Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/51177
標題: 熟雞肉系統中李斯特菌及耶爾辛尼腸炎桿菌預測生長模式之建立及其應用
Development and application of predictive model for Listeria monocytogenes and Yersinia enterocolitica in cooked chicken meat
作者: 危貴金
King, Wei Que
關鍵字: Predictive model;預測模式;Listeria monocytogene;Yersinia enterocolitica;microbiology risk assessment;李斯特菌;耶爾辛尼腸炎桿菌;微生物風險評估
出版社: 食品科學系
摘要: 
本研究主要建立李斯特菌及耶辛尼氏腸炎桿菌於熟雞肉之預測生長模式。首先利用修正Gompertz 模式建立生長曲線的一級模式,由模式中估計其生長參數值,包括最大比菌數量、最大比生長速率及遲滯期(A、μm及λ)。二級模式分別將李斯特菌及耶辛尼氏腸炎桿菌由模式所估計之生長參數建立與溫度之適當模式描述之,進一步將所得結果與PMP的資料庫加以比較,模式的適用性之,除以觀測值及預測值所得散佈圖及殘差分析圖外,並用數學/統計指標如MSE (mean square errors)、判定係數(R2)及最近發展的偏估因子 (bias factor)及準確因子(accuracy factor)加以驗證,再將李斯特菌預測生長模式進一步應用於微生物風險評估。實驗結果顯示熟雞肉中李斯特菌及耶辛尼氏腸炎桿菌之生長可利用Gompertz 模式加以配適。將所得生長參數如比生長速率(m)及遲滯期(l)分別建立與溫度的關係,遲滯期以Adair及Smith模式描述;比生長速率則以Ratkowsky及Zwietering模式描述,所選用模式之適用性則利用作圖法、MSE、R2、bias factor及 accuracy factor等方驗證。以Adair及Smith模式描述耶辛尼氏腸炎桿菌於雞肉中其遲滯期與溫度關係,所得預測值之變異較大。Ratkowsky及Zwietering兩種模式其預測所得的數值較準確。主要利用微生物預測模式建立一由於消費含有污染李斯特菌的熟雞肉系統之定量風險評估模式。本模式主要應用所得預測熟雞肉系統中李斯特菌生長的模式,並使用German model及 Weibull-Gamma model為劑量反應模式,經蒙地卡羅模擬建立每餐攝取污染李斯特菌致病機率分佈。由模擬結果可知,假設起始污染菌量為1CFU/g,於4℃下貯藏24 hrs後,由表中顯示對German model 每年最大可能發生案例約1.43件,而以Weibull-Gamma model的高危險群平均每年可能發生的案例有342件,而以Weibull-Gamma 的低危險群平均可能發生最大案例有0.04件。

Modified Gompertz function was used to predict the growth of Listeria monocytogenes Scott A and Yersinia enterocolitica in chicken meats. After the growth curves of pathogens were obtained, the modified Gompertz function was then fit to the experimental curves, the variations of the parameters A (the asymptotic value of the stationary growth phase), mm (the maximum specific growth rate), and l (lag phase). The variations of the parameters lag time (l) and specific growth rate (m) with temperature were then modeled. The variation of lag time can be described by Adair model and Smith model;on the other hand, Ratkowsky model and Zewietering model were used to describe the variation of specific growth rate. The various models were compared by graphical and mathematical analysis such as MSE, R2, bias factor and accuracy factor. Lag time modeled with Adair and Smith function had larger variance, indicating larger errors. The accuracy factor of Zwietering model was lower than Ratkowsky model, indicating the Zwietering model was more accurate than Ratowsky model in describing the parameters used in this study. The Zwietering model was found to be suitable for predicting the specific growth rate of L. monocytogenes and Y. enterocolitica in the chicken meat and parameters generated by PMP. In addition, the predicted value for the pathogens in the PMP was higher than that in the chicken meat. The study was to development risk assessment, which applied the predictive microbiology methodology, for Listeria monocytogenes in cooked chicken meats. The probability of illness per serving was calculated by using German model and Weibull-Gamma model. Monte Carlo Simulation was used to establish the distribution probability of illness. Assuming cooked chicken was contaminated by Listeria monocytogenes with the initial number content of 1 CFU/g and storaged at 4℃ for 24 hrs. The estimated maximum number of listeriosis / year predicted by German model was 1.43. On the other hand, mean number of listeriosis/year predicted by weibull-Gamma model for high risk population or low risk population were 342 and 0.04 , respectively.
URI: http://hdl.handle.net/11455/51177
Appears in Collections:食品暨應用生物科技學系

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.