Description:TECHNICAL FIELD
The present disclosure relates generally to the field of animal husbandry; and more specifically, to an electronic system and a method for animal health management.
BACKGROUND
In the realm of the livestock industry, collecting and analyzing data related to emerging disease trends occurring in animals across various geographic regions and then ensuring the optimal health and well-being of farm animals across diverse geographical regions is a formidable challenge. The existing methodologies for diagnosing animal health issues often suffer from imprecision and hinder effective disease identification and deficiency management, resulting in potential economic losses for farmers and compromised animal welfare.
Existing techniques struggle to capture the complex interplay between health conditions, diseases, deficiencies, and interventions that are unique to each farm animal. As a result, there exists a significant gap in the ability to proactively and effectively address health concerns in farm animals. It is observed that existing systems applied for monitoring farm animals for nutrition deficiency and diseases are often inaccurate, unreliable, expensive, and complex. This can lead to false positives, false negatives, and missed opportunities for early detection and treatment. For example, a false positive could lead to a farmer unnecessarily treating an animal, while a false negative could lead to a farmer missing early signs of a disease or nutritional deficiency.
In a first example, existing electronic systems use a plethora of sensors or wearable sensors tied to body of animals. It becomes a practical challenge over a period of time how to ensure if all sensors are functioning optimally or even functioning or not functioning on some animals. Thus, reliability and accuracy get hampered drastically over a period of time. Further, conventional systems often rely on indirect measures of animal health, such as feed intake, weight gain, and milk production, etc. These measures can be inaccurate and unreliable, especially in early stages of disease. For instance, a farmer may rely on a system that monitors feed intake to identify animals that are not eating enough. However, if the system is not accurate, the farmer may miss early signs of a disease that is causing the animal to lose its appetite.
Therefore, in light of the foregoing discussion, there exists a need to overcome the aforementioned drawbacks associated with the conventional systems or equipment applied to animal health monitoring and management.
SUMMARY
The present disclosure provides an electronic system of animal health management and a method implemented in an electronic system of animal health management. The present disclosure provides a solution to the existing problem of inaccuracy, unreliability, and system complexity manifested over a period of time by existing systems applied to animal health management, and how to technically reduce false positives and false negatives manifested by conventional systems in the domain of animal health management. . An aim of the present disclosure is to provide a solution that overcomes at least partially the problems encountered in prior art and provide an improved electronic system applied to a specific domain of animal health management and an improved method implemented in the electronic system of animal health management. The electronic system of the present disclosure provides a new way of monitoring and management of animal health that is accurate, reliable, affordable, and easy to use, where the accuracy and reliability instead of decreasing actually further increases over the period of time. The simplified approach lies in the fact that the electronic system uses biological markers and photographs of a plurality of target animals in a unique manner to significantly reduce false positives and false negatives as compared to conventional systems in the domain of animal health management.
One or more objectives of the present disclosure is achieved by the solutions provided in the enclosed independent claims. Advantageous implementations of the present disclosure are further defined in the dependent claims.
In one aspect, the present disclosure provides an electronic system of animal health management. The electronic system includes a hardware controller configured to receive a set of biological markers and photographs of a plurality of target animals. The hardware controller is further configured to extract one or more features for each target animal from the received set of biological markers and photographs. The hardware controller is further configured to determine variation in the one or more extracted features of each target animal. The variation in the one or more extracted features is indicative of potential discrepancies or alterations within health conditions or visual attributes of each target animal. The hardware controller is further configured to identify nutritional deficiencies or diseases in the plurality of target animals based on the determined variation in the one or more extracted features of each target. The hardware controller is further configured to form a knowledge graph based on a rumen fluid analysis (RFA) dataset from RFA conducted on a set of sample animals distributed across a plurality of different geographical regions, considering influence of a set of environmental factors on health conditions or visual attributes of the set of sample animals. The knowledge graph includes relationships of each health-related parameter of a set of health-related parameters with a specific disease and a nutritional deficiency. Each health-related parameter within the knowledge graph is associated with the set of sample animals influenced by the set of environmental factors. The hardware controller is further configured to correlate the identified nutritional deficiencies or diseases with the relationships in the knowledge graph to remediate the identified nutritional deficiencies or diseases. The hardware controller is further configured to generate tailored nutrition recommendations for each target animal of the plurality of target animals based on the correlated nutritional deficiencies or diseases and the relationships in the knowledge graph.
The integration of photographs and biological markers in the electronic system's input is leveraged to extract relevant visual features from the photographs, such as the subtle changes in an animal's physical appearance. Simultaneously, it processes biological markers, which may include complex biochemical data, to quantitatively assess the animal's health status. By considering both visual and biochemical indicators, a more comprehensive and accurate assessment which is specific to a given animal among hundreds or thousands of animals is made in subsequent operations. Further, variations in the extracted features can be subtle or complex, and their determination may not be readily achievable through manual inspection. Not only such these variations are determined by the electronic system but also used as indicators of potential health discrepancies or alterations in the target animals, enabling identification of nutritional deficiencies if any specific to each animal. Subsequently, the system utilizes a knowledge graph that integrates rumen fluid analysis (RFA) data and environmental factors. This knowledge graph, formed through comprehensive data aggregation and analysis, offers a holistic view of animal health conditions, diseases, and nutritional deficiencies taking into account inherent differences among animals across different geographical regions. In an example, there are two different geographical regions, Region A and Region B, which have distinct mineral compositions in their soil and water sources. Region A is naturally rich in iron and oxygen due to its geological characteristics. Animals in this region may have sufficient access to these minerals in their diets. The system, taking into account the regional differences, ensures that tailored nutrition recommendations for animals in Region A consider the existing abundance of iron and oxygen in their environment. This can lead to the adjustment of supplements or feed formulations to avoid over-supplementation, which can be harmful. Region B has water sources with a higher sulfur content. Animals in this region might already receive adequate sulfur intake through their water supply. The system recognizes this regional variation and avoids recommending excess sulfur supplementation, which could be detrimental to animal health. This provides a technical effect because as inherent differences among animals across different geographical regions is considered, the tailored nutrition recommendations for each target animal considers is accurate and reliable, where the false positives and false negatives is significantly reduced as compared to conventional systems, for example, about 30-40 % reduction in false positives and false negatives observed. Further, the comprehensive insight enables targeted interventions and preventive measures.
