The Spaniards are developing a trap that classifies mosquitoes by sex and gender

The expert team in entomology and arboviruses from the Animal Health Research Center (CReSA) of the Institute for Food Research and Technology (IRTA) and the company Irideon SL have developed the first optical sensor coupled with a trap that classifies mosquitoes captured automatically and reliably by gender and sex. Laboratory test results indicate that the sensor is able to classify specimens according to the genera Aedes or Culex with a reliability of 94.2%. The study, which is part of the European VECTRACK project, also shows that the sensor differentiates males from females of the genus Aedes with a reliability of 99.4% and distinguishes the sex of the Culex with complete reliability.

The research team worked with these genera because they include two of the species of most concern currently in Europe in the field of public health and surveillance of mosquito-borne viruses. These are, on the one hand, the tiger mosquito (Aedes albopictus), an urban mosquito and potential transmitter of viruses such as dengue fever, Zika, chikungunya fever, and on the other hand, the common mosquito (Culex pipiens) , which lives in urban, rural and humid areas, and can transmit diseases such as west nile virus. “We are primarily interested in sensors that identify females because they are the ones that can bite people and transmit viruses,” he says. Sandra TalaveraIRTA-CReSA researcher and head of the VECTRACK project.

For two years, within the IRTA-CReSA laboratory, they have recorded the flight of more than 4,300 specimens of tiger mosquito and mosquito common bred in the same facilities to train the optical sensor and learn to detect the frequency with which mosquitoes beat their wings – what would be understood as the buzz – measured in Hertz. In the case of mosquitoes, the flight frequency is between 300 and 900 Hz. like the size. , age and behavior during mating, as well as other environmental factors such as temperature,” he explains. Maria Elisabeth Gonzalez, predoctoral researcher at IRTA-CReSA and first author of the study. Until now, commercially available optical sensors only distinguished mosquitoes from other species and counted them, without differentiating their species, sex or other characteristics of the mosquitoes.


When a mosquito flies near the inlet funnel of the trap, it may be sucked in by the fan inside. At this moment, the sensor detects the insect thanks to an optical panel which emits light and another which receives it. When the mosquito crosses the detection area, it casts a shadow on the optical receiver. Thus, when the insect flaps its wings, the light falling on the receiver changes and causes changes in the amplitude of the light waveform recorded by the sensor. “Thanks to the Python programming language, the sensor manages to translate these optical signals into acoustic signals,” he points out. João Incarnationdirector of Irideon SL

Recordings of mosquito flights last an average of 30 milliseconds and can be downloaded from the sensor as audio files for playback and viewing. This information was combined with machine learning techniques and artificial intelligence algorithms developed to train the sensor.


Taking into account the fact that every year more than 700,000 people die on the planet due to mosquito-borne diseases, it is essential to quickly identify their species when there are cases of people affected by a disease which is transmitted through mosquito bites. This work is carried out by entomology professionals from the observation of the morphology of insects. “It is very laborious and urgent work, especially in emergencies where the weather marks the possible spread of a mosquito-borne virus,” Talavera points out.

Thanks to artificial intelligence, traps should be able to identify mosquitoes in real time and send results remotely immediately to the competent authorities to help them make decisions and thus speed up the process of surveillance and control of virus-carrying mosquitoes. One of the advantages explained by the experts is that staff resources can be prioritized and it will not be necessary to travel to the place where the entomological inspection is to be carried out if the species of interest is not detected. For example, during entomological inspections where there is a suspicion of dengue fever, if the trap indicates that there is no tiger mosquito, it will not be necessary for anyone to go there expressly to take the sample . The remote connection will also be useful for analyzing the trends and risk of transmission of mosquito-borne diseases at a general level.

In recent years, many studies have been initiated to develop technologies based on identifying mosquito wing beat even with body shape, but achieving a very reliable sensor is a challenge. “At the moment our study It showed that the optical sensor is reliable under controlled laboratory conditions, but we will soon have the results of tests carried out in the field, where environmental conditions vary and can have an influence,” Talavera concludes.

Leave a Comment