Tumbe Group of International Journals

Full Text

Covid 19 Severity Detection with CT Scan Images using Machine Learning

Namratha R1,  Anitha Devi M D2,  M Z Kurian3


1PG Student, Dept. of ECE.,

 Sri Siddhartha Institute of Technology, Tumakuru, Karnataka, India,


2Assistant Professor, Dept. of ECE.,

Sri Siddhartha Institute of Technology,

Tumakuru Karnataka, India


3Head of the Department, Dept. of ECE.,

Sri Siddhartha Institute of Technology,

Tumakuru Karnataka, India,



In December 2019, the first covid case was reported in China province. It has achieved a status as pandemic. This pandemic with continuously evolving transmission. Tracing the people who was a close contact with positive people and then quarantining them by some measures like seal-down. So, some other reliable measures have to be taken in order to control the cases of positivity.

The speed and pace and of the*transmission of severe acute respiratory syndrome coronavirus 2 also referred as Covid-19*have resulted in a*global pandemic, with significant health, financial and other implications.

The global outbreak of novel*coronavirus 2019 was declared by the World Health Organization on 30 January 2020. The*clinical symptoms of covid-19 are predominantly pulmonary, although serious cardiovascular side*effects were also*observed in a number of patients.

Existing preventative solutions, include frequent hand wash using*soap and water, hydro-alcoholic solution and digital technologies to detect and limit the spread*of the virus and track the movement of quarantined*peoples. Impact- millions of deaths, lockdown in cities, restricted movements, business losses, global*economy slowdown.

Keywords: Covid-19, Severity Detection, Machine Learning, global*economy.

  1. Introduction

The coronavirus disease 2019 (Covid-19), caused by severe acute respiratory syndrome coronavirus2 (SARS-CoV-2), is an ongoing larger epidemic. The number of citizens infected by the virus is increasing swiftly. The genuine head diagnosis Reverse transcriptase polymerase chain reaction test. The solution time and cost of this test are excessive, so further swift and reachable diagnostic instruments are needed.

The rapid increase in covid infection peoples is enormous. The healthcare system across world-wide with having only limited testing kits, so it is impossible for every covid patient with respiratory illness to test using normal techniques.

Those tests are also have long turn over time and very limited sensitivity. So, X-ray machines are already in use it may help to quarantine high risk covid affected patients stint test results are awaited.

The sequence of image-based diagnosis for COVID-19, taking thoracic CT as an example. Each subject is instructed and assisted by a technician to pose on the patient bed then CT scan images are acquired during a single breath-hold. The scans are done from the level of upper thoracic inlet to the inferior level of the costophrenic angle with the optimized parameters set by the radiologists, based on the patient’s body shape. From the acquired data, CT images are reconstructed and then transmitted through picture archiving and communication systems (PACS) for subsequent reading and diagnosis.

Artificial intelligence (AI), an emerging technology in the field of medical imaging, has contributed actively to fight COVID-19. Compared to the traditional imaging workflow that heavily relies on human labors, AI enables more safe, accurate and efficient imaging solutions. AI empowered applications in COVID-19Zmainly include the dedicated imaging platform, the lung infection region segmentation, the clinical assessment and diagnosis, as well as the basic and clinical research. Moreover, many commercial products have been developed, which successfully integrate AI to combat COVID-19 and clearly demonstrate the capability of the technology.

Due to the importance of AI in all the spectrum of the imaging-based analysis of COVID-19, this review aims to extensively discuss the role of medical imaging, especially empowered by AI in fighting the COVID-19. In this we first introduce intelligent imaging platforms for COVID-19 and then summarize popular machine learning methods in the imaging workflow including segmentation, diagnosis and prognosis. Several publicly available datasets are also introduced.

Healthcare practitioners are particularly vulnerable concerning the high risk of occupational viral exposure. Imaging specialists and technicians are of high priority, such that any potential contact with the virus could be under control. In addition, to the personal protective equipment (PPE), one may consider dedicated imaging facilities and workflows, which are significantly important to reduce the risks and save lives.



Mohamed Amine Ferrag, Lie Shu, Kim-Kwang Raymond Choo proposed a paper on FIGHTING COVID-19 AND FUTURE PANDEMICS WITH THE INTERNET OF THINGS. This paper gives a method to detect Covid-19 disease by using official contact tracing apps for android phones for example India (Arogya-setu) etc., Some medical organizations use other approaches like Covid-19 symptom tracker app, contact tracing apps, telemedicine kind of apps. It approaches a method of Internet of things to detect Covid-19.

The IoT is a system of interrelated and internet connected objects that are able to collect and transfer the data over a wireless network without human intervention. The IoT is an ideal potential network for monitoring the vaccine status, healthcare, remote infected citizen monitoring and also detecting and preventing the disease. It contains a healthcare sensor where it consists IoT devices and it includes smart hospitals, infected citizens with wearable devices and also medical people. It also consists of a fog computing layer it contains gateways, switches and routers. Also, it has cloud computing layer, in that it contains cloud servers for storage and end to end purposes.