Furthermore, by correlating identified nutritional deficiencies or diseases with knowledge graph relationships, the system gains predictive capabilities with improved accuracy, reliability, and significantly further reduces the false positives and false negatives as compared to conventional systems, for example, a further reduction in false positives and false negatives from 30-40% reduction to 41-85% reduction, is observed, which in turn improves the accuracy and reliability of the electronic system (accuracy about 95-100 % achieved irrespective of geographical regions where the system is deployed).This allows for the proactive identification of potential health issues based on patterns and trends, enabling proactive measures to prevent and manage diseases. Further, the system generates tailored nutrition recommendations for each target animal based on its specific health conditions and the knowledge graph insights. This personalized approach ensures that interventions are precise and aligned with the individual needs of each animal in practice in a real-word application.
In an implementation form, the hardware controller is further configured to train an AI engine on a training dataset stored in a first database, and wherein the training dataset comprises a set of annotated images of animals, each annotated image of a set of annotated images linked with metadata to provide context for features of an animal corresponding to each annotated image.
In this implementation, the AI engine is trained on annotated images linked with metadata, enhancing its ability to recognize variations in extracted features by considering the contextual information associated with each animal image, leading to more accurate health assessments.
In an implementation form, the determination of the variation in the one or more extracted features of each target animal comprises executing the artificial intelligence (AI) engine to detect a variation pattern in the one or more extracted features of each target animal based on a geographical region of the plurality of different geographical regions of each target animal.
The AI engine detects variation patterns in extracted features based on the geographical region of each target animal. This geospatial context enhances the system's ability to assess health conditions and visual attributes, considering environmental factors specific to different geographical regions, which is valuable for precision animal health management and reduction in false positives and false negatives.
In an implementation form, in order to determine the variation in the one or more extracted features of each target animal, the hardware controller is further configured to compare the extracted features with the set of annotated images of the training dataset.
The hardware controller uses the training dataset's annotated images to compare and analyze the extracted features of each target animal. This enables the system to identify variations by referencing a dataset with known attributes that also considers the inherent differences of health attributes across different geographical regions, enhancing the precision of health assessments.
In an implementation form, in order to identify the nutritional deficiencies or diseases in the plurality of target animals, the hardware controller is further configured to compare the determined variations in the one or more extracted features with predefined thresholds.
In this case, the hardware controller compares determined variations in extracted features with predefined thresholds to identify nutritional deficiencies or diseases in target animals. This method provides an objective and systematic approach to health issue recognition.
In an implementation form, the hardware controller is further configured to execute a feedback loop to continuously refine an accuracy parameter of the nutrition recommendation over time by detecting shifts or anomalies in the set of health-related parameters of the plurality of target animals that emerges over a predefined time period.
This implementation provides a further technical effect of dynamic nutrition adjustment, as the hardware controller continuously refines the accuracy of nutrition recommendations by detecting shifts or anomalies in health-related parameters of target animals over time. This dynamic adaptation ensures that the recommendations remain effective as the animals' health conditions change given a particular geographical location the animal lives or when shifted to a new geographical location.
In an implementation form, the one or more extracted features comprises visual characteristics, vital signs, biochemical markers, digestive metrics, reproductive insights, behavioural patterns, milking performance metrics, feeding behaviour indicators, and immune function markers.
As the system analyses a wide array of extracted features, including visual, vital, biochemical, digestive, reproductive, behavioural, milking, feeding, and immune function data. This comprehensive analysis enhances the system's ability to detect and assess animal health and well-being from multiple perspectives.
In an implementation form, the AI engine utilizes computer vision operation to analyze the set of photographs of the plurality of target animals and determine variations in visual attributes of the plurality of target animals.
This implementation provides computer vision-powered visual attribute analysis, where the AI engine employs computer vision operations to analyze photographs of target animals, enabling it to detect and quantify variations in visual attributes accurately. This enhances the system's ability to assess the animals' physical condition and appearance for health monitoring. The term “computer vision operations" refers to specific algorithms and processes used to extract, analyze, and interpret information from images or videos. These operations encompass tasks like object detection, image enhancement, and pattern recognition. The use of computer vision operations enables the AI engine to analyze photographs of target animals, allowing it to detect and quantify variations in visual attributes for assessing the animals' physical condition and appearance.
In an implementation form, the hardware controller is further configured to control display of the tailored nutrition recommendations through a user interface accessible via mobile applications or web platforms, and wherein the tailored nutrition recommendations comprises adjustments in feed composition, dosages of nutritional supplements, and specific dietary regimens customized for each target animal's health profile.
These recommendations include precise adjustments in feed composition, nutritional supplement dosages, and dietary regimens tailored to each target animal's unique health profile. This technical feature allows for convenient and highly individualized animal health management, facilitated by user-friendly interfaces.
In an implementation form, the set of environmental factors comprises climatic conditions, geographical locations, and specific farm management methodologies.
The electronic system considers a comprehensive set of environmental factors, including climatic conditions, geographical locations, and specific farm management methodologies, when assessing animal health. This contextualization enhances the system's ability to provide precise health insights, accounting for the broader environmental and geographical context in which the animals are situated.
In another aspect, the present disclosure provides a method implemented in an electronic system of animal health management. The method includes receiving, by a hardware controller, a set of biological markers and photographs of a plurality of target animal. The method further includes extracting, by the hardware controller, one or more features for each target animal from the received set of biological markers and photographs. The method further includes determining , by the hardware controller, variation in the one or more extracted features of each target animal. The variation in the one or more extracted features is indicative of potential discrepancies or alterations within the health conditions or visual attributes of each target animal. The method further includes identifying, by the hardware controller, nutritional deficiencies or diseases in each target animal based on the determined variation in the one or more extracted features of each target. The method further includes forming, by the hardware controller, a knowledge graph based on a rumen fluid analysis (RFA) dataset from RFA conducted on a set of sample animals distributed across a plurality of geographical locations. The knowledge graph includes relationships of each health-related parameter of a set of health-related parameters with a specific disease and a nutritional deficiency, wherein each health-related parameter within the knowledge graph is exhibited by the set of sample animals distributed across the plurality of geographical locations. The method further includes correlating, by the hardware controller, the identified nutritional deficiencies or diseases with the relationships in the knowledge graph to remediate the identified nutritional deficiencies or diseases. The method further includes generating, by the hardware controller, tailored nutrition recommendations for each target animal of the plurality of target animals based on the correlated nutritional deficiencies or diseases and the relationships in the knowledge graph.
The method achieves all the advantages and technical effects of the system of the present disclosure.
It is to be appreciated that all the aforementioned implementation forms can be combined. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. It will be appreciated that features of the present disclosure are susceptible to being combined in various combinations without departing from the scope of the present disclosure as defined by the appended claims. Additional aspects, advantages, features and objects of the present disclosure would be made apparent from the drawings and the detailed description of the illustrative implementations construed in conjunction with the appended claims that follow.
BRIEF DESCRIPTION OF THE DRAWINGS
The summary above, as well as the following detailed description of illustrative embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the present disclosure, exemplary constructions of the disclosure are shown in the drawings. However, the present disclosure is not limited to specific methods and instrumentalities disclosed herein. Moreover, those skilled in the art will understand that the drawings are not to scale. Wherever possible, like elements have been indicated by identical numbers.
Embodiments of the present disclosure will now be described, by way of example only, with reference to the following diagrams wherein:
FIG. 1 is a block diagram illustrating an electronic system for animal health management, in accordance with an embodiment of the present disclosure;
FIG. 2 is a flowchart of operation that illustrates a process of identifying nutritional deficiencies or diseases in animals and generating nutrition recommendations for the animals, in accordance with an embodiment of the present disclosure; and
FIG. 3 is a flowchart of a method for animal health management, in accordance with an embodiment of the present disclosure.
DETAILED DESCRIPTION OF EMBODIMENTS
The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practicing the present disclosure are also possible.
As used throughout this disclosure, the word “may” is used in a permissive sense (i.e., meaning having the potential to), rather than the mandatory sense (i.e., meaning must). Similarly, the words “include”, “including”, and “includes” mean including but not limited to.
The phrases “at least one”, “one or more”, and “and/or” are open-ended expressions that are both conjunctive and disjunctive in operation. For example, each of the expressions “at least one of A, B and C”, “at least one of A, B, or C”, “one or more of A, B, and C”, “one or more of A, B, or C” and “A, B, and/or C” means A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B and C together.
The term “a” or “an” entity refers to one or more of that entity. As such, the terms “a” (or “an”), “one or more” and “at least one” can be used interchangeably herein. It is also to be noted that the terms “comprising”, “including”, and “having” can be used interchangeably.
The term “automatic” and variations thereof, as used herein, refers to any process or operation done without material human input when the process or operation is performed. However, a process or operation can be automatic, even though performance of the process or operation uses material or immaterial human input, if the input is received before performance of the process or operation. Human input is deemed to be material if such input influences how the process or operation will be performed. Human input that consents to the performance of the process or operation is not deemed to be “material”.
The present subject matter may have a variety of modifications and may be embodied in a variety of forms, and specific embodiments will be described in more detail with reference to the drawings. It should be understood, however, that the embodiments of the present subject matter are not intended to be limited to the specific forms, but include all modifications, equivalents, and alternatives falling within the spirit and scope of the present subject matter.
FIG. 1 is a block diagram illustrating an electronic system for animal health management, in accordance with an embodiment of the present disclosure. With reference to FIG. 1, there is shown a block diagram that includes an electronic system 100 (herein after referred to as the system 100). The system 100 includes a hardware controller 104 and a memory 106. In an implementation, the system 100 further includes an artificial intelligence (AI) engine 108. The hardware controller 104 is communicatively coupled with the memory 106 and the AI engine 108 (when provided). The system 100 may be used to manage health of a plurality of farm animals facing different climatic condition or distributed in different geographical regions. Specifically, the system 100 identifies nutritional deficiencies and diagnose diseases for a target animal, by considering environmental factors. Further, the system 100 generates tailored nutrition recommendation for the target animal. In some examples, the system 100 may be used for cattle farming or other animal management or farming scanarios. Some other farming scenarios may include, but limited to, poultry farming and aqua farming.
In an implementation, the hardware controller 104, the memory 106, and the AI engine 108 may be implemented on a same server, such as a server 102. In some implementations, the system 100 further includes a storage device 110. In some implementations, the storage device 110 is configured to store a first database 112. In some implementations, the first database 112 may be stored in the same server, such as the server 102. The first database 112 includes a training dataset 114, a rumen fluid analysis (RFA) dataset 116, a biological marker dataset 118, and a photograph dataset 120 (e.g., thousands to millions of digital photographs of target animals). In some implementations, the storage device 110 is communicatively coupled to the server 102, via a communication network 122. The server 102 may be communicatively coupled to a plurality of user devices, such as a user device 124, via the communication network 122. A mobile application or a web platform may be accessible from the user device 124. The user device 124 includes a user interface 126.
