Submitted to: International Journal of Poultry Science
Publication Type: Peer reviewed journal
Publication Acceptance Date: 10/27/2000
Publication Date: 11/30/2008
Citation: Smith, D.P., Lawrence, K.C., Heitschmidt, G.W. 2008. Fertility and embryo development of broiler hatching eggs evaluated with a hyperspectral imaging and predictive modeling system. International Journal of Poultry Science. 7(10):1001-1004. Interpretive Summary: The ability for chicken egg hatcheries to cull infertile eggs that will never hatch (approx. 10% of all eggs) prior to incubation would be advantageous in several ways. They could either maximize efficiency by loading incubators to full capacity (using fewer incubators) or load the fertile-only eggs into incubators with less crowding, which improves hatchability. With either method, fewer “exploder” eggs (typically infertile of early dead eggs contaminated with bacteria and/or mold, sometimes pathogenic strains) would be in the incubators, so final chick quality would improve. Therefore an imaging system was adapted to screen fertile eggs and a computer analysis program was set up to process the data and predict initial fertility and early embryo development. After testing 192 eggs with 4 images each (768 images total) it was determined that the predictive model as constructed was unable to accurately detect fertility and development. Further work on the analysis portion of this project will be required to achieve results accurate enough to be used by the industry.
Technical Abstract: A hyperspectral imaging system and a predictive modeling technique was evaluated for determining fertility and early embryo development of broiler chicken hatching eggs. Twenty-four broiler eggs were collected (12 fertile, 12 infertile) for each of 8 replicate trials (n=192) and imaged on Days 0, 1, 2, and 3 of incubation for training and model validation. Three replications of 30 eggs each (fertile and infertile eggs randomly mixed) were collected and imaged as above for model verification (n=90). Eggs were backlit and positioned below and vertical to the imaging system (lens, spectrograph, and CCD camera). Spatial and spectral data from approximately 400 to 1000 nm were collected for each egg on each day of incubation with refinement to 550 to 899 nm. A Mahalanobis Distance (MD) supervised classifier was trained with spectral data from the first 5 replicate sets of eggs, then Principal Component Analysis (PCA) was performed. This model was applied to the next 3 sets for model validation and then to the three 30 egg sets for verification. Fertility was confirmed on Day 5 of incubation by candling and breakout. The MD/PCA model predictions for the 3 validation sets of eggs were: 71% accuracy for Day 0; 63% for Day 1, 65% for Day 2, and 83% for Day 3. For the 3 sets of verification eggs, the MD/PCA model accurately predicted 46/90 on Day 1 and 45/90 on Day 3. The data indicate that the particular MD/PCA model used is not appropriate for predicting fertility and early development.