In a recent study conducted in Brazil by researchers from the Instituto Oswaldo Cruz, the E4Warning project successfully deployed smart traps with the VECTRACK sensor developed by IRIDEON, to improve mosquito surveillance in urban settings. These smart traps, equipped with AI and IoT technologies, have been tested in two Brazilian cities—Rio de Janeiro and Brasilia—where they demonstrated high accuracy in identifying mosquito species such as Aedes aegypti and Culex quinquefasciatus, both key vectors for diseases like dengue, Zika, and chikungunya.

Over a year of field testing, the traps showed a remarkable 99.8% accuracy in distinguishing target mosquito species from other insects and a 93.7% accuracy in differentiating between the two genera of mosquitoes. The smart traps collected data on 9,151 insect flights, of which 1,383 were mosquito vectors.

These findings indicate the potential of the E4Warning tool to provide real-time monitoring and classification of mosquitoes in areas with high vector-borne disease risks: by the end of September 2024, over 9.5 million dengue cases had been reported in Brazil. In this context, the data gathered from the smart traps could be instrumental in aiding public health authorities to make timely interventions, such as targeted vector control measures, enhancing the preparedness and response to mosquito-borne disease outbreaks.  The successful implementation of the IRIDEON smart traps in Brazil showcases how cutting-edge technologies can revolutionize mosquito surveillance, making it faster, more accurate, and less reliant on manual labor.

The paper has been published in the Parasite & Vector journal: Njaime, F.C.B.F.P., Máspero, R.C., Leandro, A.d., Maciel-de-Freitas, R. Automated classification of mixed populations of Aedes aegypti and Culex quinquefasciatus mosquitoes under field conditions. Parasites Vectors 17, 399 (2024).

It can be consulted at the following link: https://parasitesandvectors.biomedcentral.com/articles/10.1186/s13071-024-06417-z

It can also be founded in the E4Warning Zenodo community: https://zenodo.org/records/13868673

In our ongoing efforts to enhance mosquito surveillance and early warning systems, the E4Warning consortium is leveraging innovative technologies, including the improvement of the IRIDEON SL smart-traps. These smart-traps are designed to automate and improve the efficiency of mosquito monitoring, addressing the challenges faced by traditional methods.

Overcoming Traditional Surveillance Challenges

Traditional mosquito surveillance methods are costly and time-consuming, requiring significant professional resources for trap placement, sample collection, and laboratory analysis. Additionally, there is an inevitable time lag between placing traps and collecting samples, which can lead to inaccurate and untimely monitoring of mosquito populations.

The IRIDEON smart traps are off-the-shelf commercial suction traps equipped with the VECTRACK optoelectronic sensor prototype developed by IRIDEON SL in Barcelona. These sensors are capable of distinguishing between mosquito species, sex, and age under both laboratory and field conditions. The VECTRACK sensor includes an emitter, an array of LEDs, and photo-transistors acting as photoreceptors, all working together to create a light field. When a mosquito is drawn into the trap, its wing flapping modulates the light field, generating a species-specific signature. This signature is processed using AI methods, including rule-based systems, genetic algorithms, artificial neural networks, and fuzzy models, to classify the mosquito by genus, species, sex, and age.

Each VECTRACK recording includes GPS coordinates, date and time of capture, ambient temperature, and relative humidity. Every 30 minutes, the field sensor transmits data batches to the server via the mobile phone network. The server then classifies each mosquito event, providing detailed and accurate data on mosquito populations in real-time. This automation reduces the need for manual intervention and enables continuous monitoring.

Field Performance and Global Testing

The VECTRACK sensor has demonstrated its effectiveness in real-world scenarios, achieving high accuracy in field trials conducted in Spain. In El Prat de Llobregat (2021) and Rubí (2022), the sensor achieved 95.5% accuracy in species detection and 88.8% accuracy in classifying the genus and sex of mosquitoes. Check the work performed by Gozález-Pérez et al.: Field evaluation of an automated mosquito surveillance system which classifies Aedes and Culex mosquitoes by genus and sex. Parasites Vectors 17, 97 (2024). https://doi.org/10.1186/s13071-024-06177-w

Ongoing Field Tests Under E4Warning

The E4Warning project is testing the efficacy of the IRIDEON smart-traps under various conditions:

• Spain (Barcelona): VECTRACK sensors are deployed at five strategic monitoring sites across the city, focusing on Culex pipiens and Aedes albopictus. Agència de Salut Pública de Barcelona – ASPB
• Brazil (Rio de Janeiro and Brasilia): Field trials with local populations of Aedes aegypti and Culex quinquefasciatus in collaboration with Instituto Oswaldo Cruz, Fiocruz. Fundação Oswaldo Cruz (Fiocruz)
• Spain (Girona): Testing in non-urban environments like the Botanical Garden of Blanes and Aiguamolls de l’Empordà Natural Park, where diverse mosquito species are present. CSIC
• Greece (Athens): Planned deployment in the Attica region, focusing on critical points of entry for Aedes aegypti and areas of significant vector importance. Benaki Phytopathological Institute

To learn more about IRIDEON smart-traps refer to:

• González-Pérez, M.I., Faulhaber, B., Aranda, C. et al. Field evaluation of an automated mosquito surveillance system which classifies Aedes and Culex mosquitoes by genus and sex. Parasites Vectors 17, 97 (2024). https://doi.org/10.1186/s13071-024-06177-w
• González-Pérez MI, Faulhaber B, Williams M, Brosa J, Aranda C, Pujol N, et al. A novel optical sensor system for the automatic classification of mosquitoes by genus and sex with high levels of accuracy. Parasit Vectors. 2022;15:190. https://doi.org/10.1186/s13071-022-05324-5.