Virus Alert

https://drive.google.com/file/d/1OcrZyKJwnvRQPwqWu20SsPy84NLOZFRV/view?usp=sharing

https://covid19-risk-factor.herokuapp.com/

Our aim's to create an easy user interface with an interactive model to aid in suppressing the spread of the pandemic by raising the awareness of the public and keeping them updated with the latest Covid- 19 news, statistics.

It can provide risk warnings based on the location and real-time meteorological factors such as Temperature, Atmospheric Pressure, wind speed, and humidity.

The application also offers a personalized experience based on an algorithm that assess the probability of infection based on current symptoms the user experience .

It also estimates the severity and risk of hospitalization according to the symptoms and users profile, whether they suffer from chronic diseases like CKD, high blood pressure, diabetes ..e. t. c. Our application is classified into three main systems; the dashboard, the warning system, and the diagnostic system.

[1]The dashboard & latest statistics;

The app starts with multiple slides that introduce new users to the app and the offered service. Once the user is signed up, he/she will be asked to insert general information and other health-related data like chronic diseases and medical history. The data can be later used in the application in severity calculations. In the dashboard, there is a summary of the latest statistics in the registered country in a simplified convenient way for the user to facilitate information delivery. All the contained information is automatically updated and collected from reliable and scientific resources. The icons appear indicating the daily cases the active cases and death. Statistics appear in easy simple graphs and charts.

[2]Warning system:

Several studies have indicated the correlation between various climate factors such as temperature, pressure, wind, humidity, and covid-19 transmission. Our aim was to estimate the influence of meteorological factors on morbidity and covid-19 risk to warn our users about the possible risk using real-time data of the weather condition. Data of daily COVID-19 cases in Egypt were collected from [1]. The collected Data ranged from date; March 3, 2020, to September 29, 2021, aqueduct data was collected to offer the highest accuracy possible. Temperature, humidity, wind, and pressure data were also collected using the NASA power website [2]. Four climate variables were tested for correlation: Temperature at 2 meters, Atmospheric Pressure, wind speed, and humidity. The temperature at 2 meters, Atmospheric Pressure, and humidity were found significantly correlated with P=0.0028, P=4.9*10^-15, and P=0.000024 , respectively while wind speed was not significantly correlated. The data were analyzed by STATA 17.0 via a linear regression model.

The coefficients of each factor form the regression table were substituted in the following formula;

Y = a + b1T + b2H+ b3P

Where a is the intersection constant, T is the temperature in Celsius,   H is the relative humidity, and P is the pressure in kpa and Y is the predicted value of cases.

The highest number of cases observed in Cairo was 563 on Jun 18, and the lowest number was 0. As result, a scale was created ranging from 0 to 500. The average number of cases was estimated to be 168 . we consedred higher cases than the average(168) will be considered as high risk . The user will be notified by the risk level and adviced to take precausion and avoid meeting.

Diagnostic system:

 We developed a diagnostic system that estimates the probability of testing positive for covid-19 based on the user (symptoms, age, gender, and whether he is in contact with the covid-19 patient at home or not) using an equation which was published on [3] resulted from the logistic regression method combined with the upsampling balancing strategy

Prediction= - 1.078 + (1.309 * loss of smell) + (0.481 * fever) + (0.407 * covid at home) + (0.338 *shortness of breath) + (0.237 * myalgia) + (0.153 * cough) + (0.035 * nausea) + (0.033 * gender) + (0.008* age) – (0.441 * sore throat) – (0.227 * coryza) – (0.045 * diarrhea)

