Portrait photo of Oleksiy Guzhva

Oleksiy Guzhva

Assistant Professor, Department of Biosystems and Technology
Phone
+4640-41 50 58
I develop responsible and sustainable AI solutions that integrate computer vision, sensors, and data science to enhance animal welfare and improve farming efficiency.

Presentation

As an assistant professor and associate senior lecturer at the Department of Biosystems and Technology (Swedish University of Agricultural Sciences – SLU), I am passionate about responsible innovation and sustainable development of AI in agriculture. My work combines sensor technology, computer vision, the Internet of Things (IoT), digital agriculture, and smart farm buildings to create farming systems that are both productive and ethically sound. With a veterinary degree (DVM), an MSc in animal science focusing on dairy production and welfare, and a PhD that explored sensor and computer‑vision methods for dairy cattle behaviour, I rely on both scientific training and practical farm experience.

My Focus

I see Precision Livestock Farming (PLF) as a tool for responsible innovation. Through continuous animal monitoring, data‑driven decision‑making and ethically aligned technology, I strive to enhance animal welfare and resource efficiency. I develop computer vision and deep learning algorithms for smart farming across animal and plant systems, and work on big data, IoT, and data infrastructure to provide farmers with actionable insights. I also apply data science – from exploratory analysis to multivariate prediction modelling – to support sustainable, transparent AI solutions. Project management, product development, and biosecurity strategies are integral parts of my approach, and I design animal-oriented facilities that promote health and well-being.

 

Research

Ongoing projects:

MINI-Moo – Modular, Intelligent Node for In-Field Assessment of Grazing Cattle Wellbeing 

Our vision is to create a solution for continuous monitoring of cattle wellbeing and pasture management that is modular, scalable, non-dependent on large quantities of manually annotated data and, of foremost importance, highly relevant for end-users and their unique challenges. Our solution aims to reduce the randomness of animal inspections and provide earlier indications of potential health issues in any animal, regardless of circumstances. We will use a highly customizable inertial measurement unit (IMU) and a Global Navigation Satellite Systems (GNSS) receiver to collect data relevant to on-the-go decision-making. Additionally, our solution will address the issue of animals escaping from enclosures or pastures. Our system will facilitate this process by tracking the animals' movement history and marking the potential area of the breach in the user-friendly application, reducing the need for manual inspection. 

Keel under pressure (KUP)

Keel bone damage (KBD) is one of the most critical welfare concerns for laying hens in modern egg production. Despite ongoing research, significant knowledge gaps remain regarding the underlying causes of this multi-factorial problem.

Our primary objective is to enhance knowledge that will improve keel bone health in hens. Specifically, we aim to determine whether offering a variety of perch types during rearing has a positive impact on keel bone health, investigate the development and welfare consequences of keel bone deviations, and develop a non-invasive method to detect KBD by assessing the birds' movement in a 3D space.

This project aims to address these gaps by (1) investigating the effect of varied perch types, offering a new perspective on the link between housing conditions and KBD, (2) studying the underexplored issue of keel bone deviations, and (3) utilising markerless computer vision (CV) technology to gather detailed biomechanical data on laying hens. The knowledge obtained in the experimental setting will also be applied to a commercial layer farm.

By introducing novel approaches to both perch provision and the use of cutting-edge computer vision, this project will contribute to better animal welfare and more sustainable egg production.

OinkScope – A Computer Vision-based Spatial Intelligence tool for Monitoring Behaviour of Group-Housed Pigs

Video recordings of ethological experiments generate a large volume of footage that must be manually annotated to extract the data. This often creates a critical bottleneck in data processing, limiting the time available for analysis and delaying the reporting of results. To meet the growing demand for accurate automated behaviour monitoring, we present a new tool for pig behaviour monitoring. OinkScope is a novel computer vision (CV) algorithm that integrates real-time object detection, customizable region-of-interest (ROI) monitoring, and automated heatmap analysis. At its core, OinkScope relies on the Detectron2 object detection and segmentation framework built on the PyTorch ecosystem.  By uniting customizable ROIs, scalable deep learning models, and an intuitive GUI, OinkScope provides insights into group dynamics, crowding patterns, and resource utilisation. Ultimately, it offers a powerful yet accessible toolkit for automatic and real-time behaviour analysis — showcasing the transformative potential of modern CV systems in ethological research.

Aside from being a Principal Investigator, I also contribute my computer vision and AI modelling expertise as an invited expert in several projects. These collaborations span across SLU as well as national and international universities and research institutes.

Teaching

Teaching and extension are key parts of my mission. I teach animal physiology, health and welfare, biosecurity and production management, and I am the course leader for Precision Livestock Farming for Sustainable Production (TN0356). I supervise undergraduate, MSC and PhD students and work closely with farmers, companies and advisers to ensure that research is translated into sustainable AI applications and best practices. 

Extension Work

By leading proof-of-concept projects and building cross-disciplinary partnerships, I help integrate ethical AI tools into real-world farming systems.

Background

Education: I hold a PhD in veterinary medicine (SLU, 2014–2018), an MSc in animal science (SLU, 2011–2013) and a DVM from Poltava State Agrarian Academy (Ukraine, 2004–2010).

Professional experience: I have been a researcher and associate senior lecturer at SLU since 2018. I also work as a Subject Area Coordinator at SLU's collaboration platform Partnership Alnarp, with responsibility for animal production systems. 

Contact

Position: Assistant Professor/Associate Senior Lecturer, Department of Biosystems and Technology, SLU.
Work: +46703869577
Mobile: +46729328191
Email: oleksiy.guzhva@slu.se
Postal address: Biosystems and Technology, Box 190, 234 22, Lomma, Sweden.
Visiting address: Sundsvägen 14, Alnarp.