View of Prediction Of Covid-19 Using Support Vector Algorithm

Turkish Journal of Computer and Mathematics Education Vol.12 No.12 (2021), 1683-1690

Research Article

1683

Prediction Of Covid-19 Using Support Vector Algorithm

Dr. J. Jayaudhaya1_{, J. Sumithra}2_{, M. Sharmila}3_{, P. Sreenidhi}4_{, P. sruthi}5

1, 2 _{Associate Professor, Department of Electrical and Electronics Engineering, R.M.D. Engineering College,}

Chennai, Tamil Nadu, India

3, 4, 5 _{Student, Department of Electrical and Electronics Engineering, R.M.D. Engineering College,}

Chennai, Tamil Nadu, India

E-mail: [email protected]

Article History: Received: 11 January 2021; Revised: 12 February 2021; Accepted: 27 March 2021; Published online:

23 May 2021

Abstract- Predicting COVID 19 has been a major challenge in the lungs nodule because the cells were overlapped with

each other. The first case was recorded in Wuhan, China since late December 2019, which subsequently became pandemic around the world. Serious illness may result in death as a consequence of substantial alveolar damage and progressive respiratory failure. Reverse transcription polymerase chain reaction (RT-PCR) is widely used for the predictions, the tests may produce false negatives. Under the pandemic situation, shortage of RT-PCR testing resources may also delay the results. Under such circumstances, Lungs CT imaging has become a valuable tool for both diagnosis and prognosis of COVID-19 patients. In this study, we propose supervised deep learning strategy for detecting and classifying COVID-19 infection from CT images. To obtain accurate infection detection and distinguish COVID-19 from non-COVID-19 cases “PREDICTION OF COVID-19 USING SUPPORT VECTOR ALGORITHM” is used .

Keywords: Covid 19, Support Vector Algorithm, Reverse transcription polymerase chain reaction (RT-PCR)

1. Introduction

Corona virus disease (COVID-19) is a respiratory disease caused by a virus. The common symptoms of this disease are fever, dry cough, and tiredness. Other symptoms are aches and pains, nasal congestion, headache, sore throat, loss of taste and smell called Anosmia. COVID- 19 can damage the lungs, causing pneumonia. The virus can exacerbate through the respiratory tract and enter into a person’s lungs. This causes damage to the air sacs or alveoli that can fill with fluid. This progression then constraints a person’s ability to take in oxygen. Continuous oxygen deprivation can damage many of the body’s organs, causing kidney failure, heart attacks, and other life-threatening conditions. People who have pre-existing conditions such as cancer, diabetes, high blood pressure, kidney or liver disease, including but not limited to asthma are at most risk of COVID-19 pneumonia. People over the age of 65 years are more prone to the intense effects of this disease. The disease has turned into a widespread pandemic where the cases and deaths seem to surge rapidly day by day. This research intends to uncover the prognosis of various parameters involved with this virus such as the increase of new cases, recoveries and deaths daily worldwide with the help of a machine learning technique called Prophet model which was developed and introduced by Face book. Existing model has RT-PCR (Reverse Transcription polymerase chain reaction) tests, High amount of false rate, it takes one day to get results and its accuracy is about 40%-45%. Proposed model uses CT-Scans (Computed Tonography), Less amount of false rate, faster response and Its accuracy is about 85%-90%.

2. Block Diagram of Proposed system

Turkish Journal of Computer and Mathematics Education Vol.12 No.12 (2021), 1683-1690

Research Article

1684

Initially, we are loading patient CT images from the Local "C" drive into MATLAB. We can't carry an image directly from the c drive to the Matlab so we are using image acquisition. Following that we are converting the load input (RGB image) to a grey image with the help of the grey image process. Even though, if the image is in grey colour, Matlab will consider that input image as a colour image only. So for further process, we must convert the image to a grey image. In this filtration process, we are going to remove the noise and disturbance in the image with the help of the "Gabor filter" which might occur in the previous stage. To check the quality of the image, we are using image quality assessment here and It is also represented in the graph. To enhance the contrast level and to brighten up the image, the adaptive histogram is used. With the help of this enhancement process, we are getting an improved version of the input image and here also we are using image quality assessment and which is also represented in the form of graph. Feature Extraction is used for the extraction of all the information like GLCM and LBP from the above processed CT image. The Support Vector Machine ( SVM) is used to compare the trained dataset CT images with the above Processed CT image. After this comparison, we are about to conclude that the patient is suspected for Covid-19 or not. The patient gets to know about their results whether they are suspected for covid negative or covid positive through mail.

ALGORITHM

STEP 1: We are importing the input images

STEP 2: We are using linear dataset. It is used to get more information about data such as features, target name etc. STEP 3: From the input image we are extracting the features.

