× Home Projects Partners FAQs Contacts

MALARIA: Assessing machine learning methods for mosquito-targeted interventions

Jan. 23, 2023
MALARIA: Assessing machine learning methods for mosquito-targeted interventions
A snip from the BMJ journal with an inset of Ifakara Health Institute researcher Emmanuel Mwanga who contributed to the study and publication. GRAPHIC | IFAKARA/KMC.

Ifakara Health Institute scientists and colleagues from the USA and the University of Glasgow, UK have argued that accurate prediction of mosquito population age can improve the evaluation of mosquito-targeted interventions. This argument challenges the standard methods for age-grading mosquitoes, which are laborious and costly, encouraging scientists to look for better methods.

Against this backdrop, the scientists conducted a study to assess whether using transfer learning and dimensionality reduction techniques could improve the generalizability of machine-learning predictions of mosquito ages from mid-infrared spectra to which they concluded by saying: “We have shown that Mid-infrared spectroscopy (MIRS) can be used to detect age-specific patterns in mosquito cuticles and thus can be used to train age-grading machine learning models."

"Combining dimensionality reduction and transfer learning can reduce computational costs and improve the transferability of both deep learning and standard machine learning models for predicting the age of mosquitoes,” the scientists added.

Published in the BMC journal on January 9, 2023, this study adds to the growing evidence of the utility of two technologies; infrared spectroscopy and machine learning in estimating mosquito age and survival.

Ifakara scientists involved in the study are: Emmanuel Mwanga, Doreen Siria, Issa Mshani, and Fredros Okumu; from the University of Glasgow are: Joshua Mitton, Francesco Baldini, Simon Babayan, Mario González-Jiménez and Klaas Wynne; and Prashanth Selvaraj from the Institute for Disease Modelling, USA.

>> Read the full publication here: https://doi.org/10.1186/S12859-022-05128-5
>> Read other Ifakara publications here: https://ihi.or.tz/publications/journals-paper