GLOBE Projects

GLOBE Side Navigation

Predicting Volume of Mosquito Borne Diseases Using A.M.E.A. Sensor/Radio Network

Student(s):Miguel Jose Bueno, James Ervin, Anushka Jain, Owen Luo, and Aarnav Tendulkar
Grade Level:Secondary School (grades 9-12, ages 14-18)
GLOBE Educator(s):Cassie Soeffing
Contributors:Dr. Rusty Low, IGES, scientist Peder Nelson, OSU, sme Dr. Erika Podest, NASA JPL, scientist Andrew Clark, IGES, EO Researcher and Data Analyst
Report Type(s):International Virtual Science Symposium Report, Mission Mosquito Report
Protocols:Earth As a System, Mosquitoes
Presentation Video: View Video
Presentation Poster: View Document
Language(s):English
Date Submitted:01/24/2023
The A.M.E.A. (AI-Powered Mosquito Environmental Analyzer) Sensor/Radio Network is an effective tool towards predicting volumes of mosquito borne diseases. In this study, data based on light, temperature, and precipitation is fed into a ML model that then processes said data, producing Mosquito Borne Disease, (MBD), predictions for that specific area and climate. Will the data received from the A.M.E.A. Network be sufficient enough in creating accurate predictions in a given location? Using a pair of computers and a transmitter (Raspberry Pi Speaker/Kenwood TS-440) and receiver (Computer microphone/Kenwood R600) unit, each connected to their respective radios, we can demonstrate the ability to transmit data using a novel technique with Slow Scan Television (SSTV) images that can be fed into an ML Model that processes and decodes the data and images to create a prediction on the volume of MBDs in a given area. These MBD predictions may then be compared to GLOBE data in order to assess if our MBD predictions are accurate based on a given location and climate. Our decision forest model had an accuracy of 64.6%, and the neural network Google Teachable Machine image-recognition ML model is able to predict the color of an SSTV image with 100% certainty. Since our AMEA sensor/radio network proved to be extremely accurate, in the future, scientists and public health professionals may better prepare for outbreaks in advance, limiting the amount of cases and casualties from mosquito-borne diseases in communities. Keywords: citizen science; machine learning model; mosquito prediction; sensor/radio network; engineering Research



Comments

This is very impressive. With such brilliant initiative, communities in the tropics with very high MBDs such as malaria can be tracked and identified easily and faster, this will in turn encourage prompt, accurate and specific interventions.

Very thoughtful and exciting. It  could bring about predictive surveillance and monitoring of MBDs in our community or local environment.