STEP 4: We are using two data sets in SVM, one is trained data set and another one is testing data set. STEP 5: We are using train SVM model process to compare trained and testing data set.

STEP 6: After comparing we are predicting COVID positive or negative. STEP 7: The result is sent through mail.

Turkish Journal of Computer and Mathematics Education Vol.12 No.12 (2021), 1683-1690

Research Article

1685

a. GREY IMAGE CONVERSION

The load input is converted to grey through image acquisition in MATLAB.

Fig. 3 shows gray image conversion

b. FILTRATION

Fig.4 shows filtering output The filtering process is done with the Gabor filter.

1686

c. CONTRAST ENHANCEMENT

Fig.6 shows contrast enhancement

The filtered image is enhanced with the help of adaptive histogram.

GRAPHICAL REPRESENTATION OF ADAPTIVE HISTOGRAM

Fig.7 shows graphical representation of Adaptive Histogram

1687

In both dilation and erosion the inner and outer parts are brightened up.

e. NODULE PART

Fig. 9 shows nodule part output

The nodule part compares the features with trained data. It will highlight the abnormal part.

GRAPHICAL REPRESENTATION OF FEATURE EXTRACTION

Fig. 10 shows graphical representation of feature extraction

1688

1689

5. Conclusion

Without the need for annotating the COVID-19 lessons in CT volumes for training, our weakly-supervised deep learning algorithm obtained strong COVID-19 detection performance. Therefore, our algorithm has great potential to be applied in clinical application for accurate and rapid COVID-19 diagnosis, which is of great help for the frontline medical staff and is also vital to control this epidemic worldwide.

References

[1] D. S. W. Ting, L. Carin, V. Dzau, and T. Y. Wong, “Digital technology and COVID-19,” Nat. Med., vol. 26, no. 4, pp. 459-461, 2020, doi: 10.1038/s41591-020-0824-5.

1690

[2] J. A. Lewnard and N. C. Lo, “Scientific and ethical basis for social- distancing interventions against COVID-19,” Lancet Infect. Dis. , vol. 20, no. 6, pp. 631-633, 2020, doi: 10.1016/S1473-3099(20)30190-0.

[3] S. Woolhandler and D. U. Himmelstein, “Intersecting U.S. Epidemics: COVID-19 and Lack of Health Insurance,” Ann. Intern. Med., vol. 173, no. 1, pp. 63-64, 2020, doi: 10.7326/M20-1491.

[4] E. Christaki, “New technologies in predicting, preventing and controlling emerging infectious diseases,” Virulence, vol. 6, no. 6, pp. 558565, 2015, doi: 10.1080/21505594.2015.1040975.

[5] T. L. Inn, “Smart City Technologies Take on COVID-19,” Penang institude, 2020. Accessed: Aug.2, 2020.[online]. Available: https://pe- nanginstitute.org/publications/issues/smart-city-technologies-take- on-covid-19

[6] L. Setti, F. Passarini, G. De Gennaro, P. Barbieri, M. G. Perrone, M. Borelli, J. Palmisani,

A. Di Gilio, P. Piscitelli, and A. Miani, “Airborne Transmission Route of COVID-19: Why 2 Meters/6 Feet of Inter-Personal Distance Could Not Be Enough,” Int. J. Environ. Res. Public Health, vol. 17, no. 8, pp. 2932-2937, 2020, doi10.3390/ijerph17082932.

[7] R. A. Calvo, S. Deterding, and R. M. Ryan, “Health surveillance dur- ing covid-19 pandemic,” Bmj, vol. 369, 2020, doi:10.1136/bmj.m1373.

[8] T. Sharon, “Blind-sided by privacy? digital contact tracing, the apple/google api and big tech’s newfound role as global health policy makers,” Ethics Inf. Technol., 2020, doi: 10.1007/s10676-020-09547-x.

[9] Y. Yin, Y. Zeng, X. Chen, and Y. Fan, “The internet of things in healthcare: An overview,” J. Ind. Inf. Integr., vol. 1, pp. 3-13, 2016, doi: 10.1016/j.jii.2016.03.004.

[10] C. F. Pasluosta, H. Gassner, J. Winkler, J. Klucken, and B. M. Esko- fier, “An emerging era in the management of Parkinson’s disease: Wearable technologies and the Internet of Things,” IEEE J. Biomed. Health Inform., vol. 19, no. 6, pp. 1873–1881, 2015, doi: 10.1109/JBHI.2015.2461555.

[11] P. A. Laplante and N. Laplante, “The Internet of Things in Healthcare: Potential Applications and Challenges,” IT Prof., vol. 18, no. 3, pp. 2– 4, 2016, doi: 10.1109/MITP.2016.42.