Deep Transfer Learning Based Unified Framework for COVID19 Classification and Infection Detection from Chest X-Ray Images

Abstract

The presentation of the COVID19 has endangered several million lives worldwide causing thousands of deaths every day. Evolution of COVID19 as a pandemic calls for automated solutions for initial screening and treatment management. In addition to the thermal scanning mechanisms, findings from chest X-ray imaging examinations are reliable predictors in COVID19 detection, long-term monitoring and severity evaluation. This paper presents a novel deep transfer learning based framework for COVID19 detection and segmentation of infections from chest X-ray images. It is realized as a two-stage cascaded framework with classifier and segmentation subnetwork models. The classifier is modeled as a fine-tuned residual SqueezeNet network, and the segmentation network is implemented as a fine-tuned SegNet semantic segmentation network. The segmentation task is enhanced with a bioinspired Gaussian Mixture Model-based super pixel segmentation. This framework is trained and tested with two public datasets for binary and multiclass classifications and infection segmentation. It achieves accuracies of 99.69% and 99.48% for binary and three class classifications, and a mean accuracy of 83.437% for segmentation. Experimental results and comparative evaluations demonstrate the superiority of this unified model and signify potential extensions for biomarker definition and severity quantization.

Disease Localization and Severity Assessment in Chest X-Ray Images using Multi-Stage Superpixels Classification

“…However, the author has not performed severity analysis and the applied superpixel segmentation utilized constant vertical and horizontal window size (that affect the discriminating potential of extracted features). Another similar work [50] presented transfer learning based COVID-19 detection and infection segmentation model using CXR images. The method achieved 99.69% and 99.48% classification accuracy for binary and three class classification; whereas 83.43% accuracy for segmentation.…”

Interactive framework for Covid-19 detection and segmentation with feedback facility for dynamically improved accuracy and trust

“…Although most research on Covid-19 segmentation applied DNNs for lung segmentation as a part of pre-processing for Covid-19 classification to extract only the lung regions from the medical images [ 27 , 28 ], more recent researches are focusing on segmenting the infections and lesions from lungs using DNN and hybrid models. Sundaram et al [ 29 ] recently proposed a two stage DL framework by combining residual SqueezeNet and SegNet (RSqz-SegNet) models to detect Covid images and segment opacities, granulomas and subtle infections of Covid patients. The model detected Covid patients from viral pneumonia and bacterial pneumonia patients with more than 99% accuracy.…”

A Systematic Review on Machine Learning and Deep Learning Based Predictive Models for Health Informatics

“…Vaiyapuri et al [27], presented an IoT enabled elderly fall detection model using optimal deep convolutional neural network (IMEFD-ODCNN) for smart homecare. Sundaram et al [29], presented a novel deep transfer learning based framework for COVID19 detection and segmentation of infections from chest X-ray images. It was realized as a twostage cascaded framework with classifier and segmentation subnetwork models.…”