Mixup-Inf-Net: A data augmentation algorithm for segmentation of new coronary pneumonia infections

In 2020, the global spread of Coronavirus Disease 2019 exposed entire world to a severe health crisis. This has limited fast and accurate screening of suspected cases due to equipment shortages and and harsh testing environments. The current diagnosis of suspected cases has benefited greatly from the use of radiographic brain imaging, also including X-ray and scintigraphy, as a crucial addition to screening tests for new coronary pneumonia disease. However, it is impractical to gather enormous volumes of data quickly, which makes it difficult for depth models to be trained. To solve these problems, we obtained a new dataset by data augmentation Mixup method for the used chest CT slices. It uses lung infection segmentation (Inf-Net [1]) in a deep network and adds a learning framework with semi-supervised to form a Mixup-Inf-Net semi-supervised learning framework model to identify COVID-19 infection area from chest CT slices. The system depends primarily on unlabeled data and merely a minimal amount of annotated data is required;therefore, the unlabeled data generated by Mixup provides good assistance. Our framework can be used to improve improve learning and performance. The SemiSeg dataset and the actual 3D CT images that we produced are used in a variety of tests, and the analysis shows that Mixup-Inf-Net semi-supervised outperforms most SOTA segmentation models learning framework model in this study, which also enhances segmentation performance. © 2023 SPIE.

SAUNet++: an automatic segmentation model of COVID-19 lesion from CT slices

The coronavirus disease 2019 (COVID-19) epidemic has spread worldwide and the healthcare system is in crisis. Accurate, automated and rapid segmentation of COVID-19 lesion in computed tomography (CT) images can help doctors diagnose and provide prognostic information. However, the variety of lesions and small regions of early lesion complicate their segmentation. To solve these problems, we propose a new SAUNet++ model with squeeze excitation residual (SER) module and atrous spatial pyramid pooling (ASPP) module. The SER module can assign more weights to more important channels and mitigate the problem of gradient disappearance;the ASPP module can obtain context information by atrous convolution using various sampling rates. In addition, the generalized dice loss (GDL) can reduce the correlation between lesion size and dice loss, and is introduced to solve the problem of small regions segmentation of COVID-19 lesion. We collected multinational CT scan data from China, Italy and Russia and conducted extensive comparative and ablation studies. The experimental results demonstrated that our method outperforms state-of-the-art models and can effectively improve the accuracy of COVID-19 lesion segmentation on the dice similarity coefficient (our: 87.38% vs. U-Net++: 84.25%), sensitivity (our: 93.28% vs. U-Net++: 89.85%) and Hausdorff distance (our: 19.99 mm vs. U-Net++: 26.79 mm), respectively.

COVID-19 Image Segmentation Algorithms Based on Conditional Generative Adversarial Network

The size and location of the lesions in CT images of novel corona virus pneumonia (COVID-19) change all the time, and the lesion areas have low contrast and blurred boundaries, resulting in difficult segmentation. To solve this problem, a COVID-19 image segmentation algorithm based on conditional generative adversarial network (CGAN) is proposed. Uses the improved DeeplabV3+ network as a generator, which enhances the extraction of multi-scale contextual features, reduces the number of network parameters and improves the training speed. A Markov discriminator with 6 fully convolutional layers is proposed instead of a common discriminator, with the aim of focusing more on the local features of the CT image. By continuously adversarial training between the generator and the discriminator, the network weights are optimised so that the final segmented image generated by the generator is infinitely close to the ground truth. On the COVID-19 CT public dataset, the area under the curve of ROC, F1-Score and dice similarity coefficient achieved 96.64%, 84.15% and 86.14% respectively. The experimental results show that the proposed algorithm is accurate and robust, and it has the possibility of becoming a safe, inexpensive, and time-saving medical assistant tool in clinical diagnosis, which provides a reference for computer-aided diagnosis. © 2022 ACM.

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

Video-Based Online Interactive System Using CNN Segmentation Method

In recent years, the COVID-19 has made it difficult for people to interact with each other face-to-face, but various kinds of social interactions are still needed. Therefore, we have developed an online interactive system based on the image processing method, that allows people in different places to merge the human region of two images onto the same image in real-time. The system can be used in a variety of situations to extend its interactive applications. The system is mainly based on the task of Human Segmentation in the CNN (convolution Neural Network) method. Then the images from different locations are transmitted to the computing server through the Internet. In our design, the system ensures that the CNN method can run in real-time, allowing both side users can see the integrated image to reach 30 FPS when the network is running smoothly. © 2023 IEEE.

