Last Updated: 06/03/2024
Advancing infrared and AI-based techniques for real time mosquito age-grading and evaluation of malaria vector control interventions in Africa
Objectives
This project will consolidate lessons from previous efforts to optimize the performance of infrared (IR) spectroscopy techniques and artificial intelligence (AI) for the age-grading of wild-collected mosquitoes. And it will demonstrate scalable field applications and deliver an operationally relevant integrated IR-AI system of hardware and software, thereby enabling age-grading to become a simple routine activity for researchers and National Malaria Control Programmes (NMCP).
Another core component of this project will be to train an initial cohort of technicians in Africa to use this new method.
Mercy Opiyo
Fredros Oketch Okumu
Maggy Sikulu
Abdoulaye Diabate
Francesco Baldini
The ability of Anopheles mosquitoes to transmit human malaria is age dependent, and females must be at least 11 (±1) days old to be infective. The age structure of mosquito populations, and in particular the proportion of females old enough to transmit diseases, can therefore be relied upon to estimate transmission risk or infer the performance of key interventions. Recent studies have demonstrated the principle of using reflectance of infrared wavelengths from mosquitoes to classify mosquitoes into age categories and to detect pathogens (including Plasmodium, Zika virus) and endosymbionts (e.g., Wolbachia) in the mosquitoes. This project will consolidate lessons from these previous efforts to optimize the performance of infrared (IR) spectroscopy techniques and artificial intelligence (AI) for age-grading of wild-collected mosquitoes. We will demonstrate scalable field applications and deliver an operationally relevant integrated IR-AI system of hardware and software, thereby enabling age-grading to become a simple routine activity for researchers and National Malaria Control Programmes (NMCP). By relying on off-the-shelf hardware, open-source software and user-friendly graphics, our system will enable fine-scale changes in Anopheles age structure to be detected and quantified, thus providing a better proxy for malaria transmission than mosquito density alone. It will not require expensive reagents or advanced mathematical skills. Our refined IR-AI systems will include handheld and smart traps to allow real time and remote surveillance. Another core component of this project will be to train an initial cohort of technicians in Africa on use of this new method. We anticipate that the resulting integrated system will complement, and in some cases lower the costs of current efforts for measuring transmission risk and evaluating vector control tools.
Oct 2021 — Sep 2024
$3.78M
