Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Automatic Detection of COVID-19 from Chest X-ray Images Using Deep Learning Model

Alloy Das, Rohit Agarwal, Rituparna Singh, Arindam Chowdhury, Debashis Nandi

TL;DR

This work evaluates U-Net and W-Net architectures for automatic COVID-19 detection from chest X-ray images, addressing the shortage of RT-PCR tests by providing a secondary screening tool. The authors build a 400×400 grayscale pipeline with semantic segmentation guiding classification, demonstrating strong performance in both binary (COVID vs Normal) and ternary (COVID vs Normal vs Pneumonia) tasks on public datasets. Across 80:20 train-test splits, U-Net and W-Net achieve accuracies up to $0.9917$ with near-perfect sensitivity and high specificity, complemented by ROC analyses and heat-map visualizations that validate focus on clinically relevant lung regions. The results suggest these models can assist radiologists in rapid screening, with future work aimed at multi-class expansion and broader clinical validation.

Abstract

The infectious disease caused by novel corona virus (2019-nCoV) has been widely spreading since last year and has shaken the entire world. It has caused an unprecedented effect on daily life, global economy and public health. Hence this disease detection has life-saving importance for both patients as well as doctors. Due to limited test kits, it is also a daunting task to test every patient with severe respiratory problems using conventional techniques (RT-PCR). Thus implementing an automatic diagnosis system is urgently required to overcome the scarcity problem of Covid-19 test kits at hospital, health care systems. The diagnostic approach is mainly classified into two categories-laboratory based and Chest radiography approach. In this paper, a novel approach for computerized corona virus (2019-nCoV) detection from lung x-ray images is presented. Here, we propose models using deep learning to show the effectiveness of diagnostic systems. In the experimental result, we evaluate proposed models on publicly available data-set which exhibit satisfactory performance and promising results compared with other previous existing methods.

Automatic Detection of COVID-19 from Chest X-ray Images Using Deep Learning Model

TL;DR

This work evaluates U-Net and W-Net architectures for automatic COVID-19 detection from chest X-ray images, addressing the shortage of RT-PCR tests by providing a secondary screening tool. The authors build a 400×400 grayscale pipeline with semantic segmentation guiding classification, demonstrating strong performance in both binary (COVID vs Normal) and ternary (COVID vs Normal vs Pneumonia) tasks on public datasets. Across 80:20 train-test splits, U-Net and W-Net achieve accuracies up to with near-perfect sensitivity and high specificity, complemented by ROC analyses and heat-map visualizations that validate focus on clinically relevant lung regions. The results suggest these models can assist radiologists in rapid screening, with future work aimed at multi-class expansion and broader clinical validation.

Abstract

The infectious disease caused by novel corona virus (2019-nCoV) has been widely spreading since last year and has shaken the entire world. It has caused an unprecedented effect on daily life, global economy and public health. Hence this disease detection has life-saving importance for both patients as well as doctors. Due to limited test kits, it is also a daunting task to test every patient with severe respiratory problems using conventional techniques (RT-PCR). Thus implementing an automatic diagnosis system is urgently required to overcome the scarcity problem of Covid-19 test kits at hospital, health care systems. The diagnostic approach is mainly classified into two categories-laboratory based and Chest radiography approach. In this paper, a novel approach for computerized corona virus (2019-nCoV) detection from lung x-ray images is presented. Here, we propose models using deep learning to show the effectiveness of diagnostic systems. In the experimental result, we evaluate proposed models on publicly available data-set which exhibit satisfactory performance and promising results compared with other previous existing methods.
Paper Structure (21 sections, 9 figures, 10 tables)

This paper contains 21 sections, 9 figures, 10 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Proposed Methodology
  4. Network Architecture
  5. W-Net
  6. Experimental Result:
  7. Data-Set:
  8. TRAIN-TEST SPLIT:
  9. Two-Class Classification:
  10. Three-Class Cassification:
  11. Result:
  12. Two-Class Clasification:
  13. U-Net:
  14. W-Net
  15. Three-Class Clasification:
  16. ...and 6 more sections

Figures (9)

  • Figure 1: (a–c) 2019-nCoV affected chest X-Ray images (d–f) Pneumonia affected chest X-Ray (g-i) Normal chest X-Ray images.
  • Figure 2: Workflow of the proposed classification system
  • Figure 3: W-NET
  • Figure 4: ROC curve U-NET
  • Figure 5: ROC curve W-NET
  • ...and 4 more figures