The present disclosure provides the system 100 for animal health management, where the system 100 employs comprehensive knowledge graphs to ensure optimal health and well-being of farm animals across diverse geographical regions. By seamlessly integrating data collection, feature extraction, disease identification, and tailored nutrition recommendations, the system 100 offers a revolutionary approach to enhancing animal health monitoring and management. Through a user-friendly mobile application, farmers and veterinarians gain insights that transcend geographical boundaries, enhancing disease detection, deficiency management, and overall farm animal welfare. The system 100 empowers users with precise and actionable information, elevating animal health management to new levels of efficiency and effectiveness. The term “knowledge graph” refers to a structured representation of interconnected information and relationships among various entities, concepts, and attributes within a specific domain of knowledge.
The server 102 includes suitable logic, circuitry, interfaces, and code that may be configured to communicate with the user device 124 via the communication network 122. In an implementation, the server 102 may be a master server or a master machine that is a part of a data center that controls an array of other cloud servers communicatively coupled to it for load balancing, running customized applications, and efficient data management. Examples of the server 102 may include, but are not limited to a cloud server, an application server, a data server, or an electronic data processing device.
The hardware controller 104 refers to a computational element that is operable to respond to and processes instructions that drive the system 100. The hardware controller 104 may refer to one or more individual processors, processing devices, and various elements associated with a processing device that may be shared by other processing devices. Additionally, the one or more individual processors, processing devices, and elements are arranged in various architectures for responding to and processing the instructions that drive the system 100. In some implementations, the hardware controller 104 may be an independent unit and may be located outside the server 102 of the system 100. Examples of the hardware controller 104 may include but are not limited to, a hardware processor, a digital signal processor (DSP), a microprocessor, a microcontroller, a complex instruction set computing (CISC) processor, an application-specific integrated circuit (ASIC) processor, a reduced instruction set (RISC) processor, a very long instruction word (VLIW) processor, a state machine, a data processing unit, a graphics processing unit (GPU), and other processors or control circuitry.
The memory 106 refers to a volatile or persistent medium, such as an electrical circuit, magnetic disk, virtual memory, or optical disk, in which a computer can store data or software for any duration. Optionally, the memory 106 is a non-volatile mass storage, such as a physical storage media. Furthermore, a single memory may encompass and, in a scenario, and the system 100 is distributed, the hardware controller 104, the memory 106 and/or storage capability may be distributed as well. Examples of implementation of the memory 106 may include, but are not limited to, an Electrically Erasable Programmable Read-Only Memory (EEPROM), Dynamic Random-Access Memory (DRAM), Random Access Memory (RAM), Read-Only Memory (ROM), Hard Disk Drive (HDD), Flash memory, a Secure Digital (SD) card, Solid-State Drive (SSD), and/or CPU cache memory.
The AI engine 108 refers to software component or an integrated circuitry comprising AI engine that incorporates artificial intelligence model, such as machine learning model, a deep learning model, to perform complex tasks and make intelligent decisions based on input data. In some implementations, the AI engine 108 also utilizes computer vision algorithms that may identify and quantify visual attributes of an image.
The storage device 110 may be any storage device that stores data and applications without any limitation thereto. In an implementation, the storage device 110 may be a cloud storage, or an array of storage devices.
The first database 112 within the system 100 serves as a repository encompassing distinct datasets for the system's functionality. As discussed above, the first database includes the training dataset 114, the RFA dataset 116, the biological marker dataset 118, and the photograph dataset 120. The training dataset 114 includes a set of annotated images of animals, aiding learning process of the AI engine 108. The RFA dataset 116 contains rumen fluid analysis data from animals across varied geographical regions, contributing to the knowledge graph. The biomarker dataset 118 includes biological markers shared by users. The photograph dataset 120 includes image of the plurality of animals shared by the users, for example, farmers. Collectively, the datasets empower the hardware controller 104 to process data, extract features, and form the knowledge graph. The first database 112 facilitates the identification of potential health issues, assessment of animal conditions, and tailored nutrition recommendations, effectively enhancing animal well-being across diverse geographical contexts.
The communication network 122 includes a medium (e.g., a communication channel) through which the user device 124 communicates with the server 102. The communication network 122 may be a wired or wireless communication network. Examples of the communication network 122 may include, but are not limited to, Internet, a Local Area Network (LAN), a wireless personal area network (WPAN), a Wireless Local Area Network (WLAN), a wireless wide area network (WWAN), a cloud network, a Long-Term Evolution (LTE) network, a plain old telephone service (POTS), a Metropolitan Area Network (MAN), and/or the Internet.
The user device 124 refers to an electronic computing device operated by a user. In some implementations, the user device 124 may be configured to receive photographs of a target animal and data related to the biomarkers of the target animal. In some implementations, the nutritional deficiency or disease of the target animal is rendered over the user interface 126. Examples of the user device 124 may include but not limited to a mobile device, a smartphone, a desktop computer, a laptop computer, a Chromebook, a tablet computer, a robotic device, or other user devices.
In operation, the hardware controller 104 is configured to receive a set of biological markers and photographs of a plurality of target animals. In some implementation, the set of biological markers and photographs of the plurality of target animals are shared by the users, for example, the farmers via an application installed in their smartphones. The set of biological markers and photographs of the plurality of target animals are stored as the biomarker dataset 118 and the photograph dataset 120 within the first database 112. The target animal refers to the specific animals that show any signs of potential health issues. Such animals are further analysed to determine the underlying causes of their condition, such as deficiencies or diseases.
The hardware controller 104 is further configured to extract one or more features 128 for each target animal from the received set of biological markers and photographs. The one or more features 128 being extracted include various health-related data points or visual attributes that provide insights into the health condition of the plurality of target animals. The one or more extracted features 128 may include parameters such as heart rate, body temperature, and respiratory rate, which provide indicators of cardiovascular and respiratory health. Visual attributes like coat condition, body condition score, and posture contribute to assessing the animals' nutritional status and overall well-being. Additionally, biochemical markers like blood glucose levels and protein concentrations offer insights into metabolic and immune health. The one or more extracted features 128 further includes reproductive parameters, digestive metrics, behavioural patterns, milking performance metrics, feeding behaviour indicators, and immune function markers.
