COVID-19 Detection using Transfer Learning with Convolutional Neural Network

TL;DR

This paper proposes a transfer learning approach using Inception V3 CNN to detect COVID-19 from chest CT images, aiming to improve diagnosis accuracy amid limited testing kits.

Contribution

It introduces a CNN model with transfer learning using Inception V3 for COVID-19 detection from CT images, enhancing feature extraction and accuracy.

Findings

Achieved high accuracy in COVID-19 detection from CT images.

Demonstrated the effectiveness of transfer learning in medical image analysis.

Provided a rapid diagnostic tool during the pandemic.

Abstract

The Novel Coronavirus disease 2019 (COVID-19) is a fatal infectious disease, first recognized in December 2019 in Wuhan, Hubei, China, and has gone on an epidemic situation. Under these circumstances, it became more important to detect COVID-19 in infected people. Nowadays, the testing kits are gradually lessening in number compared to the number of infected population. Under recent prevailing conditions, the diagnosis of lung disease by analyzing chest CT (Computed Tomography) images has become an important tool for both diagnosis and prophecy of COVID-19 patients. In this study, a Transfer learning strategy (CNN) for detecting COVID-19 infection from CT images has been proposed. In the proposed model, a multilayer Convolutional neural network (CNN) with Transfer learning model Inception V3 has been designed. Similar to CNN, it uses convolution and pooling to extract features, but this transfer learning model contains weights of dataset Imagenet. Thus it can detect features very effectively which gives it an upper hand for achieving better accuracy.