Our Innovation Director, Leslie, recently was interviewed for an article that appeared in the NVIDIA blog, that discusses Wovenware’s work developing an AI solution to help with the identification and classification of mosquitoes carrying deadly diseases. The deep learning solution is being developed for the Puerto Rico Vector Control Unit (PRVCU), to aid in its project to control the spread of diseases across the island.
As the article states, “Ask folks which animal kills the most people each year and they’ll probably say crocodiles, sharks or maybe lions. But the correct answer is a lot less obvious — the mosquito.” Thanks to the PRVCU’s hard work, maybe one day mosquitoes will lose this title and go back to being simply annoying.
We’re very proud of the fact that our deep learning solution is automating the identification and classification of Aedes Aegyptis, which is infecting people with diseases such as Zika, Dengue and Chikungunya across Puerto Rico and nationwide. The other purpose of the project is to develop safe and more effective insecticides.
Our solution is built upon convolutional neural networks (CNNs), which enable it to perform gender and species classification. The two classifier CNNs can then be consolidated into one for training. Leslie said the team is also experimenting with TensorFlow to further speed the inference process.
The article explains that while there are other initiatives underway to combat mosquito-borne diseases, Wovenware is more concerned with technology solutions that would provide the least disruption to the food chain. As Leslie says, “Every species has a reason to live,” and I don’t see why we should need to extinguish a species.”
We’re very grateful to NVIDIA for shedding light on how deep learning AI can help address some of the most pressing environmental and health-related concerns. Stay tuned for more updates as our AI solution continues to do its part in helping to eradicate mosquito-borne diseases.