Coming into the NASA SEES Internship, I had only ever experienced science through a one-dimensional lens. I always knew the results, experimenting just to prove existing knowledge. This internship taught me to look outside the box and challenge my preconceived notions to discover new relationships that exist on our earth.
Through our weekly meetings, I got to interact with guest speakers and my mentors who were taking on various projects in the field of epidemiology. Inspired by one such meeting where Dr. Chellappan used machine learning to predict the mosquito species in an image, I wanted to explore the power of this tool. I started asking questions, intrigued by how the model used mathematical values to make sense of the pixels in the image.
My swarm of questions did not go unnoticed - fellow researcher Sriram Elango picked up on my interest in AI. He messaged me in the zoom chat and expressed his interest in doing a project with machine learning as well! We decided to research together, brainstorming ideas for our project.
We realized to prevent the spread of mosquito-borne diseases, we had to focus on the root of the problem: mosquito breeding grounds. We researched existing habitat detection technology, noticing the expensive and time-consuming approach to map areas using Drones and multispectral imaging like Lidar. We knew this process had to be streamlined and applied to a macro-scale, so we built an autonomous mosquito habitat detector. We annotated hundreds of satellite images and inputted them into a convolutional neural network, classifying the habitats as rivers, ponds, and inlets. Given any satellite image, the model is able to clearly identify mosquito habitats with astounding accuracy, potentially allowing for a cheap and accessible solution to mosquito habitat detection that can be integrated into public health policies.
We have finished our research paper outlining our process of data collection and presenting our model as a proof of concept. We are currently working on publishing our paper with the help of our amazing mentors and hope to share our findings with the scientific community. The NASA SEES Internship has taught me to explore and experiment without constraint. I have come to appreciate the game-changing work that NASA has done and hope to become a part of that change as I grow into a scientist in the future.
Nandini is a high school student at Plano West Senior High School in Texas. Her virtual internship is part of a collaboration between the Institute for Global Environmental Strategies (IGES) and the NASA Texas Space Grant Consortium (TSGC) to extend the TSGC Summer Enhancement in Earth Science (SEES) internship for US high school (http://www.tsgc.utexas.edu/sees-internship/). She shares her experience this summer in this guest blog post.
