DEVELOPMENT OF TRYPANOSOMA EVANSI INFECTION DETECTION USING ELECTRONIC TONGUE AND MACHINE LEARNING COMBINATION METHOD

Abstract

Trypanosoma evansi is a parasite that can affect animal development and cause economic losses. Diagnosis of infection T. evansi generally based on anamnesis, clinical examination, and conventional methods, but these methods have low sensitivity so it is necessary to develop methods that are faster, more sensitive, and can be applied in the field. The research aims to develop infection detection T. evansi by using a combination of electronic tongue biosensors and machine learning. E-tongue and machine learning test results in blood samples resulted in a total variance of PCA of 75.8% and LDA of 97.96%, while in serum samples PCA was 40.52% and 97.46%. Accuracy of total LDA performance for blood samples is 100% and serum is 100%. The combination of e-tongue and machine learning can detect blood and serum samples from infected and uninfected mice T. evansi based on PCA and LDA data pattern plots.

 Keywords: Trypanosoma evansi; Electronic Tongue, Machine Learning, PCA, LDA, Biosensor