Starflock

Cleveland, OH | COVID-19: Calculate the Risk

COVID REAL-TIME RISK METER

High-Level Project Summary

Develop a mobile app to help users around the world gauge in real-time their risk for contracting SARS-Cov-2, their risk for hospitalization, and their risk for death based on their location and some basic user information.

Link to Project "Demo"

https://docs.google.com/presentation/d/1EaQ-1AlTuwje3mqk3fv55ou0NNp3r7MMeVU0swUWk-M/edit?usp=sharing

Link to Final Project

https://docs.google.com/presentation/d/1EaQ-1AlTuwje3mqk3fv55ou0NNp3r7MMeVU0swUWk-M/edit?usp=sharing

Detailed Project Description

Mobile app will combine user data, climate data, social data, and real-time local crowd estimates to provide a real-time risk of contracting SARS-CoV-2 as well as the risk for severe illness and death.

Environmental Factors Affecting SARS-CoV-2 Transmission

Average Air Temperature: Negative Correlation

Decreased transmission with increased temperature.
1C increase in minimum temp = 0.86% decrease in n cases. (Eslami et al., 2020 AMB Express)
Data:https://www.ncei.noaa.gov/metadata/geoportal/rest/metadata/item/gov.noaa.ncdc%3AC00532/html

Average Relative Humidity: Positive Correlation

Increased transmission with increased relative humidity (Raines et al., 2021 PLOS ONE)
Data:https://www.ncei.noaa.gov/metadata/geoportal/rest/metadata/item/gov.noaa.ncdc%3AC00532/html

Average Daily Sunlight: Negative Correlation

Decreased transmission with increased sunlight (especially UVB).
Two-fold effect: Direct sunlight degrades viral particles and increases Vitamin D production, which promotes the immune system. (Sharun et al., 2020 Annals of Medicine and Surgery)
Data:https://ceres.larc.nasa.gov/data/

Social Factors Affecting SARS-CoV-2 Transmission

Population Density (current data): Positive Correlation

Increased transmission with increased population density (urban vs rural)

Vaccination Rate (user input & current data): Negative Correlation

Decreased transmission with increased vaccination rate.
Data:https://data.cdc.gov/Vaccinations/COVID-19-Vaccinations-in-the-United-States-County/8xkx-amqh

Mask-Wearing Rate (user input & current data): Negative Correlation

Decreased transmission with increased mask-wearing.

Hospitalization Rate (current): Positive Correlation

Increased hospitalization indicates increased transmission in that area
Data:https://healthdata.gov/Hospital/COVID-19-Reported-Patient-Impact-and-Hospital-Capa/anag-cw7u

School-aged Children at home (user input): Positive Correlation

Increased transmission when school-aged children live in the home.

Social Factors Affecting SARS-CoV-2 Severe Illness/Death

Available ICU Beds: Negative Correlation

Increased risk for severe illness or death with decreased number of available ICU beds.
Data:https://healthdata.gov/Hospital/COVID-19-Reported-Patient-Impact-and-Hospital-Capa/anag-cw7u

Age (user input): Positive Correlation

Increased risk for severe illness or death with increased age.

Health Status (user input): Positive Correlation

Increased risk for severe illness or death in people who have certain health issues.
Health issues that cause the highest risk:
Cancer, chronic kidney disease, chronic lung diseases (COPD, asthma, cystic fibrosis, pulmonary hypertension), dementia, diabetes, heart conditions, HIV infection, immunocompromised, liver disease, obesity, pregnancy, sickle cell anemia, smoking/substance abuse, organ transplant, and stroke. (CDC)

Real-Time Risk via Local Crowd Size Measurements

Real-time Twitter activity within a certain radius of the user’s current lat/lon coordinates to estimate local crowd and provide a more accurate risk of contracting SARS-CoV-2. Data via Twitter API (point_radius parameter; JSON)

Reference: Botta et al. 2015 Royal Society Open Science

Bluetooth technology could be used for even more local crowd size estimates: determine the number of unique bluetooth signals within a ~30 foot boundary.

*** ALGORITHM METHOD ***

Using the datasets mentioned, calculate correlations between each factor listed and the current number of cases at the county-level resolution. Use each correlation in a model to determine overall risk.

_DEMO_

Features:

Risk Warning Label (HIGH, MEDIUM, LOW) and when the risk was last updated.
Map of user’s location (with risk heat map)
Guide: Provide recommendations to reduce the user’s risk of contracting SARS-CoV-2 based on risk level.
Profile: Where the user provides their information (vaccinated, wears a mask, medical conditions, age, school-aged children)

Space Agency Data

NOAA/NCEI (climate data)

NASA/LARC/CERES (solar data)

Hackathon Journey

Did a solo project this year. I always enjoy the challenges. Some years I can motivate my programming friends to join but this year was difficult to mobilize the group. Instead, I focused on a concept. It was fun learning about the less obvious risk factors for contracting SARS-CoV-2.

References