Probability of testing positive = e^ prediction /1 + e^ prediction where all symptoms are coded as 1 if the person self-reports the symptom and 0 if not. The gender feature is also binary, with 1 indicative of male participants and 0 representing females. According to the input that will be obtained from the user, a message will pop up with the probability of catching covid-19 and a suggestion to follow CDC's information on COVID-19 testing. We will not only estimate the probability of testing positive but also estimate the severity risk of hospitalization using the odd ratios for hospitalization that were published on [4] which explained that the people who are exposed to these risk factors are more likely to be hospitalized compared to whom are not exposed to these factors that associated with sex, age, race, BMI, exposure, symptoms and underlying health conditions. We estimated the upper value that the user could achieve according to the sum of high odd ratio of each category (supposedly that the user suffers from all the health conditions and is exposed to all the exposures). The obtained value that will be calculated according to the user input will be divided by the upper value then multiplied by 100 to obtain the severity risk of hospitalization. we grouped the severity risk into three categories - high, medium, and low risk. From 0% to 33.33% considered as low risk, 34.33% to 66.66% considered as medium risk and from 67.66% to 100% considered as high risk. if the user is at risk to be hospitalized, a message will pop up showing the places where he should avoid. As wearing mask reduces the risk of catching COVID-19 by 67% and hand hygiene reduces the risk by 53%, an advice will pop up.

The software;

We have developed a hybrid application (PWA) that works on all devices, whether phones (Android - IOS) or desktop systems, for example, Windows and Linux. it contains the following features:

- Login system

- Registration system (to collect some related user info)

- Diagnostic System and calculate risk factor(based on equations and algorithms from reliable sources mentioned below)

- Push notification system

- Latest news from worldwide about covid19

On the first use of the user, he enters the link of the application, and a notification appears at the bottom of the page (in case he opened the download link from the phone) whether he wants to install the application on his device or not. By clicking on the following to go to the membership registration page, where we take some sick data about him to put on his page. After the data is successfully registered, the user can access his control panel by logging in, and then he can go to the diagnosis page where the Covid-19 risk factor is calculated by By answering some questions, and by answering these questions, the user can obtain the rate of his infection with the Coronavirus, and the user can follow the latest Corona news through the News tab, which shows him all the updates. unfortunately we couldn't finish the notification system because of the shortage of the time (we built the Backend and the static Frontend but we have not integrated them)

• Our coding languages: FRONTEND > JS (React.js, chart.js) == backend > python (Django, pandas, celery, channels)
• APIs that we use >

https://disease.sh/docs/#/COVID-19%3A%20Worldometers/get_v3_covid_19_countries

api.openweathermap.org/data/2.5/weather?q={city name}&appid={API key}

https://rapidapi.com/microsoft-azure-org-microsoft-cognitive-services/api/bing-news-search1/

Front end repo => https://github.com/moussa32/NASA-Space-Apps-Cairo-2021-Project

Back end repo => https://github.com/osama614/virus-alert-backend

Reference;

1.“Covid-19 situation update for the EU/EEA, as of 1 October 2021,” European Centre for Disease Prevention and Control, 01-Oct-2021. [Online]. Available: https://www.ecdc.europa.eu/en/cases-2019-ncov-eueea. [Accessed: 03-Oct-2021].

2. “NASA Power,” NASA. [Online]. Available: https://power.larc.nasa.gov/. [Accessed: 03-Oct-2021].

3.L. F. Dantas, I. T. Peres, L. S. L. Bastos, J. F. Marchesi, G. F. G. de Souza, J. G. M. Gelli, F. A. Baião, P. Maçaira, S. Hamacher, and F. A. Bozza, “App-based symptom tracking to optimize SARS-COV-2 testing strategy using machine learning,” PLOS ONE. [Online]. Available: https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0248920. [Accessed: 03-Oct-2021].

4 .S. C. Knight, S. R. McCurdy, B. Rhead, M. V. Coignet, D. S. Park, G. H. L. Roberts, N. D. Berkowitz, M. Zhang, D. Turissini, K. Delgado, M. Pavlovic, A. D. N. A. S. Team, A. K. H. Baltzell, H. Guturu, K. A. Rand, A. R. Girshick, E. L. Hong, and C. A. Ball, “Covid-19 susceptibility and severity risks in a survey of over 500,000 people,” medRxiv, 01-Jan-2020. [Online]. Available: https://www.medrxiv.org/content/10.1101/2020.10.08.20209593v1. [Accessed: 03-Oct-2021]. 

