High-Level Project Summary
Since the beginning of 2020, the COVID-19 pandemic has had a great impact, causing the death of many people and affecting the lifestyle of society. In 2021, in Colombia the economical crisis and social problems have caused massive protests, generating not only the spread of the virus, but also the appearance of new variants.The current project is aimed at prototyping a Covid-19 mutation risk tool that provides an interactive map-able to predict Covid Mu, Alpha, Gamma Mutations on different geographical areas in Colombia using geo-espatial, multi-temporal and demographic data, allowing people to learn about the risk factor so that they can take action.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
Our team has developed a deep learning model that was trained with the data of the different COVID-19 Mutations present in Colombia and with satellite data of temperature, precipitation and humidity, taken from Earth Data. The tool consists of an interactive map that can predict a high risk (3 - red), medium-high (2 - red), medium (1-red) or low (0 - green) in different geographic areas of Colombia.
M-Predictor is a deep learning model that calculates the risk of a mutation, this is defined as the probability that cases increase (3), remain (2), decrease (1) or decrease a lot (1)
Additionally, you can specify your location and select a date, from there M-Predictor will enter the data in the model and the risk will be obtained as a result.
Regarding the tools, hardware and software used to develop the project, there is python, the language in which all the logic of the model was carried out. Regarding the front-end we use a library called leaflet, which allows interactive maps with javascript.
We have developed a tool that monitors cases of COVID-19 mutations in order to make a risk prediction, this is very important because it brings benefits such as support in decision-making by public health entities, in addition, informs the people to take their respective care.
We hope that this tool will be of great help to all people considering all the problems that COVID-19 has brought and the appearance of new mutations. Our goal is to inform about the risk factor for Covid Mu, Alpha and Gamma mutations
Space Agency Data
What space agency data you used in your project
Part of our data comes from Earth Data, from there values such as temperature, precipitation and humidity were taken, these data are related to external data of the cases of COVID-19 mutations in Colombia.
The data we had access to was important for our project, because thanks to this, the model was trained to finally obtain M-Predictor.
Hackathon Journey
The experience in Space Apps were challenging, The idea of create a solution in a weekend was something to big for the short amount of time, but was also gratifying because gave us the possibility to work in a group of people form different parts and with different knowledge to build a solution together.
We learn about the new variants and the distribution of those variants around the Country, We also learn to do a team work in a team with different abilities and different cultures to build a solution for a problem using new techniques as machine learing integrating those technologies with the page.
We are motivated to contribute a tool that will inform people about the risk of COVID-19 mutations, using artificial intelligence.
With the situation that has been and is still experienced in the world due to the pandemic, our team knew that it would be incredible to train and apply a model to detect the risk of COVID-19.
However, when investigating we discovered that the risk of COVID-19 mutations had not been very worked, which inspired us to work on it.
To solve the problem we used the knowledge of each member in the group toghether to think how to messure and solve the challenge, then we decided the tasks between all the members to solve the problem in the least amount of time.
Our team wants to thank the National Institute of Health, because they will guide us in the process of calculating the risk factor and also, they gave us recommendations to carry out a project that could have more impact.
References
Data and Resources
- https://data.nal.usda.gov/dataset/cropscape-cropland-data-layer/resource/035d2848-dacd-48d2-be92-b5379fc8eb4d
- Shapefiles: https://github.com/calvinmetcalf/shapefile-js
- Resources to download NASA earth data: https://search.earthdata.nasa.gov/search?m=-0.0703125!0.0703125!2!1!0!0%2C2
- Filtered out for Colombia: https://ladsweb.modaps.eosdis.nasa.gov/search/order/4/MOD05_L2--61/2021-04-01..2021-08-12/DB/Country:COL
- https://ladsweb.modaps.eosdis.nasa.gov/archive/orders/501658900/?process=ftpAsHttp&path=orders%2f501658900
- Mutation Dataset for Colombia: https://microreact.org/project/nwAwK6rDzoLbmBY5P1VGsj/f6583326
- http://www.ins.gov.co/BibliotecaDigital/Estrategia-caracterizacion-genomica-muestreo-probabilistico-SARS-CoV-2-Primer-muestreo.pdf
- https://rsaiconnect.onlinelibrary.wiley.com/doi/epdf/10.1111/rsp3.12469
Tags
#mutations #covid #variants #delta #alpha #gamma #deeplearning #machinelearning #ai #risk
Global Judging
This project has been submitted for consideration during the Judging process.