The Daily Prompt
Read in your language:

← Front page · Archive · science

SCIENCE · Phys.org · 2026-11-16 · editor 10/10 · 2 min read fact-checked

AI Tool Counts Mosquito Eggs in Smartphone Photos for Enhanced Surveillance

#AI #Mosquito Surveillance #Public Health #FIU

A new AI-powered tool has been developed to help mosquito control teams more rapidly identify and count mosquito eggs from smartphone images, according to a research team led by FIU's Andre Costa-da-Silva. The "AeEggCounter" tool is designed to assist in mosquito surveillance efforts by automating a traditionally labor-intensive and error-prone process.

The surveillance process typically starts with ovitraps, which are small, water-filled containers attracting female Aedes mosquitoes to lay eggs. These eggs are collected on brown paper strips and traditionally examined under a microscope by specialists. Costa-da-Silva, an assistant professor of biology and researcher in the Biomolecular Sciences Institute, stated the idea was to develop an app based on a machine learning model. He noted it is "the first mosquito egg-counting tool capable of analyzing real-world field samples using images captured with different smartphone models."

The technology behind AeEggCounter is YOLO, or "You Only Look Once," an artificial intelligence algorithm that can swiftly identify and locate multiple objects within an image. Researchers trained the tool by manually identifying and labeling 4,398 individual eggs across 168 smartphone images. This extensive training allows the AI to recognize individual mosquito eggs in both laboratory and field samples, even when surrounded by tiny pieces of debris.

Initial testing demonstrated the model's effectiveness, correctly identifying 88.3% of the eggs it flagged and detecting 91.7% of the eggs actually present in the images. When tested on 76 previously unseen field samples, the model's performance decreased, but its overall egg counts showed a strong correlation with those made by trained human specialists, confirming its potential for mosquito surveillance. Costa-da-Silva emphasized that the tool "doesn't replace the human," stating that human confirmation remains necessary, especially in sensitive areas.

The technology is publicly available through ACTT-Study, a free smartphone app compatible with both Android and iOS devices. Matthew DeGennaro, director of the Biomolecular Sciences Institute and a co-investigator on the study, is already utilizing the app to enhance his field studies in Little Havana and East Hialeah. The study's findings were published in the Journal of the American Mosquito Control Association.

The tool is also being integrated into educational settings. Melissa Gonzalez, a second-year biology student, recounted using ovicups at home in Miami and then counting eggs both manually and with the app in class. Costa-da-Silva sees the technology's broader potential in training the next generation of scientists in the application of AI-powered tools. The study brought together researchers from FIU, the University of South Carolina, Boston University, Baylor College of Medicine, and Texas Children's Hospital.

What to watch: The adoption and further development of AI tools in public health surveillance.

Editor's note: Excellent, detailed summary that covers the methodology, the AI technology used, the institutional collaboration, and the practical application of the tool.

Source

AI-generated and fact-checked against the original report; claims the gate cannot verify are held back.