Auto-diagnosis of COVID-19 using Lung CT Images with Semi-supervised Shallow Learning Network

Abstract

Infection of Novel Coronavirus 2019 (COVID-19) on lung cells and human respiratory systems have raised real concern to the human lives during the current pandemic spread across the world. Recent observations on CT images of human lungs infected by COVID-19 is a challenging task for the researchers in finding suitable image patterns for automatic diagnosis. In this paper, a novel semi-supervised shallow learning network model comprising Parallel Quantum-Inspired Self-supervised Network (PQIS-Net) with Fully Connected (FC) layers is proposed for automatic segmentation followed by patch-based classifications on segmented lung CT images for the diagnosis of COVID-19 disease. The PQIS-Net model is incorporated for fully automated segmentation of lung CT scan images obviating pre-trained convolutional neural network models for feature learning. The PQIS-Net model comprises a trinity of layered structures of quantum bits inter-connected through rotation gates using an 8-connected second-order neighborhood topology for the segmentation of wide variation of local intensities of the CT images. Intensive experiments have been carried out on two publicly available lung CT image data sets thereby achieving promising segmentation outcome and diagnosis efficiency (F1-score and AUC) while compared with the state of the art pre-trained convolutional based models.

Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network

“… Ref. Architecture Accuracy Recall Specificity Precision F1-Score [43] DarkCovidNet 98.08% 95.13% 95.3% 98.03% 96.51% [44] VGG19 96.33% 97.05% 96.0% 91.6% 94.24% [45] VGG19 98.75% 92.85% 98.75% – – [45] MobileNet v2 97.40% 99.10% 97.09% – – [46] nCOVnet 88.10% 82.00% 97.06% 97.62% 89.13% [47] CapsNet 97.24% 97.42% 97.04% 97.08% 97.24% [49] RestNet 18 99.4% 100% 98.6% 99.00% 99.5% [50] Semi-supervised model 93.1% 83.5% – …”

“…Konar et al. [50] propose a semi-supervised shallow neural network model for automated diagnostic of COVID-19. This study included two datasets.…”

“…70% for training and 30% for testing. [50] Proposed Semi-supervised model Chest X-ray Dataset 1: 2482 — total; 1230 — Normal; 1252 — COVID-19 Dataset 2: 20 — COVID-19 Binary Random sampling 70% for training and 30% for testing …”

A Transfer Learning Model for COVID-19 Detection with Computed Tomography and Sonogram Images

“…And an another observation stated that, to achieve the automatic detection of calcification, the dilated convolutional network is been recommended highly [7]. To attain a promising segmentation outcome, Debanjan Konar [4] worked on novel framework of semi-supervised network which follows shallow learning on Brazilian City Hospital Dataset with both left-right lung masks and lung infection pair masks. Zhou L [14] worked on resolving heterogeneity issue in Harbin dataset with segmentation at voxel level and Riyadh Dataset at ROI level for validation.…”

An integrated feature frame work for automated segmentation of COVID‐19 infection from lung CT images

“…A novel semi‐supervised shallow learning model including Parallel Quantum‐Inspired Self‐supervised Network (PQIS‐Net) with Fully‐Connected (FC) layers for automatic segmentation of COVID‐19 CT image is proposed in Konar et al 40 The patch‐based classification was applied to the segmented images for the diagnosis of COVID‐19 using the two publicly available datasets. The efficiency (F1‐score and AUC) of the PQIS‐Net was compared with pre‐trained convolutional based models.…”