Automatic Detection of Lung Ultrasound Artifacts using a Deep Neural Networks approach

The COVID-19 pandemic has challenged many of the healthcare systems around the world. Many patients who have been hospitalized due to this disease develop lung damage. In low and middle-income countries, people living in rural and remote areas have very limited access to adequate health care. Ultrasound is a safe, portable and accessible alternative;however, it has limitations such as being operator-dependent and requiring a trained professional. The use of lung ultrasound volume sweep imaging is a potential solution for this lack of physicians. In order to support this protocol, image processing together with machine learning is a potential methodology for an automatic lung damage screening system. In this paper we present an automatic detection of lung ultrasound artifacts using a Deep Neural Network, identifying clinical relevant artifacts such as pleural and A-lines contained in the ultrasound examination taken as part of the clinical screening in patients with suspected lung damage. The model achieved encouraging preliminary results such as sensitivity of 94%, specificity of 81%, and accuracy of 89% to identify the presence of A-lines. Finally, the present study could result in an alternative solution for an operator-independent lung damage screening in rural areas, leading to the integration of AI-based technology as a complementary tool for healthcare professionals. © 2023 SPIE.

A Shuffling Building Block and Augmentation Parameter Tuning Techniques to Handle Small Medical Dataset

The recent COVID-19 outbreak showed us the importance of faster disease diagnosis using medical image processing as it is considered the most reliable and accurate diagnostic tool. In a CNN architecture, performance improves with the increasing number of trainable parameters at the cost of processing time. We have proposed an innovative approach of combining efficient novel architectures like Inception, ResNet, and ResNet-Xt and created a new CNN architecture that benefits Extreme Cardinal dimensions. We have also created four variations of the same base architecture by varying the position of each building block and used X-Ray, Microscopic, MRI, and pathMNIST datasets to train our architecture. For learning curve optimization, we have applied learning rate changing techniques, tuned image augmentation parameters, and chose the best random states value. For a specific dataset, we reduced the validation loss from 0.22 to 0.18 by interchanging the architecture's building block position. Our results indicate that image augmentation parameters can help to decrease the validation loss. We have also shown rearrangement of the building blocks reduces the number of parameters, in our case, from 5,689,008 to 3,876,528. © 2023 IEEE.

Explainability of deep learning models in medical image classification

The ability to explain the reasons for one's decisions to others is an important aspect of being human intelligence. We will look at the explainability aspects of the deep learning models, which are most frequently used in medical image processing tasks. The Explainability of machine learning models in medicine is essential for understanding how the particular ML model works and how it solves the problems it was designed for. The work presented in this paper focuses on the classification of lung CT scans for the detection of COVID-19 patients. We used CNN and DenseNet models for the classification and explored the application of selected visual explainability techniques to provide insight into how the model works when processing the images. © 2022 IEEE.

Innovative and efficient CNN structure works for chest X-ray images classification

In the era of global transmission of COVID-19, it is a challenge for physicians to efficiently and accurately use chest Xray images to diagnose whether a patient is positive or not. The application of deep learning and computer vision in medical image processing solves this problem, but a highly accurate method is still needed. In this research, we proposed an innovative CNN structure used for chest X-ray classification. Based on deep learning and CNN, this new architecture has an efficient training process and the performance of accuracy is better than other classic nets. The best accuracy on the test dataset is 97.68%. It has competitive advantages over AlexNet, LeNet-5, and Vgg-16. Dropout, Data augmentation, and Grad-CAM technique are added to improve performance. © 2022 IEEE.

Lung Segmentation using MultiResUNet CNN based on Computed Tomography Image

Lung segmentation is the first step in medical image processing to determine various lung diseases. Currently, the image segmentation process will be more optimal by using deep learning through the convolution process. Various Convolution Neural Network (CNN) based architectures for image segmentation were created by many researchers, however U-Net is the current state of the art for medical image segmentation. Nevertheless, the modification of U-Net continues, and MultiResUNet is one of the new architectures claimed to be better. In this study, we use MultiResUNet for lung segmentation on Computed Tomography (CT) images as the first step to Covid-19 infection segmentation, and the results will be compared using the U-Net architecture. Based on the results of the segmentation experiment, we got satisfactory results. Using the Mean-IoU evaluation metric, it was concluded that the MultiResUNet score was slightly better than the U-Net score for patient lung segmentation, where there was an increase in the score of 1.33% (MultiResUNet=93.05%, U-Net=91.83%) in the dataset which we use. © 2022 IEEE.

Caps-3WD: A three-way diagnostic CapsNet for COVID-19

Pulmonary medical image processing is an effective diagnostic method for COVID-19, and CapsNet-based methods have achieved good performance. However, as cost-blind methods, these diagnostic methods only consider immediate and deterministic decisions, which easily lead to misdiagnosis and high costs. Therefore, based on a revised CapsNet, we propose a cost-sensitive three-way decision (3WD) method for COVID-19 diagnosis, named as Caps-3WD. To enhance the feature extraction ability for pneumonia areas, we introduce a Restage module to improve convolution layer of the original CapsNet. Further, to lighten the model, we introduce depth wise separable convolution to reconstruct decoder. Additionally, three options are considered in the decision set: infected, normal, and suspected, which are given different costs, respectively. The lowest-cost decision is chosen for each input. In the experimental analysis, we compare Caps-3WD with CNN-based and CapsNet-based methods on COVID-CXR dataset, which proves the effectiveness of 3WD and the superiority of Caps-3WD in COVID-19 diagnosis. © 2022 SPIE. Downloading of the is permitted for personal use only.

Comparing Feature Sets of Pseudo Zernike Moment and Minkowski Functionals to Classify Normal Subjects from COVID Subjects

Aim: Objective of this study is to analyze the efficiency of Pseudo Zernike Moment in differentiating COVID subjects from controls compared to Minkowski Functionals. Materials and Methods: The data for this study is obtained from a publicly available dataset. By fixing predefined values to the parameters such as effect size and algorithm power as 0.3 and 0.80 in G power tool provides the required sample size as 176. Pseudo Zernike moments and Minkowski features are extracted from the binary lung CT scans. Result: Pseudo Zernike moment feature (M2) is found to have a mean value of 0.63 for normal subjects and 0.56 for COVID subjects. Minkowski area feature is found to have the ability to differentiate COVID subject compared to its other features. Pseudo Zernike features exhibit better statistical significance (p<0.05) in differentiating normal and COVID subjects. Neural network classifier shows better classification ability with 91% classification accuracy in separating COVID subjects from normal controls. Conclusion: Compared to Minkowski features, pseudo-Zernike moments has better classification ability to differentiate normal and COVID subjects. © 2022 IEEE.

U-Net-based COVID-19 CT Image Semantic Segmentation: A Transfer Learning Approach

Deep learning (DL) algorithms are widely applied in many disciplines such as medical imaging, bioinformatics, and computer vision. DL models have been used in medical imaging to perform image segmentation, classification, and detection. During the outbreak of the COVID-19 pandemic, DL has been extensively used to develop COVID-19 screening systems. The reverse transcription-polymerase chain reaction (RT-PCR) is the gold standard method for COVID-19 screening. However, DL has been proposed to detect patients infected with COVID-19 through radiological imaging in Chest X-rays and chest computed tomography (CT) images. This paper proposes transfer learning to train modified U-Net models to segment the COVID-19 chest CT images into two regions of lung infection (ground-glass and consolidation). The proposed modified U-Net models were constructed by replacing the encoder part with a pre-trained convolutional neural network (CNN) model. Three pre-trained CNN models, namely, EfficientNet-b0, EfficientNet-b1, and EfficientNet-b2 were used. The proposed models were evaluated on the COVID-19 CT Images Segmentation dataset available in an open Kaggle challenge. The obtained results show that the proposed EfficientNet-b2_U-Net model yielded the highest FScore of 0.5666. © 2022 IEEE.

COVID-19 lesion segmentation using convolutional LSTM for self-attention

We propose a fast and robust multi-class deep learning framework for segmenting COVID-19 lesions: Ground-Glass opacities and High opacities (including consolidations and pleural effusion), from non-contrast CT scans using convolutional Long Short-Term Memory network for self-attention. Our method allows rapid quantification of pneumonia burden from CT with performance equivalent to expert readers. The mean dice score across 5 folds was 0.8776 with a standard deviation of 0.0095. A low standard deviation between results from each fold indicate the models were trained equally good regardless of the training fold. The cumulative per-patient mean dice score (0.8775±0.075) for N=167 patients, after concatenation, is consistent with the results from each of the 5 folds. We obtained excellent Pearson correlation (expert vs. automatic) of 0.9396 (p<0.0001) and 0.9843 (p<0.0001) between ground-glass opacity and high opacity volumes, respectively. Our model outperforms Unet2d (p<0.05) and Unet3d (p<0.05) in segmenting high opacities, has comparable performance with Unet2d in segmenting ground-glass opacities, and significantly outperforms Unet3d (p<0.0001) in segmenting ground-glass opacities. Our model performs faster on CPU and GPU when compared to Unet2d and Unet3d. For same number of input slices, our model consumed 0.83x and 0.26x the memory consumed by Unet2d and Unet3d. © 2022 SPIE

Graph Interaction for Automated Diagnosis of Thoracic Disease Using X-ray Images

Deep learning has shown successful performance not only in supervised disease detection but also lesion localization under the weakly supervised learning framework with medical image processing. However, few consider the semantic relationship among the diseases and lesions which plays a critical role in actual clinical diagnosis. In this work, we propose a novel framework: Feature map Graph Representational Probabilistic Class Activation Map (FGR-PCAM) to learn the graph structure of lesion-specific features and consider these relationships while also leveraging the localization ability of PCAM. Considering the relations of localized lesion-specific features has been shown to enhance both thoracic diseases classification and localization tasks on CheXpert and ChestXray14 datasets. Accurate classification and localization of Chest X-ray images would also help us fight against the COVID-19 and unveil COVID-19 fingerprints. © 2022 SPIE

Advancement of Deep Learning and Its Substantial Impact on the Diagnosis of COVID-19 Cases

A subset of machine learning is called Deep Learning (DL). Due to its intelligent behavior, it is used in various applications like speech recognition, face recognition, detection of an image, Natural Language Processing (NLP), analysis of video images, etc. Medical image processing is one of the significant area where deep learning network performance is proved outstanding. DL is used in image classification, dimensionality reduction, feature learning, detection, etc. The large volume of image data is processed and analyzed to predict disease is absent/present. In 2019, the COVID-19 virus was detected and started spreading through the community transmission rapidly, and several people lost their lives due to lack of treatment. Almost all hospitals all over the countries were overloaded heavily, and the Medical Health Care System was affected significantly. To fight against such a tough time, many researchers put their efforts day and night into designing effective deep learning models that can accurately speed up the COVID-19 viral diagnosis process. With this knowledge, a review of various deep learning algorithms and techniques for diagnosing Covid-19 cases is presented in this paper. It starts with the introduction of deep learning, its architecture, and different deep learning algorithms used to diagnose COVID-19 cases with key issues and challenges that significantly impact the detection of COVID-19 cases. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

A Review of CNN on Medical Imaging to Diagnose COVID-19 Infections*

In this paper, we study the Convolutional Neural Network (CNN) applications in medical image processing during the battle against Coronavirus Disease 2019 (COVID-19). Specifically, three CNN implementations are examined: CNN-LSTM, COVID-Net, and DeTraC. These three methods have been shown to offer promising implications for the future of CNN technology in the medical field. This survey explores how these technologies have improved upon their predecessors. Qualitative and quantitative analyses have strongly suggested that these methods perform significantly better than the commensurate technologies. After analyzing these CNN implementations, it is reasonable to conclude that this technology has a place in the future of the medical field, which can be used by professionals to gain insight into new diseases and to help in diagnosing infections using medical imaging. © 2021, EasyChair. All rights reserved.

Analysing the Changes in Shape of Lung using Pseudo Zernike Moments in Comparison to Invariant Moments

Aim: The aim of the analysis is to estimate the deformation in the shape of the lung due to incidence of COVID using pseudo Zernike moments in comparison to invariant moments. Materials and Methods: Images are obtained from Kaggle. Sample size of 176 acquired for the study using G power by considering factors effect size, standard error rate, algorithm power as 0.3, 0.05, 0.80 respectively. In this analysis the classification of normal and COVID subjects is made using seven invariant and pseudo-Zernike moment features. Classification is made using a neural network after extracting the feature values. Result: From the obtained results, the feature values of invariant moments were observed to be statistically significant (p<0.05) than pseudo-Zernike moments. The mean and standard deviation values of variance for normal and COVID subjects were (0.18\± 0.13,0.10± 0.13). For pseudo Zernike's M2 feature statistical values of normal and COVID subjects were (0.63± 0.22,0.56± 0.23). From the values, it is observed that the COVID subjects had loss in shape of lungs due to abnormality. Variance, skewness and kurtosis were found to be statistically significant in differentiating normal and COVID subjects. The accuracy and F1 score values of invariant moments were 0.98 and 0.97 respectively. Conclusion: Therefore, from this analysis it is observed that invariant moments provide significantly better classification between normal and COVID subjects when compared to pseudo Zernike moments. © 2022 IEEE.

A Review on Deep Learning based diagnosis of COVID-19 from X-ray and CT Images

More than 400 million cases of the new coronavirus (COVID-19) have been confirmed since December 2019 in more than 200 countries. Since the spread of original COVID-19 virus SARS-CoV-2, thousands of mutations have been discovered. The most dominant ones are Alpha, Beta, Gama, Delta and Omicron variants, with the Omicron variant rapidly spreading and dominating the current phase of the COVID wave across the globe. It needs early detection and self-isolation to contain the virus. Molecular tests like rRTPCR are common for its detection. However, with the current spreading rate and lack of availability of large-scale testing laboratories, rapid diagnosis has become difficult. COVID-19 diagnosis from CT and X-ray images using deep learning techniques has been the subject of a lot of research in the last two years. This work presents a review of these studies sourced from top databases such as Web of Science and highlights challenges and research gaps with future research directions. © 2022 IEEE.

