A DATA DRIVEN UNDERSTANDING OF COVID-19 DYNAMICS AND IDENTIFICATION OF THE POSITIVE COVID-19 CASES USING GENETIC ALGORITHM

Soorayadas Sudhakaran (Roll no.: B180494PE)

Farshad O P (Roll no.: B180415PE)

Rohan Raj (Roll no.: B180420PE)

Deepak P (Roll no.: B170198PE)

Sanal P (Roll no.: B180803PE)

CONTENTS

• Introduction
• Literature Review
• Methodology
• Results
• Conclusions
• Future scope
• References

​

INTRODUCTION

• Coronaviruses are a family of viruses that can cause respiratory illness in humans.

​

• The new strain of coronavirus, COVID-19, was first reported in Wuhan, China in December 2019. The virus has since spread to all continents

​

• This highly infectious disease transmit from person to person through respiratory droplets produced by infected person

​

• The measures thus taken include social distancing, restrictions on domestic as well as international travel, cancelling social events, shutting down of public as well as commercial activities etc.

​

INTRODUCTION (Contd…)

• Various technological measures and data analytics and data scientist are working to come out with this novel solution

​

• Here we study the extension of contact tracing app wherein the cluster of users are identified who all are come in contact with the Covid-19 infected person but also after taking the list, their symptoms are matched with the infected person

​

• The new algorithm has been developed GABFCov-19 which is based on genetic algorithm.

​

• This algorithm will generate high probability positive cases. So, that it can reduce the community spread.

​

LITERATURE REVIEW

LITERATURE REVIEW (Contd…)

METHODOLOGY

Contact tracing app: The Contact tracing app gives us the cluster of the entity which are came in contact with infected person within 14 days.

Figure 1: working algorithm of Contact tracing Proximity App

METHODOLOGY (Contd…)

On the basis of the cluster found by using contact tracing app, the symptoms parameter has been matched by using genetic algorithm

Figure 2:Working Methodology of GABFCov-19

METHODOLOGY (Contd…)�

• Algorithm assigns the fitness of the entity depending upon its occurrences in the symptoms that affect the covid-19.

​

• The fitness has been computed according to the matching parameters. The entity whose more parameters matched, the more will be the fitted entity.

​

• High fitness value represents the high probability of infected Covid-19 patient.

• Suppose that binary chromosomes of the positive case is 1 1 1 0 1 1
• Entity found in the cluster has been represented by the chromosomes as 1 1 0 0 0 1
• By matching parameters in both chromosomes Fitness value = 4

METHODOLOGY (Contd…)

Steps of GABFCov-19 Algorithm

• Input: Cluster’s entities
• Output: The number of expected high probability positive cases

​

STEP 1

• Initialize the parameters

​

STEP 2

• Randomly generate the m chromosomes

STEP 3

• Calculate weights to perform the Weighted Fuzzy C Means Clustering

​

STEP 4

• Perform the genetic process of selection, mutation and cross over

​

STEP 5

• The process continue until the condition meet the genetic termination

Steps of GABFCov-19 Algorithm

STEP 6

• Compute the new generation fitness of the cluster and compare with each individual to find the highest fitness of individual.

​

STEP 7

• Fitness = Matching parameters

Steps of GABFCov-19 Algorithm

RESULT

Figure 3a:Cluster 1 based on location

Figure 3b: Cluster 2 based on location

RESULT (Contd…)

• The clusters earlier have been passed as input to the GABFCov-19 algorithm.

​

• They are converted in to binary chromosomes.

​

• The binary chromosome generated has been run over the parameters mentioned in step 3 of the proposed algorithm to find the high probability positive cases.

RESULT (Contd…)

​

• Age

• Sensitivity

• Past case history

• Fever

• Cold & Cough

• Breathing Problem

• Tiredness

​

• Chest pain or pressure

• Loss of speech or movement

• Headache

• Loss of taste or smell

• A rash on skin, or discolouration of fingers or toes

• Aches and pains

• Sore throat

• Diarrhoea

TOTAL PARAMETERS: 15

Parameter tested on 12 individuals

[[0. 0. 0. 0. 0. 1. 1. 0. 0. 1. 1. 1. 1. 1. 1.]

[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]

[0. 0. 0. 0. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0.]

[1. 1. 0. 1. 0. 1. 0. 0. 0. 1. 1. 0. 0. 0. 0.]

[1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]

[1. 0. 0. 1. 1. 1. 1. 0. 1. 1. 1. 0. 1. 0. 1.]

[1. 0. 1. 1. 0. 0. 1. 0. 1. 0. 0. 0. 1. 0. 0.]

[0. 1. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]

[0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0.]

[0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0.]

[0. 1. 1. 1. 1. 1. 0. 1. 1. 0. 0. 1. 1. 1. 1.]

[1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 0. 0. 1. 1. 1.]]

RESULT (Contd…)

RESULT (Contd…)

Chromosome of the positive case generated is taken from the cluster

​

[0. 0. 0. 0. 1. 0. 0. 1. 1. 1. 1. 0. 1. 1. 0.]

Figure 4: Chromosomes of the cluster

Reference solution:

[0. 0. 0. 0. 1. 0. 0. 1. 1. 1. 1. 0. 1. 1. 0.]

Starting population:

[[0. 0. 0. 0. 0. 1. 1. 0. 0. 1. 1. 1. 1. 1. 1.]

[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]

[0. 0. 0. 0. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0.]

[1. 1. 0. 1. 0. 1. 0. 0. 0. 1. 1. 0. 0. 0. 0.]

[1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]

[1. 0. 0. 1. 1. 1. 1. 0. 1. 1. 1. 0. 1. 0. 1.]

[1. 0. 1. 1. 0. 0. 1. 0. 1. 0. 0. 0. 1. 0. 0.]

[0. 1. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]

[0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0.]

[0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0.]

[0. 1. 1. 1. 1. 1. 0. 1. 1. 0. 0. 1. 1. 1. 1.]

[1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 0. 0. 1. 1. 1.]]

AFTER APPLYING FITNESS FUNCTION

Fitness Scores: [ 8 8 11 6 7 8 6 10 9 8 7 7]

RESULT (Contd…)

RESULT (Contd…)

Parent chromosomes 1: [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.]

Parent chromosomes 2: [0. 0. 1. 0. 0. 0. 1. 0. 1. 1. 0. 0. 0. 0. 1.]

Child chromosome 1: [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1.]

Child chromosome 2: [0. 0. 1. 0. 0. 0. 1. 0. 1. 1. 1. 1. 1. 1. 1.]

​

Population before mutation:

[[0. 1. 0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 0.]

[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1.1.]]

RESULT (Contd…)

Population after mutation:

[[0. 1. 1. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0. 0. 1.]

[0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. .]]

​

Result after 200 generations:

Starting best score, percent target: 86.7

End best score, percent target: 100.0

Figure 5: Comparison of GABFCov-19 with Existing Algorithm

CONCLUSION

• A new algorithm named GABFCov-19 algorithm where the inputs are taken from the clusters generated by proximity app.

​

• Match the symptoms of a Covid-19 positive patient with the cluster

​

• Forecast the risk associated with the individuals

CONCLUSION (Contd…)

• Results have been tested with 15 parameters

​

• The process was repeated iteratively and we get the accuracy of predictive positive cases to be 86.7%

FUTURE SCOPE OF THE STUDY

• Algorithm has good accuracy

​

• High testing time is a disadvantage

​

• Using machine learning and artificial intelligence

REFERENCES

• Rizk-Allah, R.M. And Hassanien, A.E., 2020. COVID-19 forecasting based on an improved interior search algorithm and multi-layer feed forward neural network. arXiv preprint arXiv:2004.05960.

​

• Ghazaly, N.M., Abdel-Fattah, M.A. And Abd El-Aziz, A.A., Novel Coronavirus Forecasting Model using Nonlinear Autoregressive Artificial Neural Network.

​

• Al-qaness, M.A., Ewees, A.A., Fan, H. And Abd El Aziz, M., 2020. Optimization method for forecasting confirmed cases of covid-19 in china. Journal of Clinical Medicine, 9(3), p.674.

​

THANK YOU