AeEggCounter: A Computer Vision Tool for High-Throughput Screening of Aedes Eggs from Ovitrap Field Samples
Article
Costa-Da-Silva, AL, Zhou, J, Frederiksen, A et al. (2026). AeEggCounter: A Computer Vision Tool for High-Throughput Screening of Aedes Eggs from Ovitrap Field Samples
. JOURNAL OF THE AMERICAN MOSQUITO CONTROL ASSOCIATION, 42(3), 189-205. 10.2987/26-7298
Costa-Da-Silva, AL, Zhou, J, Frederiksen, A et al. (2026). AeEggCounter: A Computer Vision Tool for High-Throughput Screening of Aedes Eggs from Ovitrap Field Samples
. JOURNAL OF THE AMERICAN MOSQUITO CONTROL ASSOCIATION, 42(3), 189-205. 10.2987/26-7298
Mosquito-borne diseases are resurging globally. The Americas are experiencing record numbers of dengue cases, including the USA. A 2023 US national assessment found three-fourths of local vector control programs remain insufficiently equipped, highlighting critical preparedness gaps. Integrating artificial intelligence with traditional surveillance methods offers an opportunity to address these needs. Ovitraps are widely used to monitor Aedes populations, yet egg counting is slow, labor-intensive, and error-prone, particularly in debris-rich field samples. This work developed AeEggCounter, a smartphone application powered by a YOLO algorithm and trained on panoramic images of ovitrap linings captured with commonly used smartphone cameras. Across diverse field images and 4,398 annotated individual eggs, AeEggCounter achieved 88.3% precision, 91.7% recall, and an F1 score of 89.9%. Available for Android and iOS, this free high-throughput tool has promise to enhance vector surveillance in resource-limited settings.
