Bayesian network as an aid for Food Chain Information use for meat inspection.

Current ante mortem inspection involves a check of relevant Food Chain Information (FCI) transmitted by the farmer to the slaughterhouse on a regulatory FCI document. Since 2000, a farm sanitary form with FCI data has been used for all consignments of broiler chickens in France. However, the FCI needs to be standardized for the collection and interpretation of data. A study was conducted to develop an expert system, undertaken to elaborate on a simple decision support system capable of predicting whether the flocks will present a high condemnation risk, based on FCI. For this, 'optimal' (i.e. on-farm survey data) and 'worthy' (i.e. farmers' declaration on existing farm sanitary form) data quality conditions were considered to estimate the lower and upper reference bounds of the confidence that the decision-makers could have in such a tool. Chicken broiler flocks (404) were randomly selected in 15 slaughterhouses located in Western France in 2005. Condemnation proportion and farm sanitary form were collected for each selected flock. Information about health history and technical performances were also specifically collected on farm. Condemnation risk category was modelled from the on-farm collected information, using a Bayesian network and assuming this represented the optimal data quality conditions. Corresponding information declared by the farmer on the existing farm sanitary form was secondly used in the network to evaluate the impact of the uncertainty of such information on the condemnation classification obtained with the expert system. The learnt Bayesian network had 16 explanatory variables pertaining to technical characteristics and sanitary features of the flock. Using a threshold of 1% of condemned carcases to define high risk, the network sensitivity and specificity were 55% and 93%, respectively, corresponding to positive and negative predictive values of 70% and 87%. When declared existing information was used in the network, the sensitivity and specificity were 16% and 96%, respectively, corresponding to positive and negative predictive values of 57% and 80%. Results suggested that the predictive network developed may be insufficient for correctly classifying chicken flocks for targeting of management procedures, and in its current form, the expert system may be unlikely to be implemented in the field. However, it could help to improve the standardization of both form design and FCI interpretation at a national level.

[1]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[2]  D Herenda,et al.  Poultry abattoir survey of carcass condemnation for standard, vegetarian, and free range chickens. , 1994, The Canadian veterinary journal = La revue veterinaire canadienne.

[3]  C Lupo,et al.  Postmortem condemnations of processed broiler chickens in western France , 2008, Veterinary Record.

[4]  T Habtemariam,et al.  A computer based decision-making model for poultry inspection. , 1983, Journal of the American Veterinary Medical Association.

[5]  A Windhaus,et al.  [Results of the evaluation of food chain information for a risk-oriented meat inspection]. , 2007, DTW. Deutsche tierarztliche Wochenschrift.

[6]  A S Bremner.  Post mortem condemnation returns from poultry slaughterhouses in England and Wales , 1994, Veterinary Record.

[7]  S. Brown,et al.  Prevalence and factors associated with it, of birds dead on arrival at the slaughterhouse and other rejection conditions in broiler chickens , 2008, British poultry science.

[8]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[9]  J M Snijders,et al.  Use of information from pig finishing herds for meat inspection purposes. , 1992, The Veterinary quarterly.

[10]  Lüppo Ellerbroek.  Lebensmittelketteninformationen und visuelle Fleischuntersuchung : Kriterien zur Interpretation, Entscheidung und Durchführung , 2006 .

[11]  A. Pointon,et al.  Assessment of producer conducted antemortem inspection of market pigs in Australia. , 2006, Australian veterinary journal.

[12]  Finn V. Jensen,et al.  Bayesian Networks and Decision Graphs , 2001, Statistics for Engineering and Information Science.

[13]  Finn Verner Jensen,et al.  Introduction to Bayesian Networks , 2008, Innovations in Bayesian Networks.

[14]  C Lupo,et al.  Feasibility of screening broiler chicken flocks for risk markers as an aid for meat inspection , 2009, Epidemiology and Infection.

[15]  J. R. Landis,et al.  The measurement of observer agreement for categorical data. , 1977, Biometrics.

[16]  Linda J. Keeling,et al.  Scientific Opinion on the public health hazards to be covered by inspection of meat (swine) , 2011 .

[17]  C Lupo,et al.  Risk factors for sanitary condemnation in broiler chickens and their relative impact: application of an original multiblock approach , 2009, Epidemiology and Infection.

[18]  Maryam Ansari-Lari,et al.  Poultry abattoir survey of carcass condemnations in Fars province, southern Iran. , 2007, Preventive veterinary medicine.

[19]  Lüppo Ellerbroek.  Risk based meat hygiene--examples on food chain information and visual meat inspection. , 2007, DTW. Deutsche tierarztliche Wochenschrift.

[20]  F I Lewis,et al.  Structure discovery in Bayesian networks: an analytical tool for analysing complex animal health data. , 2011, Preventive veterinary medicine.

[21]  Nir Friedman,et al.  Bayesian Network Classifiers , 1997, Machine Learning.

[22]  Jacob Cohen.  A Coefficient of Agreement for Nominal Scales , 1960 .

[23]  K. Stärk,et al.  [Food safety achieved through herd management]. , 2000, Schweizer Archiv fur Tierheilkunde.

[24]  A. Meek,et al.  Farm management risk factors associated with cellulitis in broiler chickens in southern Ontario. , 1996, Avian diseases.

[25]  Daniel B. Mark,et al.  TUTORIAL IN BIOSTATISTICS MULTIVARIABLE PROGNOSTIC MODELS: ISSUES IN DEVELOPING MODELS, EVALUATING ASSUMPTIONS AND ADEQUACY, AND MEASURING AND REDUCING ERRORS , 1996 .

[26]  D. Altman,et al.  STATISTICAL METHODS FOR ASSESSING AGREEMENT BETWEEN TWO METHODS OF CLINICAL MEASUREMENT , 1986, The Lancet.

[27]  S. Lauritzen.  The EM algorithm for graphical association models with missing data , 1995 .

[28]  A. Beckett,et al.  AKUFO AND IBARAPA. , 1965, Lancet.

[29]  C Lupo,et al.  Risk and indicators of condemnation of male turkey broilers in western France, February-July 2006. , 2010, Preventive veterinary medicine.

[30]  A Steiner,et al.  Data quality of animal health records on Swiss dairy farms , 2008, Veterinary Record.

[31]  Wai Lam,et al.  LEARNING BAYESIAN BELIEF NETWORKS: AN APPROACH BASED ON THE MDL PRINCIPLE , 1994, Comput. Intell..

[32]  A. Meek,et al.  A prospective study of cellulitis in broiler chickens in southern Ontario. , 1996, Avian diseases.

[33]  D Meemken,et al.  [Proposals for designing the food chain information for the implementation of the risk-oriented ante- and post-mortem meat inspection]. , 2007, DTW. Deutsche tierarztliche Wochenschrift.