Skip to main content
ARS Home » Research » Publications at this Location » Publication #155549
Title: SPECTROSCOPIC DETECTION OF ABNORMALITY IN CHICKEN LIVER AS AN INSPECTION TOOL

Author
item DEY, B - USDA, FSIS, WASH.,DC
item Chan, Diane
item Chen, Yud
item GWOZDZ, FRANK - USDA, FSIS, WASH. DC

Submitted to: Proceedings of SPIE
Publication Type: Proceedings
Publication Acceptance Date: 10/29/2003
Publication Date: 12/1/2003
Citation: Dey, B.P., Chan, D.E., Chen, Y.R., Gwozdz, F.B. 2003. Spectroscopic detection of abnormality in chicken liver as an inspection tool. SPIE Proceedings entitled: Monitoring Food, Safety, Agriculture, and Plant Health. 5271:43-50.

Interpretive Summary: Food safety standards under the HACCP-based Inspection Models Project (HIMP) of the USDA Food Safety and Inspection Service require that all chickens with the septicemia/toxemia (septox) condition be condemned as unsafe for human consumption. Septox is a systemic condition in which pathogenic microorganisms, or their toxins, are present in the bloodstream, and the liver will show identifying features of this condition. A study at the USDA, ARS Instrumentation and Sensing Lab in Beltsville, MD, collected visible/near-infrared spectral data from 300 fresh chicken livers, half taken from normal chickens and half taken from birds showing the septox condition. A neural network model obtained 98% and 94% accuracies for classifying the normal and septox livers, respectively. The high rate of separation achieved shows that a visible/near-infrared spectroscopic method such as this could be very useful as a screening. This information is useful to the FSIS, and to poultry equipment manufacturers and processing plants.

Technical Abstract: Successful differentiation of normal chicken livers from septicemic chicken livers was demonstrated using visible/near-infrared (Vis/NIR) spectral data subjected to principal component analysis and then fed into a feed-forward back-propagation neural network. The study used 300 fresh chicken livers, 150 collected from normal chicken carcasses and 150 collected from chicken carcasses diagnosed with the septicemica/toxemia (septox) condition as defined for condemnation under U.S. Department of Agriculture (USDA) standards for food safety. Using a training set of 200 samples and testing set of 100 samples, the best neural network model demonstrated a classification accuracy of 98% for normal samples and 94% for septicemia/toxemia samples. These results show that Vis/NIR spectral methods have potential for use in chicken liver inspection as part of automated online systems for food safety inspection. Liver abnormalities are identifying characteristics of the septox condition; consequently, liver screening would be extremely useful as part of an automated inspection system to meet USDA food safety requirements for poultry. Automated inspection systems capable of real-time on-line operation are currently being developed, and spectroscopic liver inspection is potential tool that could be implemented as part of such systems to help poultry processors increase production while meeting food safety inspection requirements.