Here the paper by J. P. Kanne, CHEST CT FINDINGS IN 2019 NOVEL CORONAVIRUS INFECTIONS. Here the patients even without any clinical symptoms were also hospitalized or quarantined for further tests and even after tested for multiple times. The RTPCR test is the leader diagnosis method in which throat swabs were collected and tested to give the result if the person is infected or not and the time taken for the result is too long so spreading of infection is also high, in order to overcome this situation CT images, play a vital role in fighting against the pandemic.

CT scans are fast and accessible diagnosis tools and it gives actual rate of infection. By taking CT scan images and those images were reconstructed and transmitted for diagnosis to identify the actual problem.

I.D. Apostol Poulos and T. Bessiana, proposed model on COVID-19: AUTOMATIC DETECTION FROM X-RAY IMAGES UTILIZING TRANSFER LEARNING WITH CONVOLUTIONAL NEURAL NETWORKS. This paper insisted to employ a contactless workflow to avoid further risks of infection. Here chest X-ray are taken to identify the disease but some air space opacities will occur.

Hence, by using machine learning models like CNN mainly used for image analysis tasks like object detection, image recognition and segmentation for diagnosis purpose, the images are reconstructed and transmitted and compares the collected samples with datasets.

WORLD HEALTH ORGANIZATION CORONAVIRUS DISEASE (COVID-2019) SITUATION REPORTS. This represents that the first covid case was reported in China province. It has achieved a status as pandemic. The major symptoms are fever, headache, cold, muscle pain and cough. Some precautionary measures like frequently washing off hands with soaps, using alkaline based hand sanitizers, wearing masks and maintaining social distance were followed. The parameters that are taken into considerations before going to tests are temperature and saturation level of persons.

Then the citizens can undergo various types of diagnosis like RTPCR, Chest X-ray findings and Computed tomography scans. The result of reverse transcriptase polymerase chain reaction test time and cost are high so the fast and accessible diagnosis tool is computed tomography, it gives actual percentage of infection in a person body. So, further spread of infection can be controlled because of fast result and also the mortality rate can be reduced.


Sewio’s real time location system is based on UWB and it consist of hardware and software that together form an all-in-one platform for serving multiple indoor tracking cases that are ranging from asset tracking and material flow to employee location tracking for safety reasons. Advances in RTLS are precise, reliable monitoring and tracking of people, assets and heavy equipment are driving a new wave of transformation. The information is processed in real-time using RTLS studio and positional data is stored in a MySQL database. Test are done, infected people are quarantined, close contacts were also quarantined and timely monitoring those infected people, precautionary measures are taken and also telemedicine kind of treatment is done. Anchors are placed throughout the building and personal tags, each assigned to specific person, are tracked and recorded in real time and can also be visualized and further processed.

J.H. Lee, D. Kim, and M.K. Cho, COMPUTED TOMOGRAPHY APPARATUS AND METHOD OF CONTROLLING X-RAY BY USING THE SAME. This paper represents that the CT scans can be used to identify the disease or injury within various regions of the body. The scan detects hazy, patchy white spots in lung area, that tells a sign of Covid-19. A computerized tomography scan combines a series of X-ray images that are taken in various angles around the body and it uses the computer processing to create the image slices of the soft tissues and lungs in our body. For example, CT scanning is a vital screening tool for detecting possible tumours within the body. A CT scan of the heart may be ordered when various types of heart disease or abnormalities are suspected. CT scan can also be used to image the head in order to locate injuries, tumours, clots leading to stroke and other conditions. It can image the lungs in order to reveal the presence of tumours, pulmonary embolisms (blood clots), excess fluid, and other conditions such as pneumonia. A CT scan is particularly useful when imaging complex bone fractures, severely eroded joints or bone tumours, since it usually produces more detail than would be possible with a conventional x-ray. CT scanning is a vital method to diagnosis Covid-19. It is more time saving and it has a comparable sensitivity against RTPCR.

P. Forthmann and G. Pfleiderer AUGMENTED DISPLAY DEVICE FOR USE IN A MEDICAL IMAGING LABORATORY. This paper gives an information about augmented reality. It starts with a smartphone having camera loaded with a AR software. When aa user points and look at an object the software recognises through computer vision technology that will analyses the video stream. An augmented reality device used in a medical imaging laboratory housing a medical imaging device includes a headset and the cameras are mounted in the medical imaging laboratory, and the directional sensors are mounted on the headset. The cameras generate a panorama image, nothing but it uses some special equipment or a software that will capture the images horizontally with elongated fields of view. Data collected by the directional sensors is processed to determine the viewing direction. The panorama image and the determined viewing direction are processed to generate an augmented patient view image in which the medical imaging device is removed, replaced or they made partially transparent, the augmented image is presented on a display of the headset. The directional sensors may include a headset camera that provides a patient view image, which is augmented by removing or making partially transparent any portion of the medical imaging device in the patient view image by substituting corresponding portions of the panorama image.

