AI and the World's Hard Problems · Life sciences · Nature (computational biology) · 2026-10-20
Transforming mosquito surveillance LarvaFormer as a hybrid deep learning tool for rapid and accurate larval recognition
Mosquito-borne diseases cause over 700,000 deaths annually, but traditional control methods targeting adult mosquitoes are often ineffective and environmentally harmful. LarvaFormer is a hybrid deep learning framework that combines convolutional neural networks and Vision Transformers with specialized attention mechanisms to automatically classify mosquito larvae. The model achieved state-of-the-art accuracies of 98.97% and 99.50% on the MosquitoLarvae-7400 and MLMI-2024 datasets, enabling rapid, accurate larval identification that supports targeted and environmentally sustainable mosquito control.