The hardware controller 104 is further configured to determine variation in the one or more extracted features 128 of each target animal. The variation in the one or more extracted features 128 is indicative of potential discrepancies or alterations within health conditions or visual attributes of each target animal. In some implementations, the AI engine 108 utilizes the computer vision algorithms to analyze the set of photographs of the plurality of target animals and determine variations in their visual attributes. In some implementations, in order to determine the variation in the one or more extracted features 128 of each target animal, the processor 104 is further configured to compare the extracted features 128 with the set of annotated images of the training dataset 114.
The hardware controller 104 is further configured to identify nutritional deficiencies or diseases in the plurality of target animals based on the determined variation in the one or more extracted features 128 of each target animal. In order to identify the nutritional deficiencies or diseases in the plurality of target animals, the processor 104 is further configured to compare the determined variations in the one or more extracted features 128 with predefined thresholds. In other words, to identify the nutritional deficiencies or diseases in the plurality of target animals, the determined variations are matched against the predefined thresholds or reference values that serve as benchmarks. The predefined thresholds are set based on established standards or guidelines for health conditions and attributes in animals. By comparing the variations with the predefined thresholds, the processor 104 assess whether the observed changes in features are significant enough to indicate a potential issue. If the variations surpass the predefined thresholds, the processor 104 signals the possibility of a nutritional deficiency or disease in the target animal. Such approach helps the system 100 efficiently pinpoint potential health concerns among the target animals by detecting deviations from expected norms.
In an example, imagine a dairy farm where the target animals are cows. The system 100 collects data on various features such as milk production, body weight, activity levels, and vital signs from these cows. One of the features is milk production, which is a key indicator of a cow's health and well-being. The system 100 has predefined thresholds for normal milk production based on the breed and age of the cows. If a cow's milk production falls significantly below the predefined threshold, the system would detect this variation and compare it to the threshold. If the deviation is substantial, the system 100 may flag it as a potential issue, indicating that the cow might have a nutritional deficiency or underlying health problem affecting its milk production. In this case, the system 100 compares the determined variation (lower milk production) in the extracted feature (milk production) with the predefined threshold to identify a potential nutritional deficiency or health issue.
The hardware controller 104 is further configured to form a knowledge graph based on the RFA dataset 116 from RFA conducted on a set of sample animals, considering influence of a set of environmental factors on health conditions or visual attributes of the set of sample animals. In some implementations, the set of environmental factors includes climatic conditions, geographical locations, and specific farm management methodologies. In some other implementations, the set of environmental factors may include, but not limited to, factors such as seasonal changes, housing conditions, feed quality, water availability and quality, pest and disease exposure, management practices, social interactions, transportation stress, and biosecurity measures. The sample animal refers to a representative group of animals taken from various geographical regions. The sample animals are chosen for the purpose of conducting the RFA to gather data about their health-related parameters. The RFA dataset 116 collected from the set of sample animals facilitates in creating the knowledge graph, which establishes relationships between health-related parameters, diseases, deficiencies, and interventions.
In some implementations, the knowledge graph includes relationships of each health-related parameter of a set of health-related parameters with a specific disease and a nutritional deficiency. Each health-related parameter within the knowledge graph is associated with the set of sample animals influenced by the set of environmental factors.
The hardware controller 104 is further configured to correlate the identified nutritional deficiencies or diseases with the relationships in the knowledge graph to remediate the identified nutritional deficiencies or diseases. The hardware controller 104 is further configured to generate tailored nutrition recommendations for each target animal based on the correlated nutritional deficiencies or diseases and the relationships in the knowledge graph.
For example, the system 100 has identified a nutritional deficiency of vitamin D in a target animal. The knowledge graph contains relationships between various health-related parameters, diseases, and deficiencies. In this case, the graph may show that a deficiency of vitamin D is associated with bone-related issues and reduced milk production. By correlating the identified deficiency with these relationships in the knowledge graph, the system 100 may recommend tailored interventions. For instance, the processor 104 may generate a tailored nutrition recommendation suggesting adjusting the animal's diet to include feed enriched with vitamin D or providing access to sunlight to stimulate natural vitamin D production.
In accordance with an embodiment, the hardware controller 104 is further configured to train the AI engine 108 on the training dataset 114 stored in the first database 112. The training dataset 114 includes the set of annotated images of animals. Each annotated image of the set of annotated images is linked with metadata to provide context for features of an animal corresponding to each annotated image.
In accordance with an embodiment, the hardware controller 104 is further configured to display the tailored nutrition recommendations through the user interface 126 accessible via mobile applications or web platforms. The tailored nutrition recommendations includes adjustments in feed composition, dosages of nutritional supplements, and specific dietary regimens customized for each target animal's health profile.
In accordance with an embodiment, the processor 104 is further configured to execute a feedback loop to continuously refine an accuracy parameter of the nutrition recommendation over time by detecting shifts or anomalies in the set of health-related parameters of the plurality of animals that emerges over a predefined time period.
By seamlessly integrating the receiving of the set of biological markers and photographs, and the knowledge graph generation, the system 100 revolutionizes animal health management in several ways. Firstly, the system 100 automates the process of receiving biological markers and photographs from users such as farmers, eliminating manual collection of such data and reducing human interference. Secondly, in some examples, the AI engine 108 identifies subtle variations in the extracted features 128, providing early indications of potential health issues that may go unnoticed through traditional observation. This predictive capability aids in prompt and effective disease detection. Thirdly, the knowledge graph, formed through comprehensive data aggregation and analysis, offers a holistic view of animal health conditions, diseases, and nutritional deficiencies across geographical regions. The comprehensive insight enables targeted interventions and preventive measures.
By correlating the identified nutritional deficiencies or diseases with knowledge graph relationships, the system 100 gains predictive capabilities. This allows for the anticipation of potential health issues based on patterns and trends, enabling proactive measures to prevent and manage diseases. Further, the system 100 generates tailored nutrition recommendations for each target animal based on its specific health conditions and the knowledge graph insights. This personalized approach ensures that interventions are precise and aligned with the individual needs of each animal. Furthermore, ability of the system 100 to process data remotely and provide insights through a mobile application facilitates real-time monitoring of animal health across geographical locations. This remote access enhances convenience and timely decision-making for farmers. In addition, The system 100 empowers farmers with data-driven insights, allowing them to make informed decisions regarding animal health and farm management. This leads to improved animal welfare, optimized resources, and enhanced productivity. Also, the feedback loop integrated into the system 100 refines the accuracy of nutrition recommendations over time. This continuous learning process ensures that the system 100 becomes increasingly effective in diagnosing and addressing animal health issues. Beneficially, the system 100 contributes to sustainable livestock practices by minimizing the economic losses associated with undiagnosed diseases, reducing the need for antibiotics, and promoting proactive health management.
FIG. 2 is a flowchart of operation that illustrates a process of identifying nutritional deficiencies or diseases in animals and generating nutrition recommendations for the animals, in accordance with an embodiment of the present disclosure. FIG. 2 is described in conjunction with elements from FIG. 1. With reference to FIG. 2, there is shown an exemplary flowchart 200 that illustrates the process of identifying nutritional deficiencies or diseases in animals and generating nutrition recommendations for the animals. The flowchart 200 includes a series of operations from 202 to 222.The hardware controller 104 (of FIG. 1) are configured to execute the operations shown in the flowchart 200.
At operation 202, samples of rumen fluid are taken from the stomachs of the set of sample animals in various geographical regions. The samples of rumen fluid are collected for analysis to gain insights into the health condition of the set of sample animals. Further, the collected rumen fluid samples undergo various analyses. The analyses include physical tests like colour and viscosity assessments, as well as chemical tests like pH testing. Then, genomic analysis is performed to determine the presence and quantity of different organisms in the rumen fluid. This helps in understanding the microbial composition and its potential impact on health of the set of sample animals. Furthermore, based on the results of the molecular and genomic analyses, parameters are generated for each geographical region. These parameters represent the health condition of the cattle in that region.
At operation 204, the rumen fluid samples are collected from the set of sample animals in a plurality of geographical regions. The parameters generated for each region are averaged to derive an overall health condition of each sample animal in that specific area. Further, using the aggregated data, relationships and interactions between diseases, deficiencies, and vitamins/minerals are identified. For instance, if rumen fluid analysis from the set of sample animals in a specific region indicates acidity, a relationship may be established that associates reduced dry matter intake with lower acidity levels. Similarly, if a deficiency of vitamin B12 is observed, a relationship could be formed with a specific product to address this deficiency.
At operation 206, using the data collected from the RFA and the relationships formed, the knowledge graph is constructed. The knowledge graph represents the interconnectedness between various factors, including cattle health conditions, diseases, deficiencies, and recommended interventions.
At operation 208, when a farmer suspects an issue with their animal's health, the farmers provide the system with the animal's bio markers, including vital statistics, milking data, and other health parameters. This information is collected and stored in a dedicated database for the specific animal.
At operation 210, the biological marker dataset 118 containing the bio markers of each animal and the photographs dataset 120 containing photographs of each animal are formed. The biological marker dataset 118 and the photograph dataset 120 is a useful source for analysis and diagnosis. Additionally, the training dataset 114 is formed, consisting of annotated photographs of animals. These annotations, such as body condition scores, serve as references for comparison and analysis. Some of data from the biological marker dataset 118 and the photograph dataset 120 may be used as the training dataset 114.
At operation 212, from the biological marker dataset 118 and the photograph dataset 120, the set of features 128 is extracted. The one or more extracted features 128 encompass vital signs, historical health data, milking patterns, and other pertinent attributes. Collectively, the one or more extracted features 128 create a detailed profile of the animal's health status.
At operation 214, the AI engine 108, configured to identify the nutritional deficiencies and diseases, is initiated. The AI engine 108 is trained on the training dataset 114 stored in the first database 112. The AI engine 108 employs advanced machine learning techniques, including potentially using computer vision, to compare the uploaded photographs of the target animal with the annotated healthy photographs. By analyzing similarities and deviations, the AI engine 108 pinpoints potential issues and diseases.
At operation 216, using the extracted features 128 and insights from the comparison, potential deficiencies or health problems in the target animal are identified. It correlates patterns and deviations from normal markers to determine the animal's condition.
At operation 218, after the deficiency and disease have been identified in operation 216, the next step involves correlating this information with the knowledge graph created in operation 206.
At operation 220, the deficiency or disease identified for a specific farm animal is compared with the relationships established in the knowledge graph. This involves understanding how the deficiency or disease relates to various factors such as animal health conditions, diseases, nutritional deficiencies, and interventions, if taken. Based on the relationships identified, the knowledge graph may provide relevant recommendations for addressing the identified deficiency or disease. Such recommendations are informed by the insights gained from the correlation between the deficiency/disease and the various factors in the knowledge graph.
At operation 222, using the insights from the knowledge graph, the hardware controller 104 generates the tailored nutrition recommendations for the specific farm animal. Such recommendations may involve adjustments to the animal's diet, the inclusion of specific vitamins or minerals, or even introducing specialized feed or premixes that align with the animal's requirements.
FIG. 3 is a flowchart of a method for animal health management, in accordance with an embodiment of the present disclosure. FIG. 3 is explained in conjunction with elements from FIGs. 1 and 2. With reference to FIG. 3, there is shown a flowchart of a method 300. The method 300 is executed at the server 102 (of Fig. 1). The method 300 may include steps 302 to 314.
At step 302, the method 300 includes receiving, by the hardware controller 104, the set of biological markers and photographs of the plurality of target animal from the first database 112. The set of biological markers and photographs of the plurality of target animal are received by the hardware controller 104 in a structured format. At step 304, the method 300 further includes extracting, by the hardware controller 104, the one or more features 128 for each target animal from the received set of biological markers and photographs. Ability of the hardware controller 104 to extract the one or more features 128 from the received data enhances efficiency by swiftly converting complex biological information into relevant features. This integration of data collection and feature extraction enhances the overall speed and accuracy of the animal health management process.
At step 306, the method 300 further includes determining, by the hardware controller 104, the variation in the one or more extracted features 128 of each target animal. The variation in the one or more extracted features 128 is indicative of potential discrepancies or alterations within the health conditions or visual attributes of each target animal. By assessing the variations in the one or more extracted features 128, the hardware controller 104 may effectively identify potential health issues or anomalies in animals, even those that may not exhibit obvious symptoms. This advanced capability enhances the accuracy and early detection of health concerns, providing a proactive approach to animal health management and enabling timely interventions for better overall welfare and productivity.
At step 308, the method 300 further includes identifying, by the hardware controller 104, the nutritional deficiencies or diseases in each target animal based on the determined variation in the one or more extracted features 128 of each target animal. Optionally, by leveraging the precise analysis of the one or more extracted features 128 by the AI engine 108, the hardware controller 104 may swiftly and effectively identify potential health issues, enabling timely interventions. This proactive approach enhances animal welfare and prevents health concerns from escalating, ultimately leading to improved overall farm management and reduced economic losses associated with untreated health problems.
At step 310, the method 300 further includes forming, by the hardware controller 104, the knowledge graph based on the RFA dataset 116 from RFA conducted on the set of sample animals distributed across the plurality of geographical locations. The knowledge graph includes the relationships of each health-related parameter of the set of health-related parameters with the specific disease and the nutritional deficiency. Each health-related parameter within the knowledge graph is associated with the set of sample animals distributed across the plurality of geographical locations. Beneficially, the knowledge graph forms an interconnected repository of health-related parameters, diseases, and deficiencies, allowing for a holistic understanding of animal health conditions. The utilization of real-world data from diverse regions ensures a comprehensive representation of various health scenarios, enhancing the accuracy and reliability of the knowledge graph's insights. This approach empowers the system 100 with a robust foundation for correlation and analysis, resulting in precise recommendations for addressing nutritional deficiencies or diseases in the plurality of target animals.
At step 312, the method 300 further includes correlating, by the hardware controller 104, the identified nutritional deficiencies or diseases with the relationships in the knowledge graph to remediate the identified nutritional deficiencies or diseases. At step 314, the method 300 further includes generating, by the hardware controller 104, the tailored nutrition recommendations for each target animal of the plurality of target animals based on the correlated nutritional deficiencies or diseases and the relationships in the knowledge graph.
By correlating the identified nutritional deficiencies or diseases with the relationships in the knowledge graph, the hardware controller 104 precisely matches the identified deficiencies or diseases with their corresponding relationships in the knowledge graph. This not only expedites the identification of potential causes but also enables the hardware controller 104 to propose more targeted and effective solutions for remediation. Ability of the method 300 to draw upon a comprehensive repository of interrelated health parameters, diseases, and deficiencies enhances the accuracy of its recommendations. This approach ensures that the generated remedies are well-informed and specifically tailored to the individual animal's health condition and unique circumstances. As a result, the method 300 optimizes the management of animal health, maximizes efficiency, and contributes to the overall well-being of farm animals.
The steps 302 to 314 are only illustrative and other alternatives can also be provided where one or more steps are added, one or more steps are removed, or one or more steps are provided in a different sequence without departing from the scope of the claims herein. Various embodiments and variants disclosed with the aforementioned system (such as the system 102) apply mutatis mutandis to the aforementioned method 300.

Comparative analysis of the electronic system 102 with existing systems
Test Feature Electronic system 102 Existing systems
False positive rate Low (2-5%) High (30 - >50%)
False negative rate Low (1-7%) High (30 - >50%)
Accuracy High (95-99.5%) Medium-low (50-75%)
Reliability High Medium-low
Complexity Low Medium-high
The false positive rate is the percentage of healthy animals that are incorrectly identified as sick or missed opportunity of identifying nutritional deficient. The disclosed electronic system 102 has a lower false positive rate than existing systems. The false negative rate is the percentage of sick animals or nutritional deficient animal that are incorrectly identified as healthy. The electronic system 102 has a lower false negative rate than existing systems. Accuracy is the percentage of animals that are correctly identified as sick or healthy and if identified sick nutritionally deficient, what is accuracy of the tailored nutrition recommendations. Reliability is the consistency of the system's performance over time. The electronic system 102 is more reliable than existing systems. Complexity is the difficulty of using and maintaining the system. The electronic system 102 is less complex than existing systems.
To ascertain the evaluation features of the electronic system 102 over existing systems, a comparative study was conducted. The study involved a representative sample of cattle farms (100 farms), with a variety of herd sizes across a plurality of different geographical locations in a country (India). The electronic system 102 and some existing systems were used to monitor the health of the animals on each farm. Data on false positive rate, false negative rate, accuracy, reliability, and complexity was collected. The researchers reviewed the veterinary records of all animals on the farms. For each animal, the researchers determined whether the system had identified the animal as sick or healthy, and what tailored nutrition recommendation was made. The researchers then compared the system's identification to the veterinarian's diagnosis. If the system had identified a healthy animal as sick, this was counted as a false positive. If the system had identified a sick animal as healthy, this was counted as a false negative. If the system made nutritional recommendation, it was fed to a first set of sample animals and not fed to a second set of sample animals. It was observed that when the nutritional recommendation by the electronic system 102 was fed, almost every time, the animals maintained their health. Further, when not fed to the second set of animals even after detection and recommendation by the electronic system 102, almost 40-50% in many farms turned sick within 3-4 months, and over 60-70% within 6-9 months period with low milk output.
