Use of Infrared Imaging to Predict the Developmental Stages and Differences in Chicken Embryos Exposed to Different Photoperiods
Conference Paper
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
Monitoring development of chicken embryos allows determination of when an egg is not developing and when eggs are close to hatching for more efficient production. Research has been conducted on the effects of temperature fluctuations and light exposure on embryo development; similarities between chicken and mammal embryos; and the use of MRI, tomography, and ultrasound to view specific areas and processes within the embryo. However, there has been little exploration of the use of infrared imaging as a non-destructive method for analyzing and predicting embryonic development. In this study, we built an automated loading system for image acquisition. Pilot experiments were conducted to determine the overall scanning time and scanning frequency. A batch of fertilized eggs was scanned each day as the embryos continued to grow. The captured images were analyzed and categorized into three stages: Stage 1 (days 1 to 7), Stage 2 (days 8 to 14), and Stage 3 (days 15 to 21). The temperature data abstracted from the captured images were divided into two groups. Group 1, consisting of two-thirds of the data, was used to construct a model. Group 2, consisting of one-third of the data, was used to evaluate the predictive accuracy of the model. A three-layer artificial neural network model was developed to predict embryo development stage given a temperature profile. Results suggest that the neural network model is sufficient to predict embryo development stage with good accuracy of 75%. Accuracy can likely be improved if more data sets for each development stage are available. 2012 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).
