Spectral technique for detection of changes in eggshells caused by Mycoplasma synoviae
Zofia Maria Lorenc , Sławomir Paśko , Olimpia Kursa , Anna Pakuła , Leszek Sałbut
AbstractMycoplasma synoviae (MS) is a major pathogen in chicken and turkeys, causing subclinical infection. MS infections are highly prevalent and may potentate and be involved in sinovitis, respiratory syndromes, as well as lead to eggshell apex abnormality (EAA). A deformed, inhomogeneous eggshell is susceptible to cracks and breaks through which microbes get in and additionally entails higher water loss in the egg during the entire incubation process. Not all eggs with eggshell apex abnormality possess characteristic deformation and that is why some eggs may be incorrectly classified during a visual inspection. To minimize the above risk, the spectral VIS technique and the analysis based on the classification tree method—CTM is proposed. The method makes use of specially defined parameters extracted from the shape of transmittance spectra of eggshells. Directional coefficients of the lines adjusted to the specific ranges of the transmittance spectrum are used in the process of classifying samples as those from MS-carrying hens and from healthy hens. Three CTM-based classifiers were created for a group of white, brown, and mixed shells. After comparing, it can be concluded that the best results were obtained for the group of brown shells (accuracy 88%, specificity 88%, and false negative rate 13%). The authors present a non-invasive spectral method that utilizes eggshells, i.e., the natural waste from chicken farms. The method enables entering data into the classifiers described in the article. The process provides an opportunity to correctly assign, the examined shell to the group of shells with increased risk—with approx. 86% accuracy. This means that, if a few of such results are registered, the herd is eligible more specific studies targeting MS bacteria. Regular spectral testing can support the detection of egg lesions in MS positive flocks.
|Journal series||Journal of Poultry Science, ISSN 1346-7395, (N/A 40 pkt)|
|Publication size in sheets||0.5|
|Keywords in English||spectral technique, Mycoplasma synoviae, transmittance spectrum, classification tree method CTM, binary classification|
|Score||= 40.0, 20-10-2019, ArticleFromJournal|
|Publication indicators||= 0; : 2017 = 0.728; : 2017 = 0.86 (2) - 2017=0.803 (5)|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.