Using transfer learning and dimensionality reduction techniques to improve generalisability of machine-learning predictions of mosquito ages from mid-infrared spectra

Aug 5, 2026

Upload by Jiven

Old mosquito age strongly influences malaria transmission risk, making accurate mosquito age prediction crucial for evaluating mosquito-targeted interventions. Traditional age-grading methods are laborious and costly, while mid-infrared spectroscopy can capture age-related patterns in mosquito cuticles. This study tests whether dimensionality reduction and transfer learning applied to MIRS data improve cross-population generalisability for classifying mosquito age.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Abstract

Background

Methods

Results

Conclusion

Keywords

Mwanga etal. BMC Bioinformatics (2023) 24:11 BMC Bioinformatics [https://doi.org/10.1186/s12859-022-05128-5](https://doi.org/10.1186/s12859-022-05128-5) RESEARCH Open Access Using transfer learning and dimensionality reduction techniques to improve generalisability of machine-learning predictions of mosquito ages from mid-infrared spectra Emmanuel P. Mwanga 1,2*, Doreen J. Siria 1, Joshua Mitton2,3, Issa H. Mshani 1,2, Mario González‑Jiménez4, Prashanth Selvaraj5, Klaas Wynne4, Francesco Baldini2, Fredros O. Okumu 1,2,6 and Simon A. Babayan2 *Correspondence: [emwanga@ihi.or.tz](emwanga@ihi.or.tz) 1 Environmental Health and Ecological Sciences Department, Ifakara Health Institute, Morogoro, Tanzania 2 School of Biodiversity, One Health, and Veterinary Medicine, University of Glasgow, Glasgow G12 8QQ, UK 3 School of Computing Science, University of Glasgow, Glasgow G12 8QQ, UK 4 School of Chemistry, University of Glasgow, Glasgow G12 8QQ, UK 5 Institute for Disease Modelling, Bellevue, WA 98005, USA 6 School of Public Health, University of Witwatersrand, Johannesburg, South Africa Abstract Background: Old mosquitoes are more likely to transmit malaria than young ones. Therefore, accurate prediction of mosquito population age can drastically improve the evaluation of mosquito‑targeted interventions. However, standard methods forage‑grading mosquitoes are laborious and costly. We have shown that Mid‑infrared spectroscopy (MIRS) can be used to detect age‑specific patterns in mosquito cuticlesand thus can be used to train age‑grading machine learning models. However, these models tend to transfer poorly across populations. Here, we investigate whether applying dimensionality reduction and transfer learning to MIRS data can improve the transferability of MIRS‑based predictions for mosquito ages. Methods: We reared adults of the malaria vector Anopheles arabiensis in two insectar‑ ies. The heads and thoraces of female mosquitoes were scanned using an attenuated total reflection‑Fourier transform infrared spectrometer, which were grouped into two different age classes. The dimensionality of the spectra data was reduced using unsu‑ pervised principal component analysis or t‑distributed stochastic neighbour embed‑ ding, and then used to train deep learning and standard machine learning classifiers. Transfer learning was also evaluated to improve transferability of the models when predicting mosquito age classes from new populations. Results: Model accuracies for predicting the age of mosquitoes from the same popu‑ lation as the training samples reached 99% for deep learning and 92% for standard machine learning. However, these models did not generalise to a different popula‑ tion, achieving only 46% and 48% accuracy for deep learning and standard machine learning, respectively. Dimensionality reduction did not improve model generalizability but reduced computational time. Transfer learning by updating pre‑trained models with 2% of mosquitoes from the alternate population improved performance to ~ 98% accuracy for predicting mosquito age classes in the alternative population. Conclusion: Combining dimensionality reduction and transfer learning can reduce computational costs and improve the transferability of both deep learning and stand‑ ard machine learning models for predicting the age of mosquitoes. Future studies © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the mate‑ rial. If material is not included in the articl

FAQs

Why is predicting mosquito age important for malaria control?

Older mosquitoes have a higher likelihood of transmitting malaria, so accurate age prediction improves evaluation of mosquito-targeted interventions.

How are mosquito age classes predicted in this study?

Heads and thoraces are scanned with an attenuated total reflection Fourier transform infrared spectrometer, spectra are reduced using PCA or t-SNE, and classifiers are trained using deep learning or standard machine learning.

What approach improved transferability to a new mosquito population?

Updating pre-trained models with about 2% of mosquitoes from the alternate population increased performance to around 98% accuracy in the alternative population.

Aug 5, 2026

Upload by Jiven