Abstract
Accurate identification of sandfly species is critical for controlling and preventing the spread of visceral leishmaniasis, a major public health concern in Latin America. Morphological similarities between female Lutzomyia cruzi and Lutzomyia longipalpis present a significant challenge for traditional identification methods, highlighting the need for innovative alternative approaches. This study evaluates the potential of Fourier transform infrared (FTIR) spectroscopy associated with principal component analysis (PCA) and machine learning (ML) algorithms for species discrimination. Using vibrational bands predominantly assigned to lipid and carbohydrate molecules, the method achieved over 95% classification accuracy with the Linear support vector machine. Our results demonstrate that the 2970-2800 cm(-1) (C-H stretching) and 1154-1109 cm(-1) (C-O and C═C stretching) spectral ranges are particularly informative for distinguishing the species. The approach offers a rapid, cost-effective, and nondestructive solution for entomological classification, significantly enhancing vector surveillance capabilities. The integration of FTIR and machine learning (ML) techniques represents a transformative tool for entomological and epidemiological studies, providing valuable support for disease control strategies.