Comparative Evaluation of Benchtop and Portable Near-Infrared Spectrometers for Predicting the Age and Blood Feeding History of Aedes aegypti

This study is a comparative assessment of a more affordable handheld spectrometer (NIRvascan) with the traditional Labspec 4i spectrometer for predicting the chronological age and blood feeding history of female Aedes aegypti mosquitoes reared in the lab. Three separate cohorts of laboratory-reared Ae. aegypti mosquitoes were reared and collected at three age groups (1-, 10- and 17-days old). A model developed using Artificial Neural Networks (ANN) with spectra collected by the Labspec 4i NIR spectrometer predicted the age of Ae. aegypti, classifying them into two groups (< or ≥ 10 days) with a predictive accuracy of 94% (N = 366) whereas an ANN model developed using spectra collected by the NIRvascan spectrometer predicted the age of Ae. aegypti mosquitoes, classifying them into the same age group with a predictive accuracy of 90% (N = 290). ANN models developed for predicting the blood feeding history of mosquitoes were 82.8% (N = 308) and 71.4% accurate (N = 300) when Labspec 4i and NIRvascan were used, respectively. This is the first study to demonstrate that a handheld NIR instrument operated by a smart phone could potentially be used for predicting entomological parameters of mosquitoes.