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.

Semi-Supervised Modified-UNet for Lung Infection Image Segmentation

Automatic lung infection segmentation in computed tomography (CT) scans can offer great assistance in radiological diagnosis by improving accuracy and reducing time required for diagnosis. The biggest challenges for deep learning (DL) models in segmenting infection region are the high variances in infection characteristics, fuzzy boundaries between infected and normal tissues, and the troubles in getting large number of annotated data for training. To resolve such issues, we propose a Modified U-Net (Mod-UNet) model with minor architectural changes and significant modifications in the training process of vanilla 2D UNet. As part of these modifications, we updated the loss function, optimization function, and regularization methods, added a learning rate scheduler and applied advanced data augmentation techniques. Segmentation results on two Covid-19 Lung CT segmentation datasets show that the performance of Mod-UNet is considerably better than the baseline U-Net. Furthermore, to mitigate the issue of lack of annotated data, the Mod-UNet is used in a semi-supervised framework (Semi-Mod-UNet) which works on a random sampling approach to progressively enlarge the training dataset from a large pool of unannotated CT slices. Exhaustive experiments on the two Covid-19 CT segmentation datasets and on a real lung CT volume show that the Mod-UNet and Semi-Mod-UNet significantly outperform other state-of-theart approaches in automated lung infection segmentation. IEEE

HarDNet and Dual-Code Attention Mechanism based model for medical images segmentation

During the formation of medical images, they are easily disturbed by factors such as acquisition devices and tissue backgrounds, causing problems such as blurred image backgrounds and difficulty in differentiation. In this paper, we combine the HarDNet module and the multi-coding attention mechanism module to optimize the two stages of encoding and decoding to improve the model segmentation performance. In the encoding stage, the HarDNet module extracts medical image feature information to improve the segmentation network operation speed. In the decoding stage, the multi-coding attention module is used to extract both the position feature information and channel feature information of the image to improve the model segmentation effect. Finally, to improve the segmentation accuracy of small targets, the use of Cross Entropy and Dice combination function is proposed as the loss function of this algorithm. The algorithm has experimented on three different types of medical datasets, Kvasir-SEG, ISIC2018, and COVID-19CT. The values of JS were 0.7189, 0.7702, 0.9895, ACC were 0.8964, 0.9491, 0.9965, SENS were 0.7634, 0.8204, 0.9976, PRE were 0.9214, 0.9504, 0.9931. The experimental results showed that the model proposed in this paper achieved excellent segmentation results in all the above evaluation indexes, which can effectively assist doctors to diagnose related diseases quickly and improve the speed of diagnosis and patients’quality of life. Author

Towards Lifting the Trade-off between Accuracy and Adversarial Robustness of Deep Neural Networks with Application on COVID 19 CT Image Classification and Medical Image Segmentation

Deep neural networks (DNNs) are vulnerable to adversarial noises. Adversarial training is a general strategy to improve DNN robustness. But training a DNN model with adversarial noises may result in a much lower accuracy on clean data, which is termed the trade-off between accuracy and adversarial robustness. Towards lifting this trade-off, we propose an adversarial training method that generates optimal adversarial training samples. We evaluate our methods on PathMNIST and COVID-19 CT image classification tasks, where the DNN model is ResNet-18, and Heart MRI and Prostate MRI image segmentation tasks, where the DNN model is nnUnet. All these four datasets are publicly available. The experiment results show that our method has the best robustness against adversarial noises and has the least accuracy degradation compared to the other defense methods. © 2023 SPIE.

Medical image mis-segmentation region refinement framework based on dynamic graph convolution

In medical image segmentation tasks, it is hard for traditional Convolutional Neural Network (CNN) to capture essential information such as spatial structure and global contextual semantic features since it suffers from a limited receptive field. The deficiency weakens the CNN segmentation performance in the lesion boundary regions. To handle the aforementioned problems, a medical image mis-segmentation region refinement framework based on dynamic graph convolution is proposed to refine the boundary and under-segmentation regions. The proposed framework first employs a lightweight dual-path network to detect the boundaries and nearby regions, which can further obtain potentially misclassified pixels from the coarse segmentation results of the CNN. Then, we construct the pixels into the appropriate graphs by CNN-extracted features. Finally, we design a dynamic residual graph convolutional network to reclassify the graph nodes and generate the final refinement results. We chose UNet and its eight representative improved networks as the basic networks and tested them on the COVID, DSB, and BUSI datasets. Experiments demonstrated that the average Dice of our framework is improved by 1.79%, 2.29%, and 2.24%, the average IoU is improved by 2.30%, 3.53%, and 2.39%, and the Se is improved by 5.08%, 4.78%, and 5.31% respectively. The experimental results prove that the proposed framework has the refinement capability to remarkably strengthen the segmentation result of the basic network. Furthermore, the framework has the advantage of high portability and usability, which can be inserted into the end of mainstream medical image segmentation networks as a plug-and-play enhancement block.