We used the sites and resources provided by the competition, which contributed greatly to the process collecting data and linking them together through relationships, tables and equations. Such as the sites https://power.larc.nasa.gov/

Which provided us with an integrated database of climate states and their coefficients. The site also enabled us to display the weather condition that's able to be downloaded in an Excel file within the time range we choose, that helped us to apply more accurate equations to produce accurate results in the process of calculating the risk factor.

We also made use of research papers to better understand the impact of climate and other factors on the spread of the virus.The earth observing dashboard was also used to find out the population density and its impact on the spread of Covid-19 and the increase in the infection rate. It also provided us with accurate statistics of the Corona virus.The earth observing dashboard was also used to find out the population density and its impact on the spread of Covid-19 and the increase in the infection rate. It also provided us with accurate statistics of the Corona virus. https://eodashboard.org/?poi=W1-N1&indicator=N1

Just as spaceflight journeys, a race against time, directs our greatest potential and forces us to have the courage to discover it and to wade through small opportunities to make an impact.

We learned to work as a team and distribute tasks efficiently, we learned time management and applied the concept of the threefold concept of time, quality, and resources, and of course, we learned from our different fields.

We took the opportunity of the challenge to create an application that contributes to monitoring the danger of the Coronavirus because we wanted to be part of the innovators of solutions to this pandemic that still occupies the current era. To achieve this goal, we were keen to design the project to make the responsibility a decision that every person takes for the world and himself, in a simple way that helps him to be an influential member of the goal that unites us and the users.

Like all trips, obstacles must arise, which was mainly time, we have excelled in this by distributing the main tasks with high efficiency, shifting and assisting with secondary tasks to ensure that each of us receives sufficient rest to be able to focus on his mission.

The ones who deserve to be thanked is every member of our team who did their job perfectly and made new friends that are united by this great goal.

And of course, we thank our mentor for her support, and we thank NASA Space app for giving us this great opportunity.

Thank you

Reference;

1.“Covid-19 situation update for the EU/EEA, as of 1 October 2021,” European Centre for Disease Prevention and Control, 01-Oct-2021. [Online]. Available: https://www.ecdc.europa.eu/en/cases-2019-ncov-eueea. [Accessed: 03-Oct-2021].

2. “NASA Power,” NASA. [Online]. Available: https://power.larc.nasa.gov/. [Accessed: 03-Oct-2021].

3.L. F. Dantas, I. T. Peres, L. S. L. Bastos, J. F. Marchesi, G. F. G. de Souza, J. G. M. Gelli, F. A. Baião, P. Maçaira, S. Hamacher, and F. A. Bozza, “App-based symptom tracking to optimize SARS-COV-2 testing strategy using machine learning,” PLOS ONE. [Online]. Available: https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0248920. [Accessed: 03-Oct-2021].

4 .S. C. Knight, S. R. McCurdy, B. Rhead, M. V. Coignet, D. S. Park, G. H. L. Roberts, N. D. Berkowitz, M. Zhang, D. Turissini, K. Delgado, M. Pavlovic, A. D. N. A. S. Team, A. K. H. Baltzell, H. Guturu, K. A. Rand, A. R. Girshick, E. L. Hong, and C. A. Ball, “Covid-19 susceptibility and severity risks in a survey of over 500,000 people,” medRxiv, 01-Jan-2020. [Online]. Available: https://www.medrxiv.org/content/10.1101/2020.10.08.20209593v1. [Accessed: 03-Oct-2021]

https://eodashboard.org/?poi=W1-N1&indicator=N1

#virus alart #covid19, #covid19_Risk_Factor, #covid19_Risk_Factor_calculator

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