Bioscience Biotechnology Research Communications

An Open Access International Journal

P-ISSN: 0974-6455 E-ISSN: 2321-4007

Bioscience Biotechnology Research Communications

An Open Access International Journal

V Anantha Natarajan, D. Ganesh, Macha Babitha and M. Sunil Kumar

Department of CSE Sree Vidyanikethan Engineering College, Tirupati, India

Corresponding author email:

Article Publishing History

Received: 19/05/2021

Accepted After Revision: 28/07/2021


At recent times as the COVID 19 pandemic surges in global level, this research article aims at presenting an efficient support system for the physicians in diagnosing the COVID 19 disease using deep learning architectures. The automated diagnosis is made available primarily based on evaluation of medical images (Chest CT images) in diagnosing COVID-19. This COVID 19 affects the normal functioning of the lungs and it damages the tiny air sacs called alveoli. The Chest CT, especially in case of diagnosis of severely infected patients has higher importance and also for immediate COVID 19 screening before certain emergency surgeries and treatment procedures. Till now, the diagnosis of chest CT depends on the visual analysis of radiologists, which may be prone to error at times. First of all, chest CT holds hundreds of slices, which takes a while to diagnose. Next, COVID-19, as a pulmonary disease, has a similar instance with diverse varieties of pneumonia. This research attempts to diagnose the severity of COVID-19 by detecting the abnormalities based on the radiomics features of the chest CT images (pre-processed). These features help to discriminate between the normal opaque region, GGO’s, and high intensity region including blood vessels and other consolidations.  This (classification of chest CT image using radiomic features for COVID 19 diagnosis using neural network) approach can lessen subjective variability and improves diagnostic efficiency when compared to modern-day qualitative evaluation techniques.


COVID 19, Chest CT, Radiomics, GGO’s, Classification, Neural Network

Download this article as:

Copy the following to cite this article:

Copy the following to cite this URL: