Accurately distinguishing mosquito species and sexes is crucial for research and public health protection. Only with precise identification of species and sex can the spread and threat potential of mosquito populations be effectively monitored and targeted control measures implemented efficiently. Mosquitofi supports researchers and health authorities by enabling automated identification of mosquito species and sexes based on smartphone photos. The app guides users through a clearly structured workflow: from creating new studies, to assigning traps and trays, all the way to uploading and analyzing mosquito images.

The results are stored in a database and can be accessed and analyzed via a REST API. Temporal dynamics play a role in tracking the development of breeding sites over time. In this way, the system supports targeted vector control measures and contributes to the prevention of dengue fever.

The underlying machine learning pipeline is based on state-of-the-art deep learning models in PyTorch, specifically trained to recognize different mosquito species and their sexes. MLflow ensures efficient and transparent documentation of development, training, and experiment tracking.