Eating the frog

Lima | COVID-19: Calculate the Risk

COVID Tracker

High-Level Project Summary

COVID Tracker is a mobile application that predicts the risk of COVID19 infection in real-time. We are solving the challenge by calculating the probability of getting infected by the virus and then advising the user on how to stay safe. This solution has the potential to reduce the spread of COVID-19 infections by promoting a prevention culture in our society. Moreover, it has a high impact in Peru because our code is capable of taking information related to COVID from different regions of the country. The model for calculating the percentage of contagion risk takes into account epidemiological and environmental variables at the user’s current location, as well as his vaccination status.

Link to Project "Demo"

https://www.youtube.com/watch?v=u9Qx7hBVnjQ

Link to Final Project

https://sites.google.com/utec.edu.pe/eatingthefrog/about

Detailed Project Description

The project called “COVID Tracker” is a mobile application that calculates an infection risk by introducing personal data and geographical data (by GPS or manually). After obtaining this percentage, general and personalized recommendations appear on the user screen and can be displayed at any moment. A feature of the app is called “Global facts” and it links graphs of relevant COVID variables around the world so any user can choose and see information of a determined country; this data would be collected from the ESA, specifically the RACE database.

The model for calculating the percentage of contagion risk is based on a multi-layer approach with epidemiological, environmental, and other variables (such as reinfection rate and vaccination status), supported by the most recent literature. 

The OpenCovid Peru web portal was used to obtain pandemic status data for each region of the country and the Waqi portal was used to obtain data on environmental variables. Both of them updated on a weekly and daily basis respectively. 

The whole project consists of three main parts: statistics model, backend, and frontend. For ease of use, Python was selected as the main programming language to develop the model and backend. The first one was made using popular data science libraries such as Jupyter, Matplotlib, Seaborn, Scikit learn among others, and following best python coding practices such as modularity, relative imports, and so on. The backend is used as an API between the model and any potential consumer using the REST model, HTTP requests, and the JSON format. The third component consumes this API and shows the information to the user in an intuitive manner using popular concepts such as accessibility, UI, and UX. This one was made using the Typescript programming language as the core. To provide interoperability between major mobile platforms React Native was used as the main library. Again, most of the recommended coding practices in React and Typescript were used including reuse of components, modularity, function components, hooks, etc. 

To host the API we used Heroku as our main hosting provider. This allows the use of the model in any potential consumer with an internet connection. 

In order to keep a log of the project changes and improve team collaboration, Git and Github were used as the main version control system. 

Finally to provide a brief overview of each codebase a README.md file was created using markdown and including the most relevant information about each component and its repository.

The diagrams (sequence diagram and use case diagram) were made in Miro and LucidChart.  

Space Agency Data

RACE-ESA (RAPID ACTION ON CORONAVIRUS AND EO - EUROPEAN SPACE AGENCY)

We will use mobility data from this database. Additionally, in the "Home" section of our app, there is an option to display graphs of relevant COVID data worldwide; this option is expected to be linked to RACE-ESA database.

Local database: OpenCovid Peru

Provides data on the current pandemic status in Peru by region. This database contains the most updated and reliable information in the country.

Waqi

Provides real-time data on air quality around the world.

openQA

A repository of database used to collect historical data of air-quality data.

Hackathon Journey

The Space App Challenge was a great event for all of us because it permitted us to learn how to create a viable solution in a short time. Moreover, we gained soft and hard skills. For instance, we learned how to work efficiently in a team and how to communicate science ideas with each other. In the hard skills we learned how to code in Python, how to design an app mockup taking into consideration UI and UX criteria and also, we learned how to build up our solution in a context where everything is virtual. Our team chose the Covid challenge due to the current situation. Particularly, the difficult pandemic situation faced by Peru was our main inspiration. One challenge was learning software engineering terminology since most of us come from a different educational background but we were able to solve it thanks to Mario’s help. Finally, a difficulty we had was to find enough local literature to develop our model; this was solved by further research in international literature. 

References

Data:

Papers:

  1. E. E. Félix-Arellano, A. Schilmann, M. Hurtado-Díaz, J. L. Texcalac-Sangrador, and H. Riojas-Rodríguez, “Revisión rápida: contaminación del aire y morbimortalidad por Covid-19,” salud publica mex, vol. 62, no. 5, pp. 582–589, Sep. 2020.
  2. M. V. Beusekon, Apr 12 and 2021, “Previous COVID-19 may cut risk of reinfection 84%,” CIDRAP. https://www.cidrap.umn.edu/news-perspective/2021/04/previous-covid-19-may-cut-risk-reinfection-84 (accessed Oct. 02, 2021).
  3. M. V. Beusekon, Jul 22 and 2021, “Study: 2 COVID vaccine doses much more effective than 1 against Delta,” CIDRAP. https://www.cidrap.umn.edu/news-perspective/2021/07/study-2-covid-vaccine-doses-much-more-effective-1-against-delta (accessed Oct. 02, 2021).
  4. M. Rochabrun and R. Liu, “Peru study finds Sinopharm COVID vaccine 50.4% effective against infections,” Reuters, Aug. 13, 2021. Accessed: Oct. 02, 2021. [Online]. Available: https://www.reuters.com/world/americas/peru-study-finds-sinopharm-covid-vaccine-504-effective-against-infections-2021-08-13/
  5. “Coronavirus: vacunas contra la COVID-19 en el Perú.” https://www.gob.pe/11571-coronavirus-vacunas-contra-la-covid-19-en-el-peru (accessed Oct. 02, 2021).
  6. “Lo que debe saber sobre la vacuna BNT162b2 de Pfizer-BioNTech contra la COVID-19.” https://www.who.int/es/news-room/feature-stories/detail/who-can-take-the-pfizer-biontech-covid-19--vaccine (accessed Oct. 02, 2021).
  7. “La vacuna de Oxford/AstraZeneca contra la COVID-19: lo que debe saber.” https://www.who.int/es/news-room/feature-stories/detail/the-oxford-astrazeneca-covid-19-vaccine-what-you-need-to-know (accessed Oct. 02, 2021).
  8. “Todo lo que se debe saber sobre la vacuna de Sinopharm contra la COVID-19.” https://www.who.int/es/news-room/feature-stories/detail/the-sinopharm-covid-19-vaccine-what-you-need-to-know (accessed Oct. 02, 2021).
  9. “HHS Protect Public Data Hub.” https://protect-public.hhs.gov/ (accessed Oct. 02, 2021).

Tools:

  • GitHub
  • Jupyter
  • Python
  • Java
  • Heroku
  • Flask
  • React Native
  • Figma
  • Canva
  • Lucidchart
  • Miro

Tags

#software, #app, #covid19, #peru

Global Judging

This project has been submitted for consideration during the Judging process.

COVID-19: Calculate the Risk

COVID-19 continues to be a global problem even though vaccination efforts are underway to control its propagation. Your challenge is to use environmental data and other information (such as epidemiological, social, policy, and economic data) to build a smartphone application that provides individualized, geolocated, COVID-19 risk warnings to guide social awareness, response, and health security.