animal diseases are reduced, so that the problems that in the prior art, timely early warning is difficult to carry out on the health conditions of the animals, and early warning cannot be carried out on the infection conditions of the animal diseases are solved.
Based on the same concept of the above method embodiment, the embodiment of the present invention further provides a social graph network-based animal disease early warning system, which is used for implementing the above method of the present invention.
Referring to fig. 7, fig. 7 is a schematic structural diagram of an animal disease warning system based on a social graph network according to an embodiment of the present invention. As shown in fig. 7, the animal disease early warning system includes:
the information acquisition module 110 is used for acquiring health information and position information of a plurality of animals in real time;
the first modeling module 120 is configured to model health information of a plurality of animals by using a long-short term memory (LSTM) model to obtain current health state information of the plurality of animals;
a social graph network generating module 130, configured to generate a social graph network of the multiple animals according to the location information of the multiple animals;
the second modeling module 140 is configured to model the social graph network by using a graph neural network algorithm to obtain a social graph network model, and initialize the social graph network model by using the current health state information;
and the disease early warning module 150 is configured to obtain node characteristics of the social graph network model, and perform disease early warning on each animal according to the node characteristics.
In summary, according to the animal disease early warning system based on the social graph network provided by the embodiment of the present application, first, the information acquisition module 110 acquires health information and location information of a plurality of animals, so that the social graph network generation module 130 can be used to model the social graph network by using a graph neural network algorithm according to the location information of the plurality of animals, so as to obtain a social graph network model, wherein the social graph network can reflect social relationships and affinity degrees of the plurality of animals; the first modeling module 120 models health information of a plurality of animals by using an LSTM model to obtain current health state information of each animal to preliminarily determine whether the animal has a disease, and initializes the social graph network model by using the current health state information to obtain initial node characteristics of the social graph network model, wherein the social graph network model is a neural network model and can perform machine learning on the current health state of each animal to determine the disease condition of each animal, so that the disease early warning module 150 can perform disease early warning on each animal in the graph according to the node characteristics. Through the animal disease early warning scheme based on the social graph network, the physiological indexes and daily activities of animals can be fused, early warning is carried out on the infection conditions of animal diseases, the accuracy of animal disease early warning is improved, and the false report condition of the animal diseases are reduced, so that the problems that in the prior art, timely early warning is difficult to carry out on the health conditions of the animals, and early warning cannot be carried out on the infection conditions of the animal diseases are solved.
As a preferred embodiment, as shown in fig. 8, in the animal disease early warning system, the information collecting module 110 includes:
an identification setting sub-module 111 for setting a unique identification for each of the plurality of animals;
the body temperature acquisition submodule 112 is used for acquiring a body temperature index sequence of each animal in real time by using a body temperature sensor to serve as health information of each animal;
a position acquisition submodule 113 for acquiring position information of each animal in real time using a position sensor;
a relationship establishing sub-module 114, configured to establish a corresponding relationship between the unique identifier and the health information and the location information, respectively.
As a preferred embodiment, as shown in fig. 9, in the animal disease warning system, the social graph network generating module 130 includes:
a node setting submodule 131 for setting a plurality of nodes representing a plurality of animals;
a contact identification submodule 132 for identifying contact conditions among the plurality of animals according to the position information of the plurality of animals;
an edge generation submodule 133, configured to generate edges between the plurality of nodes and set weights of the edges according to contact conditions of the plurality of animals; the contact condition of the animal comprises contact time and contact times of a plurality of animals.
And the network acquisition sub-module 134 is used for updating the weights of the nodes, the edges and the edges in real time so as to acquire the social graph network of the animals.
As a preferred embodiment, as shown in fig. 10, in the animal disease early warning system, the second modeling module 140 includes:
the algorithm input submodule 141 is configured to input information of the social graph network to a message passing algorithm for modeling, so as to obtain a social graph network model;
and the model initialization submodule 142 is used for initializing the node characteristics of each node in the social graph network model by using the current health state information of the animals.
As a preferred embodiment, as shown in fig. 11, in the animal disease early warning system, the disease early warning module 150 includes:
the multilayer convolution sub-module 151 is configured to perform multilayer convolution operation on the social graph network model according to the initialized node feature information of each node;
a feature obtaining submodule 152, configured to obtain node feature information of each node in a last layer of convolutional layer of the social graph network model;
and the characteristic input submodule 153 is used for inputting the node characteristic information of each node in the last convolution layer into the full connection layer of the message passing algorithm to obtain the disease early warning information of each animal.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. An animal disease early warning method based on a social graph network is characterized by comprising the following steps:
acquiring health information and position information of a plurality of animals in real time;
modeling the health information of the animals by using a long-short term memory (LSTM) model to obtain the current health state information of the animals;
generating a social graph network of the plurality of animals according to the position information of the plurality of animals;
modeling the social graph network by using a graph neural network algorithm to obtain a social graph network model, and initializing the social graph network model by using the current health state information;
and acquiring node characteristics of the social graph network model, and performing disease early warning on each animal according to the node characteristics.