Modifications to embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as "including", "comprising", "incorporating", "have", "is" used to describe and claim the present disclosure are intended to be construed in a non-exclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural. The word "exemplary" is used herein to mean "serving as an example, instance or illustration". Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments and/or to exclude the incorporation of features from other embodiments. The word "optionally" is used herein to mean "is provided in some embodiments and not provided in other embodiments". It is appreciated that certain features of the present disclosure, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the present disclosure, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable combination or as suitable in any other described embodiment of the disclosure.
, Claims:What is claimed is
1. An electronic system (100) for animal health management, the electronic system (100) comprising:
a hardware controller (104) configured to:
receive a set of biological markers and photographs of a plurality of target animals ;
extract one or more features (128) for each target animal from the received set of biological markers and photographs;
determine variation in the one or more extracted features (128) of each target animal, wherein the variation in the one or more extracted features (128) is indicative of potential discrepancies or alterations within health conditions or visual attributes of each target animal;
identify nutritional deficiencies or diseases in the plurality of target animals based on the determined variation in the one or more extracted features (128) of each target animal;
form a knowledge graph based on a rumen fluid analysis (RFA) dataset (116) from RFA conducted on a set of sample animals distributed across a plurality of different geographical regions, considering influence of a set of environmental factors on health conditions or visual attributes of the set of sample animals,
wherein the knowledge graph comprises relationships of each health-related parameter of a set of health-related parameters with a specific disease and a nutritional deficiency, wherein each health-related parameter within the knowledge graph is associated with the set of sample animals influenced by the set of environmental factors;
correlate the identified nutritional deficiencies or diseases with the relationships in the knowledge graph to remediate the identified nutritional deficiencies or diseases; and
generate tailored nutrition recommendations for each target animal of the plurality of target animals based on the correlated nutritional deficiencies or diseases and the relationships in the knowledge graph.

2. The electronic system (100) as claimed in claim 1, wherein the hardware controller (104) is further configured to train an AI engine (108) on a training dataset (114) stored in a first database (112), and wherein the training dataset (114) comprises a set of annotated images of animals, each annotated image of a set of annotated images linked with metadata to provide context for features of an animal corresponding to each annotated image.

3. The electronic system (100) as claimed in claim 1 or 2, wherein the determination of the variation in the one or more extracted features (128) of each target animal comprises execute the artificial intelligence (AI) engine (108) to detect a variation pattern in the one or more extracted features (128) of each target animal based on a geographical region of the plurality of different geographical regions of each target animal.

4. The electronic system (100) as claimed in any of claims 1-3, wherein, in order to determine the variation in the one or more extracted features (128) of each target animal, the hardware controller (104) is further configured to compare the extracted features (128) with the set of annotated images of the training dataset (114).

5. The electronic system (100) as claimed in claim 1, wherein, in order to identify the nutritional deficiencies or diseases in the plurality of target animals, the hardware controller (104) is further configured to compare the determined variations in the one or more extracted features (128) with predefined thresholds.

6. The electronic system (100) as claimed in claim 1, wherein the hardware controller (104) is further configured to execute a feedback loop to continuously refine an accuracy parameter of the nutrition recommendation over time by detecting shifts or anomalies in the set of health-related parameters of the plurality of target animals that emerges over a predefined time period.

7. The electronic system (100) as claim in claim 1, wherein the one or more extracted features (128) comprises visual characteristics, vital signs, biochemical markers, digestive metrics, reproductive insights, behavioural patterns, milking performance metrics, feeding behaviour indicators, and immune function markers.

8. The electronic system (100) as claimed in claim 1, wherein the AI engine (108) utilizes computer vision algorithms to analyze the set of photographs of the plurality of target animals and determine variations in visual attributes of the plurality of target animals.

9. The electronic system (100) as claimed in claim 1, wherein the hardware controller (104) is further configured to control display of the tailored nutrition recommendations through a user interface (126) accessible via mobile applications or web platforms, and wherein the tailored nutrition recommendations comprises adjustments in feed composition, dosages of nutritional supplements, and specific dietary regimens customized for each target animal's health profile.

10. The electronic system (100) as claimed in claim 1, wherein the set of environmental factors comprises climatic conditions, geographical locations, and specific farm management methodologies.

11. A method (300) implemented in an electronic system of animal health management, the method (300) comprising:
receiving, by a hardware controller (104), a set of biological markers and photographs of a plurality of target animal from a first database (112);
extracting, by the hardware controller (104), one or more features (128) for each target animal from the received set of biological markers and photographs;
determining, by the hardware controller, variation in the one or more extracted features (128) of each target animal, wherein the variation in the one or more extracted features (128) is indicative of potential discrepancies or alterations within the health conditions or visual attributes of each target animal;
identifying, by the hardware controller (104), nutritional deficiencies or diseases in each target animal based on the determined variation in the one or more extracted features (128) of each target animal;
forming , by the hardware controller (104), a knowledge graph based on a rumen fluid analysis (RFA) dataset (116) from RFA conducted on a set of sample animals distributed across a plurality of different geographical locations,
wherein the knowledge graph comprises relationships of each health-related parameter of a set of health-related parameters with a specific disease and a nutritional deficiency, wherein each health-related parameter within the knowledge graph is associated with the set of sample animals distributed across the plurality of geographical locations;
correlating, by the hardware controller (104), the identified nutritional deficiencies or diseases with the relationships in the knowledge graph to remediate the identified nutritional deficiencies or diseases; and
generating, by the hardware controller (104), tailored nutrition recommendations for each target animal of the plurality of target animals based on the correlated nutritional deficiencies or diseases and the relationships in the knowledge graph.