2024-12-12 –, Mission 1
In Precision Livestock Farming (PLF), deep learning-based approaches are increasingly employed to study animal behavior on farms. These behavioral studies enable animal phenotyping, which can be used for genetic selection and social network analysis in large animal groups. While considerable attention is often given to models in the deep learning field, the models themselves do not function in isolation. Efficient deep-learning workflows require systems that bridge research and prototyping with production operations.
We introduce the IMAGEN Data Analytics Platform under the IMAGEN program, which brings research into animal behavior together with computer science to improve the health and welfare of pigs and lay hens and reduce the ecological footprint of food production. This platform on the Surf HPC cluster supports the development of deep learning models for animal phenotyping and addresses various data challenges through a DataOps approach. Although initially designed for the animal phenotype detection domain, the platform is domain-neutral and can be applied to similar cases in other application domains.
In the context of animal breeding, genomic selection is the process of estimating the future performance of breeding individuals for the traits of interest, such as pecking, tail biting, and temperament, via a large number of markers distributed across the whole genome.
Various barriers impede the adoption of data-driven efforts in genomic selection programs. Apart from developing new algorithms and techniques to mine and analyze the data, a data analytics platform is needed for efficient data storage, meta-data management, data analysis, and data serving.
By implementing data-driven animal welfare measures in the livestock housing facilities, the efficiency of the livestock-based protein production systems can be improved significantly.
To further the research and innovation efforts in the field of precision livestock farming, the larger IMAGEN (AnIMAl Group sENsor - Integrating behavioural dynamics and social genetic effects to improve health, welfare and ecological footprint of livestock (IMAGEN) | NWO) and SmartTurkeys (Smart Turkeys, NWO Open Technology program (breed4food.com)) programs were initiated under the research grants from the Netherlands Organization for Scientific Research (NWO).
Both programs use multi-modal big data such as videos, sensors, and RFID readers to understand animal social behavior better and link it to individual genes. For this purpose, the IMAGEN Data Analytics Platform is being developed to detect and study animal behaviors.
The Surf: Spider High-performance computing (HPC) cluster-based IMAGEN Data Analytics Platform through the PiCas framework is a versatile solution designed for data scientists, researchers, and engineers. It comprises three core components:
• Intuitive User Interface: IMAGEN Data Analytics Platform’s user-friendly interface simplifies access to platform features. Whether you’re an expert or a novice, the UI abstracts complexities, allowing seamless task submission, compute configuration, and progress monitoring. Role-based access control enhances security.
• Integration with a Pilot Job Framework: Traditional job submission methods can lead to inefficiencies. IMAGEN Data Analytics Platform addresses this by encapsulating smaller components within a Pilot-Job. This decouples task submission from resource assignment, streamlining execution and simplifying job management.
• DataOps Principles: Automation is key in modern data analytics. IMAGEN Data Analytics Platform implements automated data pipelines for data movement, quality assurance, and visualization. These pipelines enhance efficiency and empower researchers in their work.
Enkhzol Dovdon is a postdoctoral researcher in the Mathematics and Computer Science department at Eindhoven University of Technology, specializing in Applied Data Science. His research focuses on the intersection of data-driven animal science and technology, leveraging AI models for the IMAGEN program to enhance livestock health, welfare, and ecological footprint. The program aims to advance animal well-being and sustainable practices through innovative software solutions.
With over 16 years of experience in IT projects, research, and teaching, he has worked as a Software Engineer and researcher at universities and companies. He has led several software and system development projects. His expertise includes high-performance computing (HPC) clusters, public cloud platforms, and AI models.