2. The method for warning of animal diseases according to claim 1, wherein the step of collecting health information and location information of a plurality of animals in real time comprises:
setting a unique identification for each animal of the plurality of animals;
acquiring a body temperature index sequence of each animal in real time by using a body temperature sensor to serve as health information of each animal;
acquiring position information of each animal in real time by using a position sensor;
and establishing corresponding relations between the unique identification and the health information and between the unique identification and the position information respectively.
3. The method as claimed in claim 1, wherein the step of generating the social graph network of the plurality of animals according to the location information of the plurality of animals comprises:
providing a plurality of nodes representing the plurality of animals;
identifying contact conditions among the animals according to the position information of the animals;
generating edges among the nodes according to the contact conditions of the animals and setting the weight of the edges;
and updating the weight of the nodes, the edges and the edges in real time to obtain the social graph network of the animals.
4. The method of claim 1, wherein the step of modeling the social graph network using a graph neural network algorithm and initializing the social graph network model using current health status information comprises:
inputting the information of the social graph network into a message passing algorithm to learn and update a graph network model to obtain the social graph network model;
initializing node characteristics of each node in the social graph network model using the current health status information of the plurality of animals.
5. The method as claimed in claim 4, wherein the step of obtaining node features of the social graph network model and performing disease warning on each animal according to the node features comprises:
performing multilayer convolution operation on the social graph network model according to initialized node characteristic information of each node;
acquiring node characteristic information of each node in the last layer of convolution layer of the social graph network model;
and inputting the node characteristic information of each node in the last layer of the convolutional layer into a full-connection layer of the message passing algorithm to obtain the disease early warning information of each animal.
6. The method as claimed in claim 5, wherein the step of performing a multi-layer convolution operation on the social graph network model according to initialized node feature information of each node comprises:
using the initialized node characteristic information of each node according to a formula
Sequentially obtaining each convolution layer of the social graph network modelNode characteristic information of the node; wherein,
7. An animal disease early warning system based on a social graph network, comprising:
the information acquisition module is used for acquiring the health information and the position information of a plurality of animals in real time;
the first modeling module is used for modeling the health information of the animals by using a long-short term memory (LSTM) model to obtain the current health state information of the animals;
the social graph network generating module is used for generating a social graph network of the animals according to the position information of the animals;
the second modeling module is used for modeling the social graph network by using a graph neural network algorithm to obtain a social graph network model and initializing the social graph network model by using the current health state information;
and the disease early warning module is used for acquiring the node characteristics of the social graph network model and carrying out disease early warning on each animal according to the node characteristics.
8. The system of claim 7, wherein the social graph network generating module comprises:
a node setting submodule for setting a plurality of nodes representing the plurality of animals;
the contact identification submodule is used for identifying the contact condition among the animals according to the position information of the animals;
the edge generation submodule is used for generating edges among the nodes and setting the weight of the edges according to the contact conditions of the animals;
and the network acquisition submodule is used for updating the weights of the nodes, the edges and the edges in real time so as to acquire the social graph network of the animals.
9. The animal disease warning system of claim 7, wherein the second modeling module comprises:
the algorithm input sub-module is used for inputting the information of the social graph network into a message passing algorithm to learn and update a graph network model so as to obtain the social graph network model;
and the model initialization submodule is used for initializing the node characteristics of each node in the social graph network model by using the current health state information of the animals.
