An Efficient way of Predicting Covid-19 using Machine and Deep Learning Algorithms

N, Meenakshi and Angappan, Kumaresan and A, Sandhya and V, Naga Susmitha and K, Vaishnavi (2021) An Efficient way of Predicting Covid-19 using Machine and Deep Learning Algorithms. In: I3CAC 2021, 7-8 June 2021, Bharath University, Chennai, India.

[img]
Preview
Text (PDF)
eai.7-6-2021.2308785.pdf - Published Version

Download (770kB) | Preview

Abstract

Recently the novel coronavirus disease pushed the world into the dramatic situation. The tough thing to deal with novel corona virus is the prediction. In the beginning RT PCR test is the golden standard test for the prediction of COVID, which takes more time, more licensed laboratories, trained personnel and prediction accuracy will be not fruitful. In our System, We used current technology for the prediction, which involves: An Efficient Random Forest, a machine learning classification model which predicts whether the person is Corona affected or not using routine blood reports and a deep learning model, Modified DenseNet121 which was pre-trained to predict theCovid using CT scan images. To analyze the machine learning model performance, 5744 blood report samples have been collected from Kagglerepository;similarly, 2482 CT scan samples have been collected from the Kaggle repository, for prediction using Random Forest and DenseNet121 model. The proposed model which is developed using machine and deep learning techniques can be deployed easily and can be used for rapid and accurate prediction of Covid19.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: covid-19 clinical data analysis machine learning deep learning blood test ct scan efficient random forest modified densenet121
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
QA75 Electronic computers. Computer science
Depositing User: EAI Editor IV
Date Deposited: 11 Jun 2021 08:03
Last Modified: 11 Jun 2021 08:03
URI: https://eprints.eudl.eu/id/eprint/3874

Actions (login required)

View Item View Item