Drawing insights from COVID-19-infected patients using CT scan images and machine learning techniques: a study on 200 patients

Abstract

As the whole world is witnessing what novel coronavirus (COVID-19) can do to the mankind, it presents several unique features also. In the absence of specific vaccine for COVID-19, it is essential to detect the disease at an early stage and isolate an infected patient. Till today there is a global shortage of testing labs and testing kits for COVID-19. This paper discusses about the role of machine learning techniques for getting important insights like whether lung computed tomography (CT) scan should be the first screening/alternative test for real-time reverse transcriptase-polymerase chain reaction (RT-PCR), is COVID-19 pneumonia different from other viral pneumonia and if yes how to distinguish it using lung CT scan images from the carefully selected data of lung CT scan COVID-19-infected patients from the hospitals of Italy, China, Moscow and India? For training and testing the proposed system, custom vision software of Microsoft azure based on machine learning techniques is used. An overall accuracy of almost 91% is achieved for COVID-19 classification using the proposed methodology.

RETRACTED ARTICLE: GraphCovidNet: A graph neural network based model for detecting COVID-19 from CT scans and X-rays of chest

“…Pedro et al 15 have utilized the EfficientNet 16 model along with transfer learning citetranferlearning and have achieved accuracies 87.60% and 98.99% for COVID-CT dataset 11 and SARS-CoV-2 CT-scan dataset 8 respectively. Sharma et al 17 have applied ResNet 14 on the database consisting of datasets: (i) GitHub COVID-CT dataset 11 , (ii) COVID dataset provided by Italian Society of Medical and Interventional Radiology 18 , (iii) dataset provided by hospitals of Moscow, Russia 19 , (iv) dataset provided by SAL Hospital, Ahmedabad, India 20 and have obtained almost 91% accuracy.…”

The effect of anticancer treatment on cancer patients with COVID‐19: A systematic review and meta‐analysis

“…The diagnosis of COVID-19 should be based on RT-PCR or antibody tests. Due to the changing standard for diagnosis, the shortage of testing kits in some regions, and the unsatisfactory accuracy of laboratory tests, 13,14 a CT finding or a consensus based on symptoms by ≥2 skillful physicians was also acceptable. No restriction was cast on cancer types, but cancer needed to be concurrent with COVID-19, and a cancer history was obviously unacceptable.…”

A machine learning-based framework for diagnosis of COVID-19 from chest X-ray images

“…They used 1262 CT images of COVID-19-positive cases and 1230 CT images of normal patients to secure an accuracy of 96.25% while F1 score of 96.29% with addition of data augmentation technique. Similarly, Sharma [ 40 ] designed a customized CNN-based ResNet50 architect as an automated tool for coronavirus disease diagnosis using 2200 computed tomography images of lungs (800 belongs to COVID-19 positive while 1400 of other pneumonia and normal patients) that attained 91.0% accuracy with sensitivity of 92.1%.…”