Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

OSegNet: Operational Segmentation Network for COVID-19 Detection using Chest X-ray Images

Aysen Degerli, Serkan Kiranyaz, Muhammad E. H. Chowdhury, Moncef Gabbouj

TL;DR

The study addresses the challenge of reliable COVID-19 detection from chest X-rays amid limited data by proposing OSegNet, an autoencoder-like network that segments COVID-19 pneumonia using a decoder built from Self-Organized Operational Neural Networks. It leverages an extended QaTa-COV19 dataset with ground-truth masks to train and evaluate segmentation and detection, employing a hybrid dice/focal loss and transfer-learning encoders (e.g., Inception-v3, DenseNet-121). Key findings show that OSegNet achieves state-of-the-art-like performance, with detection accuracy up to 99.65% and segmentation F2-scores up to 88.75% (best at Q=3), while reducing model depth and parameters relative to strong baselines. This work provides a large public benchmark and a segmentation-driven approach that improves reliability and interpretability for clinical COVID-19 diagnosis from CXRs.

Abstract

Coronavirus disease 2019 (COVID-19) has been diagnosed automatically using Machine Learning algorithms over chest X-ray (CXR) images. However, most of the earlier studies used Deep Learning models over scarce datasets bearing the risk of overfitting. Additionally, previous studies have revealed the fact that deep networks are not reliable for classification since their decisions may originate from irrelevant areas on the CXRs. Therefore, in this study, we propose Operational Segmentation Network (OSegNet) that performs detection by segmenting COVID-19 pneumonia for a reliable diagnosis. To address the data scarcity encountered in training and especially in evaluation, this study extends the largest COVID-19 CXR dataset: QaTa-COV19 with 121,378 CXRs including 9258 COVID-19 samples with their corresponding ground-truth segmentation masks that are publicly shared with the research community. Consequently, OSegNet has achieved a detection performance with the highest accuracy of 99.65% among the state-of-the-art deep models with 98.09% precision.

OSegNet: Operational Segmentation Network for COVID-19 Detection using Chest X-ray Images

TL;DR

The study addresses the challenge of reliable COVID-19 detection from chest X-rays amid limited data by proposing OSegNet, an autoencoder-like network that segments COVID-19 pneumonia using a decoder built from Self-Organized Operational Neural Networks. It leverages an extended QaTa-COV19 dataset with ground-truth masks to train and evaluate segmentation and detection, employing a hybrid dice/focal loss and transfer-learning encoders (e.g., Inception-v3, DenseNet-121). Key findings show that OSegNet achieves state-of-the-art-like performance, with detection accuracy up to 99.65% and segmentation F2-scores up to 88.75% (best at Q=3), while reducing model depth and parameters relative to strong baselines. This work provides a large public benchmark and a segmentation-driven approach that improves reliability and interpretability for clinical COVID-19 diagnosis from CXRs.

Abstract

Coronavirus disease 2019 (COVID-19) has been diagnosed automatically using Machine Learning algorithms over chest X-ray (CXR) images. However, most of the earlier studies used Deep Learning models over scarce datasets bearing the risk of overfitting. Additionally, previous studies have revealed the fact that deep networks are not reliable for classification since their decisions may originate from irrelevant areas on the CXRs. Therefore, in this study, we propose Operational Segmentation Network (OSegNet) that performs detection by segmenting COVID-19 pneumonia for a reliable diagnosis. To address the data scarcity encountered in training and especially in evaluation, this study extends the largest COVID-19 CXR dataset: QaTa-COV19 with 121,378 CXRs including 9258 COVID-19 samples with their corresponding ground-truth segmentation masks that are publicly shared with the research community. Consequently, OSegNet has achieved a detection performance with the highest accuracy of 99.65% among the state-of-the-art deep models with 98.09% precision.
Paper Structure (8 sections, 7 equations, 1 figure, 5 tables)

This paper contains 8 sections, 7 equations, 1 figure, 5 tables.

Table of Contents

  1. Introduction
  2. Methodology and Materials
  3. OSegNet: Operational Segmentation Network
  4. QaTa-COV19 Dataset
  5. Experimental Evaluation
  6. Experimental Setup
  7. Experimental Results
  8. Conclusions

Figures (1)

  • Figure 1: The proposed OSegNet model for COVID-19 pneumonia segmentation is illustrated, where the transfer learning is performed at the encoder block, and operational layers (Oper2D) are used at the decoder block.