Security in Big Data Health Care System

There has been a fundamental shift in the way firms in every industry manage, examine, and utilize their data. Health care is one of the most promising industries in which the use of big data may make a positive impact. Healthcare technology is being improved at a fast rate as an outcome of growing information and innovative innovation. In healthcare, there are different articles of big data. Digital medical data, biometric data, medical image processing, biosensor data, physician data, patient information, and administrative data are examples of these types. Many combined technologies are being deployed to modify healthcare systems in the COVID-19 pandemic. The security of medical data is required for the management of an integrated healthcare solution. In this paper, we found that many researchers face significant hurdles in encrypting sensitive patient information to prevent misuse or leakage. Our aim is to provide a focus on security issues in healthcare system and try to give a solution. © 2021 IEEE.

A Comprehensive Analysis of Identifying Lung Cancer via Different Machine Learning Approach

Now a day's lungs are very important body part due to indomitable cancer problem rice in patient. Lung cancer is the most harmful disease in human life. There are many patients, how safer from cancer now a day's also suffer from covid related problems. In this survey paper discuss the different machine learning approach to disease lung cancer. for the detection of lung cancer. using machine learning first we collection the training data set for testing data set with the help of training of data set we have to learn system learn machine to disease the lung cancer. In this article discuss different machine learning approach for cancer detection using medical image processing (MIP) techniques. Image proceeding's help to detection the cancer and machine learning technique prediction cell origination. Deep neural network is powerful tool for detection such type of deceases detection. In the review discuss the different techniques and it's specification. © 2021 IEEE.

X-ray Images Detection of COVID-19 Based on Deepwise Separable DenseNet

The Coronavirus Disease (COVID-19) is spreading worldwide. X-ray imaging plays an important role in the diagnosis of COVID-19. In order to help doctors diagnose COVID-19 effectively, we proposed a novel model (DS-DenseNet), which based on depth separable dense. By adding an improved depth separable convolution layer, we reduced the amount of parameters and make the model lighter. In the viral pneumonia, COVID-19 and normal lung, 2905 sets of chest X-ray images were collected, and the restricted contrast limited adaptive histogram equalization (CLAHE) algorithm was applied to preprocess the images and the preprocessed images were input into the model. Meanwhile, SDensenet, VGG16, Resnet18, Resnet34 and Densenet121 were introduced as baseline models. Compared with Resnet34, the sensitivity, accuracy and specificity of DS-Densenet are increased by 2.5%, 2.0% and 1.5% respectively;compared with SDensenet, the number of parameters is reduced by 44.0%, but the effect is not reduced. The experimental results show that the depth separable convolution can effectively reduce the model parameters, and the proposed DS-Densenet has a good classification effect, which has a certain significance for the auxiliary diagnosis of COVID-19. © 2021 IEEE.

Survey on using deep learning to detection COVID-19 disease

The spread of COVID-19 disease rapidly worldwide and the increase in deaths are a threat to humanity. This threat prompted researchers in deep learning (DL) to find ways to diagnose COVID-19 from computed tomography (CT) or x-rays. Working deep learning identifies the infection accurately through medical imaging, and the practising radiologist can diagnose the illness. This survey will discuss the reason behind deep learning and the technology used in medical image processing. Exposure to the most common research in the recent period uses deep learning techniques in the medical field. We will then collect research related to diagnosing COVID-19 by using medical images, studying them, discussing the better future suggestion and methods proposed by other researchers. We focus on initial available research that detects COVID-19 by deep learning and sees how they can save time and effort in this field. © 2021 IEEE

Reinforcement learning based diagnosis and prediction for COVID-19 from CT images

As we all know, COVID-19 is causing more and more human infections and deaths. In order to quickly and efficiently detect COVID-19, this paper has firstly proposed a detection framework based on reinforcement learning for COVID-19 diagnosis. We use the accuracy of the validation set as the reward value, and obtain the initial model for the next epoch by searching the model corresponding to the maximum reward value in each epoch. We also have proposed a prediction framework that integrates multiple detection frameworks using parameter sharing to predict the progression of patients' disease. We experimented with our own dataset screened by professional physicians and obtained more excellent results. In external validation, we still achieved a high accuracy rate without additional training. Finally, the experimental results show that our classification accuracy can reach 96.81%, and the precision, sensitivity, specificity, and AUC (Area Under Curve) are 95.47%, 98.64%, 95.91%, and 0.9698, respectively. The accuracy of external verification can reach 93.04% and 90.85%. The accuracy of our prediction framework is 91.04%. A large number of experiments have proved that our proposed method is effective and robust for COVID-19 detection and prediction. © SPIE 2021.