Home

Welcome to the vector-born-diseases wiki!

My project aims to identify vector carriers (i.e. insects, arachnids, etc. that carry disease) through deep learning, or artificial intelligence. This page contains my research, background information and possible applications.

Vector-borne diseases cause more than 700,000 deaths annually, and account for more than 17% of all infectious diseases.

My project aims to use image recognition through Neural Networks to correctly identify harmful vectors (i.e. insects/arachnids) that may carry disease by identifying key features associated with different genera of vector carries and comparing/contrasting against animals that do not carry disease. For example, an anopheles mosquito has its characteristic striped body compared to a common mosquito, which has a stereotypical plain brown body. It also examines positioning of the body and anatomy that differ from other types of animals.

Animals trained on include:

• Anopheles mosquito
• Aedes mosquito
• Culex mosquito
• Various genera of ticks (including lone star tick and blacklegged tick)
• Blackfly
• Triatoma (a.k.a. kissing bug)

Code is written in python using tensorflow, keras, numpy, skimage, and PIL libraries.