Medical Image Diagnosis Using Deep Learning Classifiers for COVID-19

COVID-19 and other diseases must be precisely and swiftly classified to minimize disease spread and avoid overburdening the healthcare system. The main purpose of this study is to develop deep-learning classifiers for normal, viral pneumonia, and COVID-19 disorders using CXR pictures. Deep learning image classification algorithms are used to recognize and categorise image data to detect the presence of illnesses. The raw image must be pre-processed since deep neural networks perform the most important aspect of medical image identification, which includes translating the raw image into an intelligible format. The dataset includes three classifications, including normal and viral pneumonia and COVID-19. To aid in quick diagnosis and the proposed models leverage the performance validation of several models, which are summarised in the form of a recall, Fl-score, precision, accuracy, and AUC, to distinguish COVID-19 from other types of pneumonia. When all the deep learning classifiers and performance parameters were analyzed, the ResNetl0lV2 achieved the highest accuracy of COVID-19 classifications is 97.S2%, ResNetl0lV2 had the greatest accuracy of the normal categorization is 92.04% and the Densenet201 had the greatest accuracy of the pneumonia classification is 99.92%. The suggested deep learning system is an excellent choice for clinical use to aid in the COVID-19, normal, and pneumonia processes for diagnosing infections using CXR scans. Furthermore, the suggested approaches provided a realistic technique to implement in real-world practice, assisting medical professionals in diagnosing illnesses from CXR images. © 2023 IEEE.

Multi-Attentional U-Net for Medical Image Segmentation

Deep learning (DL) approaches for image segmentation have been gaining state-of-the-art performance in recent years. Particularly, in deep learning, U-Net model has been successfully used in the field of image segmentation. However, traditional U-Net methods extract features, aggregate remote information, and reconstruct images by stacking convolution, pooling, and up sampling blocks. The traditional approach is very inefficient due of the stacked local operators. In this paper, we propose the multi-attentional U-Net that is equipped with non-local blocks based self-attention, channel-attention, and spatial-attention for image segmentation. These blocks can be inserted into U-Net to flexibly aggregate information on the plane and spatial scales. We perform and evaluate the multi-attentional U-Net model on three benchmark data sets, which are COVID-19 segmentation, skin cancer segmentation, thyroid nodules segmentation. Results show that our proposed models achieve better performances with faster computation and fewer parameters. The multi-attention U-Net can improve the medical image segmentation results. © 2022 IEEE.

AI-enabled radiologist in the loop: novel AI-based framework to augment radiologist performance for COVID-19 chest CT medical image annotation and classification from pneumonia.

A SARS-CoV-2 virus-specific reverse transcriptase-polymerase chain reaction (RT-PCR) test is usually used to diagnose COVID-19. However, this test requires up to 2 days for completion. Moreover, to avoid false-negative outcomes, serial testing may be essential. The availability of RT-PCR test kits is currently limited, highlighting the need for alternative approaches for the precise and rapid diagnosis of COVID-19. Patients suspected to be infected with SARS-CoV-2 can be assessed using chest CT scan images. However, CT images alone cannot be used for ruling out SARS-CoV-2 infection because individual patients may exhibit normal radiological results in the primary phases of the disease. A machine learning (ML)-based recognition and segmentation system was developed to spontaneously discover and compute infection areas in CT scans of COVID-19 patients. The computable assessment exhibited suitable performance for automatic infection region allocation. The ML models developed were suitable for the direct detection of COVID-19 (+). ML was confirmed to be a complementary diagnostic technique for diagnosing COVID-19(+) by forefront medical specialists. The complete manual delineation of COVID-19 often requires up to 225.5 min; however, the proposed RILML method decreases the delineation time to 7 min after four iterations of model updating.

A prediction error based reversible data hiding scheme in encrypted image using block marking and cover image pre-processing.

A drastic change in communication is happening with digitization. Technological advancements will escalate its pace further. The human health care systems have improved with technology, remodeling the traditional way of treatments. There has been a peak increase in the rate of telehealth and e-health care services during the coronavirus disease 2019 (COVID-19) pandemic. These implications make reversible data hiding (RDH) a hot topic in research, especially for medical image transmission. Recovering the transmitted medical image (MI) at the receiver side is challenging, as an incorrect MI can lead to the wrong diagnosis. Hence, in this paper, we propose a MSB prediction error-based RDH scheme in an encrypted image with high embedding capacity, which recovers the original image with a peak signal-to-noise ratio (PSNR) of ∞ dB and structural similarity index (SSIM) value of 1. We scan the MI from the first pixel on the top left corner using the snake scan approach in dual modes: i) performing a rightward direction scan and ii) performing a downward direction scan to identify the best optimal embedding rate for an image. Banking upon the prediction error strategy, multiple MSBs are utilized for embedding the encrypted PHR data. The experimental studies on test images project a high embedding rate with more than 3 bpp for 16-bit high-quality DICOM images and more than 1 bpp for most natural images. The outcomes are much more promising compared to other similar state-of-the-art RDH methods.

Automatic COVID-19 and Common-Acquired Pneumonia Diagnosis Using Chest CT Scans.

Even with over 80% of the population being vaccinated against COVID-19, the disease continues to claim victims. Therefore, it is crucial to have a secure Computer-Aided Diagnostic system that can assist in identifying COVID-19 and determining the necessary level of care. This is especially important in the Intensive Care Unit to monitor disease progression or regression in the fight against this epidemic. To accomplish this, we merged public datasets from the literature to train lung and lesion segmentation models with five different distributions. We then trained eight CNN models for COVID-19 and Common-Acquired Pneumonia classification. If the examination was classified as COVID-19, we quantified the lesions and assessed the severity of the full CT scan. To validate the system, we used Resnetxt101 Unet++ and Mobilenet Unet for lung and lesion segmentation, respectively, achieving accuracy of 98.05%, F1-score of 98.70%, precision of 98.7%, recall of 98.7%, and specificity of 96.05%. This was accomplished in just 19.70 s per full CT scan, with external validation on the SPGC dataset. Finally, when classifying these detected lesions, we used Densenet201 and achieved accuracy of 90.47%, F1-score of 93.85%, precision of 88.42%, recall of 100.0%, and specificity of 65.07%. The results demonstrate that our pipeline can correctly detect and segment lesions due to COVID-19 and Common-Acquired Pneumonia in CT scans. It can differentiate these two classes from normal exams, indicating that our system is efficient and effective in identifying the disease and assessing the severity of the condition.

A robust NIfTI image authentication framework to ensure reliable and safe diagnosis.

Advancements in digital medical imaging technologies have significantly impacted the healthcare system. It enables the diagnosis of various diseases through the interpretation of medical images. In addition, telemedicine, including teleradiology, has been a crucial impact on remote medical consultation, especially during the COVID-19 pandemic. However, with the increasing reliance on digital medical images comes the risk of digital media attacks that can compromise the authenticity and ownership of these images. Therefore, it is crucial to develop reliable and secure methods to authenticate these images that are in NIfTI image format. The proposed method in this research involves meticulously integrating a watermark into the slice of the NIfTI image. The Slantlet transform allows modification during insertion, while the Hessenberg matrix decomposition is applied to the LL subband, which retains the most energy of the image. The Affine transform scrambles the watermark before embedding it in the slice. The hybrid combination of these functions has outperformed previous methods, with good trade-offs between security, imperceptibility, and robustness. The performance measures used, such as NC, PSNR, SNR, and SSIM, indicate good results, with PSNR ranging from 60 to 61 dB, image quality index, and NC all close to one. Furthermore, the simulation results have been tested against image processing threats, demonstrating the effectiveness of this method in ensuring the authenticity and ownership of NIfTI images. Thus, the proposed method in this research provides a reliable and secure solution for the authentication of NIfTI images, which can have significant implications in the healthcare industry.

New Binary Archimedes Optimization Algorithm and its application

Optimization problem, as a hot research field, is applied to many industries in the real world. Due to the complexity of different search spaces, metaheuristic optimization algorithms are proposed to solve this problem. As a recently introduced optimization method inspired by physics, Archimedes Optimization Algorithm (AOA) is an efficient metaheuristic algorithm based on Archimedes' law. It has the advantages of fast convergence speed and balance between local and global search ability when solving continuous problems. However, discrete problems exist more in practical applications. AOA needs to be further improved in dealing with such problems. On this basis, to make Archimedes Optimization Algorithm better applied to solve discrete problems, a Binary Archimedes Optimization Algorithm (BAOA) is proposed in this paper, which incorporates a novel V-shaped transfer function. The proposed method applies the BAOA to COVID-19 classification of medical data, segmentation of real brain lesion, and the knapsack problem. The experimental results show that the proposed BAOA can solve the discrete problem well.

