Implementation of a deep learning model for automated classification of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse) in real time

• Main Authors: Ahmad, H. (Author), Isawasan, P. (Author), Majid, A.H.A (Author), Nair, G. (Author), Ong, S.-Q (Author)
• Format: Article
• Language: English
• Published: Nature Research 2021
• Series: Scientific Reports
• Subjects: adult; Adult; Aedes; Aedes aegypti; Aedes albopictus; anatomy and histology; animal; Animals; article; classification; comparative study; computer assisted diagnosis; controlled study; convolutional neural network; Datasets as Topic; deep learning; Deep Learning; dengue; Dengue; Entomology; feasibility study; female; Female; human; human experiment; Humans; Image Interpretation, Computer-Assisted; information processing; insecticide resistance; Insecticide Resistance; male; Male; middle aged; Middle Aged; mosquito control; Mosquito Control; mosquito vector; Mosquito Vectors; nonhuman; procedures; Video Recording; videorecording; virology; zoology
• Online Access: View Fulltext in Publisher; View in Scopus

 Classification of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse) by humans remains challenging. We proposed a highly accessible method to develop a deep learning (DL) model and implement the model for mosquito image classification by using hardware that could regulate the development process. In particular, we constructed a dataset with 4120 images of Aedes mosquitoes that were older than 12 days old and had common morphological features that disappeared, and we illustrated how to set up supervised deep convolutional neural networks (DCNNs) with hyperparameter adjustment. The model application was first conducted by deploying the model externally in real time on three different generations of mosquitoes, and the accuracy was compared with human expert performance. Our results showed that both the learning rate and epochs significantly affected the accuracy, and the best-performing hyperparameters achieved an accuracy of more than 98% at classifying mosquitoes, which showed no significant difference from human-level performance. We demonstrated the feasibility of the method to construct a model with the DCNN when deployed externally on mosquitoes in real time. © 2021, The Author(s).