AI for Animal Health

We worked with Vaxxinova on a solution for analyzing and classifying the parasite that causes avian coccidiosis using Artificial Intelligence.

Avian coccidiosis causes annual losses exceeding
$14.3
billion

Case Context

Avian coccidiosis is one of the most economically impactful diseases in poultry farming, causing global industry losses of over $14.3 billion each year.

It is caused by seven species of the protozoan from the Eimeria genus, which infect birds through oocysts—small, resistant organisms that spread in the environment and can be ingested by birds.

Vaxxinova’s role is to provide solutions for the prevention and control of avian coccidiosis through vaccines. This production requires strict controls and processes, including the analysis of coccidiosis oocysts.

Project Challenges

The analysis of each sample is quite complex. It requires counting and classifying the oocysts and identifying the Eimeria species to determine the appropriate control strategy.

The process is time-consuming, demands a high level of expertise, and is prone to errors, which can compromise the accuracy of the analysis.

  • Complex analysis
  • Manually performed
  • Prone to errors

Developed Solution

In partnership with Vaxxinova, we developed an advanced Artificial Intelligence model with Computer Vision to perform the analysis and counting of oocysts.

The model identifies patterns and is capable of detecting, classifying, and quantifying the oocysts. It can differentiate between:

  • Sporulated (infectious) and non-sporulated (non-infectious) oocysts.
  • Eimeria species, identifying the seven species responsible for avian coccidiosis - something the human eye cannot visually distinguish.

With each new sample, the model learns more and becomes even more accurate.

Technologies Used

Artificial Intelligence
Machine Learning
Computer Vision
Deep Learning
Captura de imagem de oocistos
The solution captures the different oocyst species, represented here in various colors.
Captura de imagem e diferenciação dos oocistos
In addition to species, the solution also distinguishes between infectious (blue) and non-infectious (red) oocysts.

Results

The process is now automated
Reduced sample analysis time
Increased accuracy in each analysis
Support for vaccine production and quality control
See other success stories

Testimonials

Jony Yoshida, Coordenador de Suporte Técnico em Bioprocessos
Jony Yoshida
Technical Support Coordinator in Bioprocesses

The project with Venturus was innovative in an area previously seen as unfeasible or requiring complex and costly solutions.

Even without prior experience in animal health, Venturus’ technological expertise, combined with our technical knowledge, allowed us to develop a high-value solution.

Beyond the practical results, this collaboration paves the way for future innovations, strengthening the transformative potential in our field.

Sandra Fernandez, diretora global do programa coccidiose
Sandra Fernandez
Global Director of the Coccidiosis Program

The partnership between the two companies was essential to the success of this project, which brought innovation and improvements to our production process.

We developed a fast and accurate method with high applicability, representing a significant advancement in the field.

The solution offers possibilities for use in various future contexts, reinforcing the strategic value of the collaboration and opening doors for new applications.