Researchers at the University of Illinois Urbana-Champaign announced on Apr. 27 that they have developed new methods using near infrared and hyperspectral imaging to predict chick embryo mortality and determine the sex of embryos in chicken eggs before hatching.
These findings could lead to more efficient, safe, and humane poultry production by addressing longstanding challenges in hatchery operations such as embryo mortality, fertility assessment, sex determination, and evaluation of eggshell characteristics. The research team published a series of studies describing their work with these imaging technologies.
Lead author Md. Wadud Ahmed said, “If there is a genetic disorder or other inherent issue, some eggs don’t produce healthy chicks, and the embryo dies. This poses a health hazard, as dead embryos can harbor bacteria. If we can detect and remove them early in the incubation period, we can avoid biosecurity issues.” Hatcheries currently test for mortality by shining light through each egg—a process that requires significant time and resources.
In their experiments with 300 chicken eggs from the university’s poultry farm incubated under commercial conditions, researchers used hyperspectral cameras to capture images both before incubation and after four days. They found that machine learning models built from these images could identify dead versus living embryos with up to 97% accuracy on day four.
Another study focused on early sex determination to address animal welfare concerns around culling male chicks—an estimated six billion are culled annually in the United States alone because they are not economically viable for meat or egg production. “Male chicks are considered a byproduct because they don’t lay eggs and they are not economically feasible for meat production… If we can identify the embryos early, we can avoid the culling of males,” Ahmed said. Their model achieved about 75% accuracy at day zero (early incubation) when classifying male versus female embryos.
The team also investigated non-destructive methods for measuring other egg characteristics like shell strength using near infrared spectroscopy (NIR), which is less costly than hyperspectral imaging but does not provide sufficient molecular detail for all assessments. Mohammed Kamruzzaman said: “Conventional testing methods are destructive; for example, to measure shell strength you need to break the eggs… With NIR and HSI, we do not need to destroy the eggs.”
Kamruzzaman added that automation will be key if hatcheries adopt these techniques: “We are working on developing a system with a robotic arm that can separate the eggs… after identification.” He noted potential applications beyond poultry: “NIR and HSI technology have applications in agriculture, food, environment, and biomedicine… I think implementing it could be very useful for the industry’s processing or farm side.”
The researchers have made datasets available publicly for further research use.
According to the official website, The University of Illinois College of Agricultural, Consumer and Environmental Sciences allocates $3.8 million annually in scholarships supporting students; promotes social well-being through nutrition programs; utilizes its own agricultural experiment station; maintains high retention (96%) and graduation rates (78% over four years); provides more than 400 study abroad opportunities; and aims to advance knowledge globally across agriculture-related fields.
