key: cord-0940099-xg19nye3
authors: Amin, Javeria; Anjum, Muhammad Almas; Sharif, Muhammad; Rehman, Amjad; Saba, Tanzila; Zahra, Rida
title: Microscopic segmentation and classification of COVID‐19 infection with ensemble convolutional neural network
date: 2021-08-26
journal: Microsc Res Tech
DOI: 10.1002/jemt.23913
sha: d3a142ca5c2a0512c374de205f1335ffec7b5e7d
doc_id: 940099
cord_uid: xg19nye3

The detection of biological RNA from sputum has a comparatively poor positive rate in the initial/early stages of discovering COVID‐19, as per the World Health Organization. It has a different morphological structure as compared to healthy images, manifested by computer tomography (CT). COVID‐19 diagnosis at an early stage can aid in the timely cure of patients, lowering the mortality rate. In this reported research, three‐phase model is proposed for COVID‐19 detection. In Phase I, noise is removed from CT images using a denoise convolutional neural network (DnCNN). In the Phase II, the actual lesion region is segmented from the enhanced CT images by using deeplabv3 and ResNet‐18. In Phase III, segmented images are passed to the stack sparse autoencoder (SSAE) deep learning model having two stack auto‐encoders (SAE) with the selected hidden layers. The designed SSAE model is based on both SAE and softmax layers for COVID19 classification. The proposed method is evaluated on actual patient data of Pakistan Ordinance Factories and other public benchmark data sets with different scanners/mediums. The proposed method achieved global segmentation accuracy of 0.96 and 0.97 for classification.

This research work is mainly focused on the detection of problems. The noisy CT images having poor contrast at the same time, the noise removal is a challenging task (Ejaz et al., 2021; Rehman et al., 2020) . Accurate lesion segmentation is a complex task because of the irregular size and shape of the abnormal region. Classification among the Healthy/COVID19 is another exciting task because it depends upon the feature extraction and selection methods. COVID-19 identification in its early stages is a big issue due to nucleic acidbased laboratory testing limitations. AI-based methods can be utilized as front-line health care for accurate and quickly diagnosed COVID-19 (He et al., 2021) . As a result, the deep learning methodology is introduced for accurate COVID-19 identification. The core contribution steps of the proposed modal are:

1. The proposed modal addresses the adverse effect of noisy images on detection rate DcNN methodology employed on input images for noise removal. The proposed research is presented in five sections; related work is discussed in Section 2, suggested method is mentioned in Section 3 and achieved results & conclusion is written in Section 4 & Section 5, respectively.

COVID-19 spread from one person to another through respiratory droplets (Singhal, 2020) has caused adverse impact on the industries, educational institutes and shopping malls, etc. It is also causing a loss to the industry (Nicola et al., 2020) . The computerized methodologies might provide help for precise COVID-19 detection because manual evaluation of the CT scan takes up to 15 min while DL takes a few seconds and improves the clinical assessments' efficiency (McCall, 2020) . In the literature, extensive work is done on COVID-19 detection using CNN; some of the latest work is discussed in this section (Rodriguez-Morales et al., 2020; Sohrabi et al., 2020) . The CNNs (Amin, Sharif, Raza, Saba, & Anjum, 2019; Amin, Sharif, Raza, Saba, & Rehman, 2019; Amin, Sharif, Yasmin, Saba, & Raza, 2019; Khan, Nazir, et al., 2019; Khan, Javed, Sharif, Saba, & Rehman, 2019; Rehman, Sadad, Saba, Hussain, & Tariq, 2021b) are helpful in extracting the meaningful features from lung CT images for identification of pulmonary nodules, ground-glass haze, and pneumonia (Choe et al., 2019) . Patchy shadows with bilateral distribution are symptoms of COVID-19 which are easily detected using CNN (Jamil & Hussain, 2020; Ramzan et al., 2020) . The 3D DL model is used to extract the local (2D) and global (3D) features (Amin et al., 2020; Fahad, Ghani Khan, Saba, Rehman, & Iqbal, 2018; Liaqat et al., 2018) from lung CT images, method is evaluated on CT scan of the 3,506 patients, which are collected from different six medical centers . The pretrained such as Xception & ResNet101 are utilized for extraction of COVID-19 features (Ardakani, Kanafi, Acharya, Khadem, & Mohammadi, 2020) . The inception model is modified with pretrained weights for classification and results are reported on 453 CT slices (Guo et al., 2020) . Sixteen pretrained CNN models are used for the analysis of the COVID-19 features and the results are reported on the publicly benchmark Chinese data set (Pham, 2020) . The method is assessed on the local CT images as well as publicly available benchmark data sets. The COVID-19 detection is still a challenging task due to the complex structure of the lung CT images. The lesions regions appear in a variable shape and size on the border. Therefore, the actually infected region is also segmented as a healthy region. In this research, a novel approach is proposed for noise removal by utilizing denoise convolutional neural network (DnCNN) and semantic segmentation based on U-Net model. The proposed SSAE model provides significant features from the segmented images for accurate classification.

The methodology employed is depicted in Figure 1 

The CT is the best imaging modality for lung nodule diagnosis. It is used to analyze the functional and structural information related to human body parts . However, the quality of the CT images is degraded due to the radiation, which affects the Radiologists decision and diagnosis. Therefore noise removal is a challenging task for the accurate detection of COVID-19 (Mahersia, Zaroug, & Gabralla, 2015) . That is why pretrained DnCNN (Zhang, Zuo, Chen, Meng, & Zhang, 2017) 3.2 | Semantic segmentation using DeeplabV3 and ResNet-18 model In this study, a semantic segmentation model is proposed utilizing deeplabv3 convolutional neural network as a backbone of the pretrained ResNet18 (He, Zhang, Ren, & Sun, 2016) . The deeplabv3 (Chen, Zhu, Papandreou, Schroff, & Adam, 2018) network utilize encoder/decoder model dilated convolutional and skip connections to segments the multi-scaled objects. Therefore, the reported research combination of the deeplabv3 and Resnet-18 models is employed for segmentation of COVID-19. The designed segmentation model consists of the 98 layers such as 01 input, 30 convolutional, 25 ReLU, 01 maxpooling, 28 batch-normalization, 08 addition, 02 depth concatenation, 01 crop 2D, 01 softmax, and 01-pixel classification. The proposed architecture with activation units of different layers is illustrated in Figure 3 . Table 1 lists the configuration settings for the recommended segmentation technique and segmented lesion regions are depicted in Figure 4 .

Segmented CT images are obtained after applying semantic segmentation model and subsequently supplied to the SSAE model. The proposed SSAE (Olshausen & Field, 1997) model is designed by two sparse autoencoders (SAEs) with a softmax layer as shown in Figure 5 .

A detailed description of autoencoder and SAE is discussed in the succeeding paragraph.

Autoencoder (AE) is an unsupervised network in which training is performed to replicate input size as its output. The training depends upon the cost optimization function. Cost function (CF) computes error among input I(x, y) and reconstructed output image b I x,y ð Þ. The autoencoder input vector I R DI , encoder maps vector I into another vector z R D I ð Þ as

Here, superscript (1) represents the first hidden layer, h 1 ð Þ :

and bias matrix b 1 ð Þ R D 1 ð Þ . The decoder maps z block encoded representation into input vector I that is defined as:

Superscript (2) 

Autoencoder sparsity is probable through the addition of regularizers to CF. The average output value activation of each neuron is computed by function of the regularizers, which is mathematically

where n represent training samples, I j is jth training samples, w i 

where Kullback-Leibler divergence (kL) computes difference among the distributions, which takes zero value when ρ and b ρ i are equal and larger when they diverge each other. Sparsity regularizers apply a sparsity constraint of the hidden layer output. Sparsity, regularization is added which takes higher value when average activations values b ρ i of neuron i and desired value ρ are not close such as Kullback-Leibler divergence sparsity regularization.

where L denotes hidden layers, n represents observations, and k training samples. L2 regularization is used in training of the SAE, in which set small value of the sparsity regularizer by increase the weights values and minimize the z values. The training CF in which SAE adjust the mean squared error (MSE) as follows

where λ denote L2 regularization coefficient. β represents sparsity regularization. Figure 7 .

The stacked network is connected with softmax. The SSAE network is mathematically express as: 

where A i denote ith neuron of the kernel vector, J i denote SSAE output, and φ represent the softmax. images with performance measures are also illustrated in Figure 8 . The empirical results image by image are mentioned in Table 3 , the numerical result shows significant enhancement in PSNR after applying the DnCNN method. The average enhancement results are also computed separately on benchmark data sets as mentioned in Table 4 . 

The proposed model is validated in terms of mean intersection over union (mIoU), global accuracy (GAc), Weighted IoU (wIoU), mean accuracy (mAc), and F1 score on COVID-19 segmentation and

Pakistani Hospitals data sets. In addition, the average segmentation outcomes are mentioned in Table 5 .

The results in Table 5 show that the proposed method achieved 0.97 GAc on the POF Hospital data set, whereas 0.96

GAc on COVID-19 segmentation data set. Figures 9 and 10 demonstrate the segmentation of lung CT photographs using ground truth.

A combination of SAE1 and SAE2 designs SSAE model. The configuration parameters such as hidden units, sparsity proportion, L2weight regularization, and number of epochs of SAE1 & SAE2 are selected after extensive experimentation that provides help to learn the complicated patterns as mentioned in Table 6 . The computation time of the training model is depicted in Table 7 .

The selected configuration parameters with error rates are graphically shown in Figure 11 . POF Hospital data sets than another benchmark data set. The proposed method performance is also plotted in terms of ROC in Figure 13 .

The proposed model achieves higher accuracy and ROC as presented in Tables 8-11 and Figure 13 , authenticating the proposed model's effectiveness and contribution. Finally, the proposed approach outcomes are compared to the latest existing works, as depicted in Table 12 . 

The authors declare that they have no conflicts of interest to report regarding the present study.

The data that support the findings of this study are openly available in 

Correlation of chest CT and RT-PCR testing in coronavirus disease 2019 (COVID-19) in China: A report of 1014 cases

Brain tumor detection using statistical and machine learning method

Brain tumor classification: feature fusion

Brain tumor detection: A long short-term memory (LSTM)-based learning model

Use of machine intelligence to conduct analysis of human brain data for detection of abnormalities in its cognitive functions

Application of deep learning technique to manage COVID-19 in routine clinical practice using CT images: Results of 10 convolutional neural networks

Computer-aided covid-19 patient screening using chest images (X-ray and CT scans)

Encoder-decoder with atrous separable convolution for semantic image segmentation

Deep learning-based image conversion of CT reconstruction kernels improves Radiomics reproducibility for pulmonary nodules or masses

COVID-19 CT segmentation dataset

Hybrid segmentation method with confidence region detection for tumor identification

Targeted Self Supervision for Classification on a Small COVID-19 CT Scan Dataset

Microscopic abnormality classification of cardiac murmurs using ANFIS and HMM

The origin, transmission and clinical therapies on coronavirus disease 2019 (COVID-19) outbreak-an update on the status

Viral reverse engineering using artificial intelligence and big data COVID-19 infection with long short-term memory (LSTM)

Deep residual learning for image recognition

Synergistic learning of lung lobe segmentation and hierarchical multi-instance classification for automated severity assessment of COVID-19 in CT images

Retinal imaging analysis based on vessel detection

Automatic Detection of COVID-19 Infection from Chest X-ray using Deep Learning, medRixv

Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem

Multimodel deep neural network based features extraction and optimal selection approach for skin lesion classification

Prediction of COVID-19 -pneumonia based on selected deep features and one class kernel extreme learning machine

Hot-Spot zone detection to tackle COVID19 spread by fusing the traditional machine learning and deep learning approaches of computer vision

Lungs nodule detection framework from computed tomography images using support vector machine

Automated ulcer and bleeding classification from WCE images using multiple features fusion and selection

Lung cancer detection on CT scan images: A review on the analysis techniques

COVID-19 and artificial intelligence: Protecting healthcare workers and curbing the spread

The socio-economic implications of the coronavirus and COVID-19 pandemic: a review

Sparse coding with an overcomplete basis set: A strategy employed by V1? Vision Research

A comprehensive study on classification of COVID-19 on computed tomography with pretrained convolutional neural networks

A deep learning approach for automated diagnosis and multi-class classification of Alzheimer's disease stages using resting-state fMRI and residual neural networks

Microscopic melanoma detection and classification: A framework of pixel-based fusion and multilevel features reduction

Deep learning-based COVID-19 detection using CT and X-ray images: Current analytics and comparisons

Real-time diagnosis system of COVID-19 using X-ray images and deep learning

American Network of Coronavirus Disease 2019-COVID-19 Research (LANCOVID-19). (2020). Clinical, laboratory and imaging features of COVID-19: A systematic review and meta-analysis

Automated lung nodule detection and classification based on multiple classifiers voting

Machine learning techniques to detect and forecast the daily total COVID-19 infected and deaths cases under different lockdown types

A review of coronavirus disease-2019 (COVID-19)

World Health Organization declares global emergency: A review of the 2019 novel coronavirus (COVID-19)

Evaluation of coronavirus in tears and conjunctival secretions of patients with SARS-CoV-2 infection

COVID-CTdataset: A CT scan dataset about COVID-19. ArXiv e-Prints

COVID-19: What has been learned and to be learned about the novel coronavirus disease

Beyond a gaussian denoiser: Residual learning of deep CNN for Image Denoising

Covid-ct-dataset: A CT scan dataset about covid-19

Microscopic segmentation and classification of COVID-19 infection with ensemble convolutional neural network