10. The animal disease warning system of claim 9, wherein the disease warning module comprises:
the multilayer convolution submodule is used for carrying out multilayer convolution operation on the social graph network model according to initialized node characteristic information of each node;
the characteristic obtaining submodule is used for obtaining node characteristic information of each node in the last layer of convolution layer of the social graph network model;
and the characteristic input submodule is used for inputting the node characteristic information of each node in the last layer of the convolutional layer to the full-connection layer of the message passing algorithm to obtain the disease early-warning information of each animal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210577566.5A CN114864105A (en) | 2022-05-25 | 2022-05-25 | Animal disease early warning method and system based on social graph network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210577566.5A CN114864105A (en) | 2022-05-25 | 2022-05-25 | Animal disease early warning method and system based on social graph network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114864105A true CN114864105A (en) | 2022-08-05 |
Family
ID=82639160
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210577566.5A Pending CN114864105A (en) | 2022-05-25 | 2022-05-25 | Animal disease early warning method and system based on social graph network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114864105A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115040095A (en) * | 2022-08-15 | 2022-09-13 | 北京九叁有方物联网科技有限公司 | Aging-suitable multifunctional autonomous non-invasive dynamic physiological signal monitoring and analyzing system |
| CN116602663A (en) * | 2023-06-02 | 2023-08-18 | 深圳市震有智联科技有限公司 | Intelligent monitoring method and system based on millimeter wave radar |
| CN116756667A (en) * | 2023-08-18 | 2023-09-15 | 北京九叁有方物联网科技有限公司 | Pet wearable device with behavior recognition function and recognition method thereof |
| CN116860904A (en) * | 2023-08-31 | 2023-10-10 | 河北金锁安防工程股份有限公司 | Target tracking method and device based on decentralization network |
-
2022
- 2022-05-25 CN CN202210577566.5A patent/CN114864105A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115040095A (en) * | 2022-08-15 | 2022-09-13 | 北京九叁有方物联网科技有限公司 | Aging-suitable multifunctional autonomous non-invasive dynamic physiological signal monitoring and analyzing system |
| CN115040095B (en) * | 2022-08-15 | 2023-01-24 | 北京九叁有方物联网科技有限公司 | Aging-suitable multifunctional autonomous non-invasive dynamic physiological signal monitoring and analyzing system |
| CN116602663A (en) * | 2023-06-02 | 2023-08-18 | 深圳市震有智联科技有限公司 | Intelligent monitoring method and system based on millimeter wave radar |
| CN116602663B (en) * | 2023-06-02 | 2023-12-15 | 深圳市震有智联科技有限公司 | Intelligent monitoring method and system based on millimeter wave radar |
| CN116756667A (en) * | 2023-08-18 | 2023-09-15 | 北京九叁有方物联网科技有限公司 | Pet wearable device with behavior recognition function and recognition method thereof |
| CN116756667B (en) * | 2023-08-18 | 2023-12-12 | 北京九叁有方物联网科技有限公司 | Pet wearable device with behavior recognition function and recognition method thereof |
| CN116860904A (en) * | 2023-08-31 | 2023-10-10 | 河北金锁安防工程股份有限公司 | Target tracking method and device based on decentralization network |
| CN116860904B (en) * | 2023-08-31 | 2024-06-04 | 河北金锁安防工程股份有限公司 | Target tracking method and device based on decentralization network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114864105A (en) | Animal disease early warning method and system based on social graph network | |
| US11404145B2 (en) | Medical machine time-series event data processor | |
| Nashif et al. | Heart disease detection by using machine learning algorithms and a real-time cardiovascular health monitoring system | |
| CN103778312A (en) | Remote home health care system | |
| CN108133274A (en) | For system and method for protocol adherence | |
| CN114419528A (en) | Anomaly identification method and device, computer equipment and computer readable storage medium | |
| US20160267254A1 (en) | System and method for classifying respiratory and overall health status of an animal | |
| Danjuma et al. | Evaluation of predictive data mining algorithms in erythemato-squamous disease diagnosis | |
| JP2018147474A (en) | Learning device, learning result utilization device, learning method, and learning program | |
| US20190051383A1 (en) | Intelligent sepsis alert | |
| Sanchez et al. | Hidden markov models for activity recognition in ambient intelligence environments | |
| García-Temza et al. | Comparison of different machine learning approaches to model stroke subtype classification and risk prediction | |
| CN114241270A (en) | Intelligent monitoring method, system and device for home care | |
| Noor et al. | Fetal health classification using supervised learning approach | |
| JP6846369B2 (en) | Programs, devices and methods for classifying unknown multidimensional vector data groups | |
| CN118038548A (en) | Abnormal behavior detection method, device, electronic equipment and storage medium | |
| CN114730392A (en) | Analysis device, analysis method, and analysis program | |
| KR102395173B1 (en) | A wearable apparatus to manage health care for companion animal | |
| Bastaki | Application of hierarchical temporal memory to anomaly detection of vital signs for ambient assisted living | |
| CN115937903A (en) | Fall risk assessment method and device, electronic equipment and storage medium | |
| Dhanusha et al. | Enhanced deep hierarchical classification model for smart home-based alzheimer disease detection | |
| EP4086811A1 (en) | Problematic behavior classification system and method based on deep neural network algorithm | |
| CN115176255A (en) | Parameter adjustment device, inference device, parameter adjustment method, and parameter adjustment program | |
| Paris et al. | Using hidden semi-Markov model for learning behavior in smarthomes | |
| Perez et al. | A cluster-based approach to predict serious adverse events in surgery |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |