Skip to main content
ARS Home » Plains Area » Manhattan, Kansas » Center for Grain and Animal Health Research » Stored Product Insect and Engineering Research » Research » Publications at this Location » Publication #378120

Research Project: Impacting Quality through Preservation, Enhancement, and Measurement of Grain and Plant Traits

Location: Stored Product Insect and Engineering Research

Title: An autoencoder and artificial neural network-based method to estimate parity status of wild mosquitoes from near-infrared spectra

Author
item MILALI, MASABHO - Marquette University
item KIWARE, SAMSON - Marquette University
item GOVELLA, NICODEM - Ifakara Health Institute
item OKUMU, FREDROS - Ifakara Health Institute
item BANSAL, NAVEEN - Marquette University
item BOZDAG, SERDAR - Marquette University
item CHARLWOOD, JACQUES - Liverpool School Of Tropical Medicine
item MAIA, MARTA - Kenya Medical Research Institute
item OGOMA, SHEILA - Clinton Health Access Initiative
item Dowell, Floyd
item CORLISS, GEORGE - Marquette University
item SIKULU-LORD, MAGGY - University Of Queensland
item POVINELLI, RICHARD - Marquette University

Submitted to: PLoS ONE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/27/2020
Publication Date: 6/18/2020
Citation: Milali, M.P., Kiware, S.S., Govella, N.J., Okumu, F., Bansal, N., Bozdag, S., Charlwood, J.D., Maia, M.F., Ogoma, S.B., Dowell, F.E., Corliss, G.F., Sikulu-Lord, M.T., Povinelli, R.J. 2020. An autoencoder and artificial neural network-based method to estimate parity status of wild mosquitoes from near-infrared spectra. PLoS One. 15(6):e0234557. https://doi.org/10.1371/journal.pone.0234557.
DOI: https://doi.org/10.1371/journal.pone.0234557

Interpretive Summary: Female mosquitoes need animal blood to develop their eggs. In the process of acquiring blood, they may acquire pathogens which may cause different diseases in humans such as malaria, zika, dengue, and chikungunya. Therefore, it is important to know if a mosquito has fed on blood so that they can be controlled as needed to minimize disease transmission. The typical method to determine if a mosquito has fed on blood is to dissect mosquito ovaries, which is very slow and tedious. An alternative method is to use a non-destructive and rapid method such as near-infrared spectroscopy (NIRS). In this study, we develop mathematical models to determine if NIRS can be used to determine if a mosquito has had a blood meal. Our results show that a combination of an autoencoder and artificial neural networks trained on NIR spectra could be used to determine if a mosquito had fed on blood with accuracies of about 90%. This method is a high-throughput, reagent-free, and simple-to-use technique compared to ovary dissections.

Technical Abstract: After mating, female mosquitoes need animal blood to develop their eggs. In the process of acquiring blood, they may acquire pathogens, which may cause different diseases in humans such as malaria, zika, dengue, and chikungunya. Therefore, knowing the parity status of mosquitoes is useful in control and evaluation of infectious diseases transmitted by mosquitoes, where parous mosquitoes are assumed to be potentially infectious. Ovary dissections, which are currently used to determine the parity status of mosquitoes, are very tedious and limited to few experts. An alternative to ovary dissections is near-infrared spectroscopy (NIRS), which can estimate the age in days and the infectious state of laboratory and semi-field reared mosquitoes with accuracies between 80 and 99%. No study has tested the accuracy of NIRS for estimating the parity status of wild mosquitoes. In this study, we train an artificial neural network (ANN) models on NIR spectra to estimate the parity status of wild mosquitoes. We use four different datasets: An. arabiensis collected from Minepa, Tanzania (Minepa-ARA); An. gambiae s.s collected from Muleba, Tanzania (Muleba-GA); An. gambiae s.s collected from Burkina Faso (Burkina-GA); and An.gambiae s.s from Muleba and Burkina Faso combined (Muleba-Burkina-GA). We train ANN models on datasets with spectra preprocessed according to previous protocols. We then use autoencoders to reduce the spectra feature dimensions from 1851 to 10 and re-train the ANN models. Before the autoencoder was applied, ANN models estimated parity status of mosquitoes in Minepa-ARA, Muleba-GA, Burkina-GA and Muleba-Burkina-GA with out-of-sample accuracies of 81.9±2.8 (N = 274), 68.7±4.8 (N = 43), 80.3±2.0 (N = 48), and 75.7±2.5 (N = 91), respectively. With the autoencoder, ANN models tested on out-of-sample data achieved 97.1±2.2% (N = 274), 89.8±1.7% (N = 43), 93.3±1.2% (N = 48), and 92.7±1.8% (N = 91) accuracies for Minepa-ARA, Muleba-GA, Burkina-GA, and Muleba-Burkina-GA, respectively. These results show that a combination of an autoencoder and an ANN trained on NIR spectra to estimate the parity status of wild mosquitoes yields models that can be used as an alternative tool to estimate parity status of wild mosquitoes, especially since NIRS is a high-throughput, reagent-free, and simple-to-use technique compared to ovary dissections.