Chick sexing method offers precision, scalability

In recent years, the poultry industry has been increasingly interested in determining gender of newly hatched chicks so birds can be sorted into separate sex groups for rearing. However, at hatch, accurately sexing chicks is difficult as both sexes exhibit similar characteristics until approximately one month of age.

Traditionally, chick sexing has been based on feather and cloaca examination, but these methods have a low accuracy rate due to human error and inter-operator variability and can require years of training. Recently, researchers have explored other methods to examine the internal anatomy of chicks, but these methods have been found to have limited effectiveness.

A group of researchers in Thailand recently studied a novel approach to chick sexing that integrates Optical Coherence Tomography (OCT) and deep learning to enable high-resolution, non-invasive gender determination. OCT provides micrometer-scale visualization of cloacal structures, allowing precise differentiation between male and female chicks based on internal anatomical markers.

These researchers described their approach in a recent edition of Poultry Science.

The OCT imaging system captures high-resolution images of cloacal tissue structures, providing detailed anatomical insights for classification. Laboratory and field experiments demonstrated that these images provided sufficient detail for accurate gender differentiation. The OCT system was designed for portability, cost-efficiency, and compactness, making it suitable for field applications.

To minimize human error and reliance on skilled personnel, the researchers integrated OCT with computer vision and deep learning techniques to automate the classification step.

Their approach achieved 79% accuracy, comparable to traditional vent sexing, but with added advantages of automation, reduced processing time, and scalability for high-throughput hatcheries.

The researchers noted that the OCT-provided images were crucial for distinguishing subtle differences between male and female cloacas. Also, the non-invasive method offers a reliable alternative to vent sexing, reducing the risk of error and the need for skilled personnel. Since this process can be automated, there are significant implications for scalability and efficiency in poultry farming.

    

The full paper, “Automated chick gender determination using optical coherence tomography and deep learning,” can be found in Poultry Science and online here.

DOI: 10.1016/j.psj.2025.105033