Indian Journal of Radio & Space Physics

Open Access Open Access  Restricted Access Subscription Access

Accurate prognosis of Covid-19 using CT scan images with deep learning model and machine learning classifiers


Affiliations
1 Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun 248 001, India
2 Department of Computer Science and Engineering, Graphic Era Hill University, Dehradun 248 001, India
 

The Covid-19 disease is caused by coronavirus or SARS-CoV-2 has wrecked havoc globally. This epidemic severely impacted the economy of most of the countries across the world and has taken away many lives. To control the pandemic situation many researchers, organizations, and institutes have come up with the pathogenesis and developing vaccines to decimate this disease. Out of the several techniques, one of the techniques use image patterns on Computed Tomography (CT) to detect whether a patient is Covid-19 positive or not. In this work, the SARS-COV-2 dataset has been used for the detection of Covid-19 images and normal images. These dataset images have been fed to various deep learning models for extracting the features and finally passed to various ML classifiers which classify the images as Covid-19 or normal images. The results have established that the VGG19 model along with Logistic Regression (LR) classifier gives the maximum AUC and accuracy of 98.5% and 94.6%.
User
Notifications
Font Size

Abstract Views: 116

PDF Views: 62




  • Accurate prognosis of Covid-19 using CT scan images with deep learning model and machine learning classifiers

Abstract Views: 116  |  PDF Views: 62

Authors

Siddharth Gupta
Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun 248 001, India
Palak Aggarwal
Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun 248 001, India
Nisha Chaubey
Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun 248 001, India
Avnish Panwar
Department of Computer Science and Engineering, Graphic Era Hill University, Dehradun 248 001, India

Abstract


The Covid-19 disease is caused by coronavirus or SARS-CoV-2 has wrecked havoc globally. This epidemic severely impacted the economy of most of the countries across the world and has taken away many lives. To control the pandemic situation many researchers, organizations, and institutes have come up with the pathogenesis and developing vaccines to decimate this disease. Out of the several techniques, one of the techniques use image patterns on Computed Tomography (CT) to detect whether a patient is Covid-19 positive or not. In this work, the SARS-COV-2 dataset has been used for the detection of Covid-19 images and normal images. These dataset images have been fed to various deep learning models for extracting the features and finally passed to various ML classifiers which classify the images as Covid-19 or normal images. The results have established that the VGG19 model along with Logistic Regression (LR) classifier gives the maximum AUC and accuracy of 98.5% and 94.6%.