ATFormer: Advanced transformer for medical image segmentation

Combining transformers and convolutional neural networks is considered one of the most important directions for tackling medical image segmentation problems. To learn the long-range dependencies and local contexts, previous approaches embedded a convolutional layer into feedforward neural network inside the transformer block. However, a common issue is the instability during training since large differences in amplitude across layers by pre-layer normalization. Furthermore, multi-scale features were directly fused using the transformer from the encoder to decoder, which could disrupt valuable information for segmentation. To address these concerns, we propose Advanced TransFormer (ATFormer), a novel hybrid architecture that combines convolutional neural networks and transformers for medical image segmentation. First, the traditional transformer block has been refined into an Advanced Transformer Block, which adopts post-layer normalization to obtain mild activation values and employs the scaled cosine attention with shifted window for accurate spatial information. Second, the Progressive Guided Fusion module is introduced to make multi-scale features more discriminative while reducing the computational complexity. Experimental results on the ACDC, COVID-19 CT-Seg, and Tumor datasets demonstrate the significant advantage of ATFormer over existing methods that rely solely on convolutional neural networks, transformers, or their combination.

The application of deep learning for COVID-19 diagnosis and treatment

The coronavirus disease 2019 (COVID-19) pandemic has resulted in a considerable increase in hospitalizations, leading to an increasing demand for accurate and efficient techniques for diagnosis. The CT-based diagnosis can provide pathologic information to assist treatment but be restricted due to inefficient and relatively complicated implementation. With the advent of deep learning and advanced hardware, an AI-assisted method diagnosis and segmentation for COVID-19 are proposed. In this paper, many traditional machine learning methods for imaging classification and segmentation are discussed, such as k-Nearest Neighbours (KNN), support vector machines (SVM), edge-based or region-based segmentation. In addition, we proposed a ResNet-based model and an improved U-Net for medical tasks of classification and segmentation, respectively. Our proposed model achieved desirable accuracy in medical applications. © 2023 SPIE.

COVID-19 image classification based on Deep Learning

COVID-19 is diagnosed by nucleic acid testing, aided by Computed Tomography. In order to rapidly screen CT images of COVID-19, Squeeze-And-Excitation Network based network model combined with Deep Learning is proposed, which can adapt to learn important parts of the feature channel. Firstly, the feature Squeeze is carried out along the space dimension, and the output dimension matches the number of input feature channels. Secondly, the feature channel learns the feature channel characteristics by capturing the channel dependencies in the previous step. Finally, the weight is updated to model the correlation of feature channels. The Precision, Recall and Specificity were selected to be 92.8%, 92.8% and 93.7%, the Accuracy of the model was 93.24% for the whole sample specificity. Compared with the mainstream model, the experimental results of this model are improved greatly. © 2022 ACM.

Detection of Covid-19 using Mask R-CNN

Deep learning has been widely used to analyze radiographic pictures such as chest scans. These radiographic pictures include a wealth of information, including patterns and cluster-like formations, which aid in the discovery and conformance of COVID-19-like pandemics. The COVID-19 pandemic is wreaking havoc on global well-being and public health. Until present, more than 27 million confirmed cases have been recorded globally. Due to the increasing number of confirmed cases and issues with COVID-19 variants, fast and accurate categorization of healthy and infected individuals is critical for COVID-19 management and treatment. In medical image analysis and classification, artificial intelligence (AI) approaches in general, and region-based convolutional neural networks (CNNs) in particular, have yielded promising results. In this study, a deep Mask R-CNN architecture based on chest image classification is suggested for the diagnosis of COVID-19. An effective and reliable Mask R-CNN classification was difficult due to a lack of sufficient size and high-quality chest image datasets. These complications are addressed with Mask Region-based convolutional neural networks (R-CNNs) as a framework for detecting COVID-19 patients from chest pictures using an open-source dataset. First, the model was evaluated using 100 photos from the original processed dataset, and it was found to be accurate. The model was then validated against an independent dataset of COVID-19 X-ray pictures. The suggested model outperformed all other models in general and specifically when tested using an independent testing set. © 2023 IEEE.

