Can AI identify mosquito species? VectorCAM, a pocket-sized device, uses machine learning to differentiate species with 95% accuracy, enhancing malaria surveillance efforts

Transcript

Not all mosquitoes are created equal. Of the more than three thousand species, only a limited number of the Anopheles genus can transmit malaria. Even within that subset, subtle physiological differences affect how malaria spreads. Some mosquitoes prefer to bite indoors, while others outdoors. Some need large bodies of water to breed, while others only need a small puddle. Distinguishing these species is critical for effective malaria control—whether using bed nets, indoor spraying, or outdoor larval management. But identifying them by eye takes expert, entomological knowledge. Could AI help? The VectorCAM team at Johns Hopkins is working on just that. Their pocket-sized device uses a small light and magnifying lens, allowing a phone camera to capture close-up images of mosquitoes placed on slides. With up to 95% accuracy, it can identify mosquito species based on morphology in seconds. The hope is that VectorCAM will help health teams better understand mosquito populations, paving the way for more targeted and relevant malaria control efforts.

Source

Towards transforming malaria vector surveillance using VectorBrain: a novel convolutional neural network for mosquito species, sex, and abdomen status identifications (Scientific Reports)

About The Podcast

The Johns Hopkins Malaria Minute podcast is produced by the Johns Hopkins Malaria Research Institute to highlight impactful malaria research and to share it with the global community.