Omdena
Published in

Omdena

COVID’19 Analysis in Pakistan

Under the Umbrella of Omdena Pakistan Chapter.

Abstract

During the current worldwide COVID19 pandemic, public health measures have been taken to reduce community transmission of the virus, including staying home orders, isolation, and social distancing measures. In addition, these limitations also have important neuropsychiatric consequences. Initial analysis of coronavirus virus 2019 (COVID19), is required in controlling this disease. COVID19 is spreading around the world in no time. The vaccines against this virus have almost stopped the inevitable disaster. Fast and systematic screening for COVID19 is possible with CT Scan Images. The Computer Vision techniques are used in the proposed method. This project focuses on detecting CT Scan Images and X-ray images of COVID19 and non-COVID 19 CT using different deep learning techniques. Doctors can use automated diagnostics of COVID19 from CT scans and X-rays as a fast and efficient method for testing for COVID19.

COVID’19 Analysis Pakistan

The Background

The economic impact of the COVID19 pandemic in Pakistan has been largely disconcerting. According to the Ministry of Statistics, Pakistan’s growth in the fourth quarter of fiscal 2020 decreased by 3.1%. If we could map vaccine shortages and absences in different states of Pakistan against vaccine waste across the country, that could help increase the effectiveness of vaccinations.

Much of the death rate from covid19 in Pakistan is due to the early failure to identify the presence of the coronavirus and its symptoms. Identifying the symptoms of Covid19 and mapping its severity will not only make the diagnosis of viral illness more rapid and effective but will also help reduce drug-induced mortality much earlier.

The Problem

A 4-weeks project with the goal, building a model to analyse and model the change in economic trends over the past couple of years in Pakistan due to the outbreak of covid and correlate it with various factors that contribute to the rise and fall of the economy. Objectives of the project include building computer vision models to early detect the presence of covid through X-ray images or CT scans and classify the severity of Covid-19. The project is made open-source and the results obtained are helping in creating awareness about the spread of Covid19 and its impact on the Pakistan economy.

The Project Goals

The goal of the project is to develop a “Web Application” that assists Drs in identifying the symptoms of Covid-19 through X-ray and CT scans in a short time, to count the unemployment rate during the Covid19, and to analyze the most recent features of Covid-19 through tableau.

Week 1: Collect covid related data from the various up to date sources

The week 1 task is based on a collection of the data for the Analysis of Covid-19 is collected from different sources which include Our World in Data, Kaggle, World Health Organization, Humanitarian Data Exchange, Open Data Pakistan, and The Pakistan Government Website of Covid 19 Analysis.

Collected Dataset

Week 2: Data Pre-processing and Exploratory Analysis

Pre-Processing

The task for Second Week was to do Pre-processing of the dataset of Covid-19 Pakistan. During analysis of the shared dataset, the working team found 80 blank records during the early 4 months.

Pre-processed Data

Exploratory Data Analysis

Exploratory Data Analysis was also done on the dataset to take out meaningful insights from the data.

EDA

Tableau

The participants started working on the dashboard on tableau using the same dataset.

Data Modelling

To solve this real-world data Hierarchical Technique of data modelling was used which can be seen below in the image.

Hierarchical Data Modeling

Week 3: An interactive Plot /Map displaying current covid analytics of each state

To better analyse current Covid-19 different types of plots and maps were generated, so that maximum meaningful data is accessed. Below you can see one of the charts that is displaying the covid19 situation of each state.

Current Covid Analytics

AlexNet Model

AlexNet is used to load the pre-trained version of the network trained on more than a million images. The pre-trained network classified images into 1000 object categories depending on the symptoms.

Week 4: Forecast the relation between covid vs economic change in past time and estimate the effect of covid in the future through Time-series analysis.

Pakistan TimeSeries Analysis

To analyse Pakistan TimeSeriesn the dataset of Pakistan Economy was utilized, the data consists of features such as GDP, CPI, and Unemployment Rate. The model LSTM from RNN is used to create the forecast of the time series.

According to the Investmentopedia definition of GDP, economists typically use the Gross Domestic Product (GDP) to measure economic progress. If GDP grows, the economy is strong and the country moves forward. On the other hand, if the gross domestic product falls, the economy may suffer and the country loses its footing. As we see the trend of the data from 1991 to 2020, the GDP starting in 2019 when the Covid19 pandemic occurred had a huge impact on the economy of Pakistan. The unemployment rate has increased over time, especially after the covid19 pandemic, it has increased.

GDP Annual Growth of Pakistan

Unemployment Rate

Covid-19 has badly affected the employment rate. The share of people without a job was increased and 1.5 million people applied for a peon position recently.

Prediction of Covid19 Through Computer Vision

  1. Predict the likelihood of covid positivity from chest CT Scan images and respiratory patterns. From demographic and clinical data (the patient’s age and sex, exposure history, symptoms, and laboratory tests) were put into a machine-learning model to classify COVID-19 positivity.

This application is developed to identify the symptoms of Covid-19 through CT scans of the patient. The user has to open the web app and browse the CT scan, then click the “Predict” button to see the Covid-19 result in percentage.

The CT-scan result below shows the person has 73.06% of Covid-19 symptoms.

3. A web application to visualize the severity of covid from chest X-ray images

In the end, the participants have successfully deployed the Web app. Below you can see two images of the Web app of Covid Diagnosis Through X-rays. The web app carries an option “Browse Files”, the user just needs to put the x-ray here and the results will be shown just after analysis.

Web App

The X-ray uploaded in the web app below shows that the person is Covid-19 Positive.

Analysis Through X-ray

Tableau of COVID-19 Pakistan

Here is the final presentation through Tabluea. The Tableau COVID-19 Data Hub contains resources to help you visualize and analyze the most recent data on the COVID-19 outbreak.

Covid-19 status of Pakistan 2020–2021

How to contact Omdena Pakistan Chapter?

Omdena Pakistan chapter is highly active on social media platforms to assist you with any concerned issue. You can follow us on bellow mentioned social accounts to stay updated about ongoing projects and workshops.

Facebook: Omdena Pakistan Chapter

Linkedin: Omdena Pakistan Chapter

Github: Qasim Hassan

Medium: Iqra Anwar

--

--

--

Omdena is a collaborative platform for building AI solutions to real-world problems through participating in AI and Data Science projects.

Recommended from Medium

Our official publication

Data Readiness and Quality: The Big New Challenges for all Companies

The data roundup — week 2021–10–15

Components of Data S

Contextualizing Airbnb by Building Knowledge Graph

Save a life or receive cash? Which do recipients want?

Improving engagement of analytics users through gamification

Kubeflow is your perfect Machine Learning workstation

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Iqra Anwar

Iqra Anwar

BSCS graduate from BBSUL Karachi, Pakistan. Writer at Omdena Pakistan Chapter | I write articles on Computer Science | Blogs on Travel, Fashion, and more.

More from Medium

Association Rule Learning(ARL)

Place Your Products with Their Mates: Association Rule Learning

The Math behind Rapid At-home Covid-19 Test Accuracy

Pokemon Competitive Analysis Project