Identification and Localization of COVID-19Abnormalities on Chest Radiographs using Computer Vision

The Corona Virus (COVID 19) pandemic is quickly becoming the world's most deadly disease. The spreading rate is higher and the early detection helps in faster recovery. The existence of COVID 19 in individuals shall be detected using molecular analysis or through radiographs of lungs. As time and test kit are limited RT- PCR is not suitable to test all. The RT- PCR being a time-consuming process, diagnosis using chest radiographs needs no transportation as the modern X-ray systems are digitized. Deep learning takes an edge over other techniques as it deduces the features automatically and performs massively parallel computations. Multiple feature maps will help in accurate prediction. The objective of the proposed work is to develop a Computer Aided Deep Learning System identify and localize COVID-19 virus from other viruses and pneumonia. It helps to detect COVID-19 within a short period of time thereby improving the lifetime of the individuals. SIIM-FISABIO-RSNA benchmark datasets are used to examine the proposed system. Recall, Precision, Accuracy-rate, and F-Measure are the metrics used to prove the integrity of the system. © 2023 IEEE.

Deep learning model for binary classification of COVID-19 based on Chest X-Ray

COVID-19 is a novel virus infecting the upper respiratory tract and lungs. On a scale of the global pandemic, the number of cases and deaths had been increasing each day. Chest X-ray (CXR) images proved effective in monitoring a variety of lung illnesses, including the COVID-19 disease. In recent years, deep learning (DL) has become one of the most significant topics in the computing world and has been extensively applied in several medical applications. In terms of automatic diagnosis of COVID-19, those approaches had proven to be very effective. In this research, a DL technology based on convolution neural networks (CNN) models had been implemented with less number of layers with tuning parameters that will take less time for training for binary classification of COVID-19 based on CXR images. Experimental results had shown that the proposed model for training had achieved an accuracy of 96.68%, Recall of 94.12%, Precision of 93.49%, Specificity of 97.61%, and F1 Score of 93.8%. Those results had shown the high value of utilizing DL for early COVID-19 diagnosis, which can be utilized as a useful tool for COVID-19 screening. © 2023 IEEE.

Secure NIfTI Image Authentication Scheme for Modern Healthcare System

Advances in digital neuroimaging technologies, i.e., MRI and CT scan technology, have radically changed illness diagnosis in the global healthcare system. Digital imaging technologies produce NIfTI images after scanning the patient's body. COVID-19 spared on a worldwide effort to detect the lung infection. CT scans have been performed on billions of COVID-19 patients in recent years, resulting in a massive amount of NIfTI images being produced and communicated over the internet for diagnosis. The dissemination of these medical photographs over the internet has resulted in a significant problem for the healthcare system to maintain its integrity, protect its intellectual property rights, and address other ethical considerations. Another significant issue is how radiologists recognize tempered medical images, sometimes leading to the wrong diagnosis. Thus, the healthcare system requires a robust and reliable watermarking method for these images. Several image watermarking approaches for .jpg, .dcm, .png, .bmp, and other image formats have been developed, but no substantial contribution to NIfTI images (.nii format) has been made. This research suggests a hybrid watermarking method for NIfTI images that employs Slantlet Transform (SLT), Lifting Wavelet Transform (LWT), and Arnold Cat Map. The suggested technique performed well against various attacks. Compared to earlier approaches, the results show that this method is more robust and invisible.

A GoogLeNet Performance Approach for COVID-19 Detection using Chest X-ray Images

Coronavirus disease (COVID-19) is a major pandemic disease that has already infected millions of people worldwide and affects many aspects, especially public health. There are many clinical techniques for the diagnosis of this disease, such as RT-PCR and CT-Scan. X-ray image is one of the important techniques for medical diagnosis and easily accessible in classifying suspected cases of COVID-19 infection. In this study, we classified COVID-19 images with four classes: COVID-19, Normal, Lung opacity and Viral pneumonia by compared three models: EfficientNetB0, MobileNet and GoogLeNet for the performance of classification using 1,000 chest X-ray images from Kaggle dataset within scenario of resource limitations. The experiment reveals that GoogLeNet shows superiority over other models that produced the highest accuracy results of 88% and F1 score of 0.88 with a total time of 1 hour and 15 minutes. Along with its confusion matrix that shows model can better classify images than other models. © 2023 IEEE.