Gunther correia bacillar, Mallikarjuna chandrappa, Rajlakshman Kulkarni, Soumava dey COVID-19 CHEST X-RAY IMAGE CLASSIFICATION USING DEEP LEARNING. This paper gives an idea about segregation of images, pre-training, image processing, optimization of model, performance evaluation and accuracy check. Here the segregation part contains the division of images into various categories that corresponds to a different-objects. It extracts the objects that we want for next processing and also it is used to isolate the dedicated object from the image. The pre training step involves to train the model with a task that can be used for some other tasks. And this is a model which is created by someone to solve a similar problem. Here it contains a VGG classifier which is one of the most popular pre-trained models for image classification purpose. The image pre-processing stage involves taking the steps to format the images that are used by training model. Optimization of model refers to the action of making effective use of a resource and it involves maximizing or minimizing some functions that are related to some set. The performance evaluation involves in measuring how precise the predictions are and we use it to measure total predictions including positives and also negatives.

Pillalamarry Mahesh, Yakkala Gnana Prathyusha, Botlagunta Sahiti, S Nagendra COVID-19 DETECTION FROM CHEST X-RAY USING CNN. This paper gives a clear idea about the CNN. The convolution neural network is mainly used in image analysis tasks such as image recognition, object detection and image segmentation. The CNN’s extracts the image features without making any loss in its attribute. Image representation refers to the way that brings information like colour is coded digitally. Edge detection refers for detecting the object boundaries within the image and it allows the user to observe the features of an image and it also involves of computing an image gradient to quantify the magnitude and direction of the edges in an image.


Assessment of infectiousness of the COVID-19Zcases during their early symptomatic phase is critical in designing preventative strategies for pandemic control. Household contacts of COVID-19 cases are most vulnerable population and needs special attention especially when all the countries have imposed lockdowns and are advocating “stay home”. Asymptomatic cases have lesser chances of spreading the disease [3]. However, immediate isolating of the cases upon development of the symptoms reduces the risk of disease drastically. So, enough measures should be taken to limit the contact.


  1. Mohamed Amine Ferrag, Lie Shu, senior member, IEEE and Kim-Kwang Raymond Choo, senior member, IEEE vol 8, no 9, September 2021 Fighting Covid-19 and Future Pandemics with the Internet of Things.
  2. J. P. Kanne, "Chest CT findings in 2019 novel coronavirus (2019-nCoV) infections from Wuhan, China: key points for the radiologist," Radiology, p. 200241, 2020.
  3. I. D. Apostol Poulos and T. Bessiana, "Covid-19: Automatic detection from X-Ray images utilizing transfer Learning with convolutional neural networks," arXiv:2003.11617, 2020.
  4. World Health Organization Coronavirus disease (COVID-2019) situation reports website”, available online [10.4.2020] at https://www.who.int/emergencies/diseases/novel-coronavirus- 2019/situation-reports
  5. “SEWIO UWB Real-Time Location System (RTLS)”, available online [5.4.2020] at https://www.sewio.net/real-time-location-system-rtls-onuwb/
  6. C. Sagonas, E. Antonakos, G, Tzimiropoulos, S. Zafeiriou, M. Pantic. 300 faces In-the-wild challenge: Database and results. Image and Vision Computing (IMAVIS), Special Issue on Facial Landmark Localisation "In-The-Wild". 2016.
  7. J.H. Lee, D. Kim, and M.K. Cho, Computed Tomography Apparatus and Method of Controlling X-ray by Using the Same, August 2021
  8. P. Forthmann and G. Pfleiderer Augmented Display Device for Use in a Medical Imaging Laboratory 25 march 2021
  9. Gunther correia bacillar, Mallikarjuna chandrappa, Rajlakshman Kulkarni, Soumava dey COVID-19 Chest X-ray Image Classification Using Deep Learning, July 19, 2021
  10. Pillalamarry Mahesh, Yakkala Gnana Prathyusha, Botlagunta Sahiti, S Nagendra COVID-19 Detection from Chest X-ray Using CNN, journal of physics conference series, ICMAICT 2020
  11. Yogeesh N. "Mathematical Maxima Program to Show Corona (COVID-19) Disease Spread Over a Period." TUMBE Group of International Journals, vol. 3, no. 1, 2020, pp. 14-16.

Sign In  /  Register

Most Downloaded Articles

Acquire employability in Indian Sinario

Department of Mathematics @ GFGC Tumkur

The Pink Sonnet

ಗ್ರಾಮೀಣ ಪ್ರದೇಶದಲ್ಲಿ ಆಯಗಾರಿಕೆ ಸಂಸ್ಕೃತಿ

Knowledge and Education- At Conjecture

© 2018. Tumbe International Journals . All Rights Reserved. Website Designed by ubiJournal