Utilizing Serum Fluorescence Spectra and Machine Learning Algorithms for Efficient Diagnosis of Sheep Brucellosis

ABSTRACT

In sheep, the infectious disease brucellosis is caused by Brucella melitensis . Traditional serological techniques for detecting Brucella in sheep are slow and not very accurate, necessitating a faster and more precise screening method. This study seeks to assess the potential for diagnosing brucellosis seropositive sheep through the application of serum fluorescence spectroscopy in conjunction with principal component analysis‐linear discriminant analysis (PCA‐LDA), support vector machine with linear (SVM‐linear), support vector machine with radial basis function (SVM‐RBF), k‐nearest neighbors (KNN), and decision tree (DT) algorithms. The study revealed differences at 470, 515, 560, 670, and 710 nm through the analysis of serum fluorescence spectra from sheep with and without brucellosis. Among them, the diagnostic effect of the PCA‐LDA algorithm is the best (accuracy 91.0% ± 6.1%). In summary, the combination of serum fluorescence spectroscopy and the PCA‐LDA algorithm holds considerable potential for detecting brucellosis seropositive sheep.