Calculador-COVID19

🚀 Technologies

This project was developed with the following technologies:

• C
• libAllegro
• R

💻 Project

This was my first project upon entering university, a COVID-19 calculator that calculates the probability of death for COVID-19 patients based on their comorbidity history. It can be found on youtube at the link: https://youtu.be/dEZyf03uDvI

:memo: Info

This was my first project from the information systems course, it was developed in C and the graphic library Allegro was used.

The project is a calculator for the COVID-19 virus, the calculator calculates the probability of death of the patient in case he/she contracts the COVID-19 virus. The model used for the calculation was the Logistic Regression model, and the data was formulated in R. As a source a sample of Alagoas from 06/06/2020 that contained 15,537 patients was used.

Below is the formula:

• 1/(1+exp(-1*(-7.09+0.4456sex(1 male, 0 female)+0.058age + 0.9866 * diabetes(1 has diabetes, 0 does not) + 0.9176 * hypertension(1 has hypertension, 0 does not) + 0.5628 * elderly(1 if elderly, 0 if not) + 1.6053 * obesity (1 if obese 0 if not)))

This calculation returns a value between 0 and 1 because the probability result can only be from 0 to 1

Here are some interesting considerations about the program:

Logistic regression is a non-linear regression, i.e. it is possible to predict the value of this response variable whether it is 0 or 1

In it we have the response variable that in this case is the patient's death, that is, 0 or 1

The objective of the logistic regression is to predict the probability of occurrence of the response variable based on the explanatory variables (gender, age, diabetes, hypertension, >60 years, obesity.)

These were the explanatory variables selected at the 99% statistical confidence level (this was possible by predicting the regression model, and these explanatory variables explain the outcome of the response variable)

The regression model was made in "R" which is a statistical technology.

The average probability of death of the people who died was 15% chance of dying.

The average of the patients who didn't die had a 3% chance of not dying.

There are infinite combinations because I didn't limit the age

• 44% of the people who received a 50% chance of dying actually died.

• 3% of the people who received probability below 50% actually died.