Integrating deep learning algorithms for the lung segmentation and body-part-specific anatomical classification with Medical Imaging and Data Resource Center (MIDRC)

Medical Imaging and Data Resource Center (MIDRC) has been built to support AI-based research in response to the COVID-19 pandemic. One of the main goals of MIDRC is to make data collected in the repository ready for AI analysis. Due to data heterogeneity, there is a need to standardize data and make data-mining easier. Our study aims to stratify imaging data according to underlying anatomy using open-source image processing tools. The experiments were performed using Google Colaboratory on computed tomography (CT) imaging data available from the MIDRC. We adopted the existing open-source tools to process CT series (N=389) to define the image sub-volumes according to body part classification, and additionally identified series slices containing specific anatomic landmarks. Cases with automatically identified chest regions (N=369) were then processed to automatically segment the lungs. In order to assess the accuracy of segmentation, we performed outlier analysis using 3D shape radiomics features extracted from the left and right lungs. Standardized DICOM objects were created to store the resulting segmentations, regions, landmarks and radiomics features. We demonstrated that the MIDRC chest CT collections can be enriched using open-source analysis tools and that data available in MIDRC can be further used to evaluate the robustness of publicly available tools. © 2023 SPIE.

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

Implementation of Image Quality Improvement Methods and Lung Segmentation on Chest X-Ray Images Using U-Net Architectural Modifications

COVID-19 is an infectious disease that causes acute respiratory distress syndrome due to the SARS-CoV-2 virus. Rapid and accurate screening and early diagnosis of patients play an essential role in controlling outbreaks and reducing the spread of this disease. This disease can be diagnosed by manually reading CXR images, but it is time-consuming and prone to errors. For this reason, this research proposes an automatic medical image segmentation system using a combination of U-Net architecture with Batch Normalization to obtain more accurate and fast results. The method used in this study consists of pre-processing using the CLAHE method and morphology opening, CXR image segmentation using a combination of U-Net-4 Convolution Block architecture with Batch Normalization, then evaluated using performance measures such as accuracy, sensitivity, specificity, F1-score, and IoU. The results showed that the U-Net architecture modified with Batch Normalization had successfully segmented CXR images, as seen from all performance measurement values above 94%.

U-Net based Semantic Segmentation for Touchless Fingerprint Technology: A Reflective Review

Touch-based fingerprints are widely used in today's world;even with all the success, the touch-based nature of these is a threat, especially in this COVID-19 period. A solution to the same is the introduction of Touchless Fingerprint Technology. The workflow of a touchless system varies vastly from its touch-based counterpart in terms of acquisition, pre-processing, image enhancement, and fingerprint verification. One significant difference is the methods used to segment desired fingerprint regions. This literature focuses on pixel-level classification or semantic segmentation using U-Net, a key yet challenging task. A plethora of semantic segmentation methods have been applied in this field. In this literature, a spectrum of efforts in the field of semantic segmentation using U-Net is investigated along with the components that are integral while training and testing a model, like optimizers, loss functions, and metrics used for evaluation and enumeration of results obtained. © 2022 IEEE.

Diagnosis of COVID-19 from the X-Ray images using BAT Algorithm with Deep Convolutional Neural Network

The COVID-19 widespread has posed a chief contest to the scientific community around the world. For patients with COVID-19 illness, the international community is working to uncover, implement, or invent new approaches for diagnosis and action. A opposite transcription-polymerase chain reaction is currently a reliable tactic for diagnosing infected people. This is a time- and money-consuming procedure. Consequently, the development of new methods is critical. Using X-ray images of the lungs, this research article developed three stages for detecting and diagnosing COVID-19 patients. The median filtering is used to remove the unwanted noised during pre-processing stage. Then, Otsu thresholding technique is used for segmenting the affected regions, where Spider Monkey Optimization (SMO) is used to select the optimal threshold. Finally, the optimized Deep Convolutional Neural Network (DCNN) is used for final classification. The benchmark COVID dataset and balanced COVIDcxr dataset are used to test projected model's performance in this study. Classification of the results shows that the optimized DCNN architecture outperforms the other pre-trained techniques with an accuracy of 95.69% and a specificity of 96.24% and sensitivity of 94.76%. To identify infected lung tissue in images, here SMO-Otsu thresholding technique is used during the segmentation stage and achieved 95.60% of sensitivity and 95.8% of specificity. © 2023 IEEE.

Deep learning-based COVID-19 triage tool: An observational study on an X-ray dataset

Background: Easy availability, low cost, and low radiation exposure make chest radiography an ideal modality for coronavirus disease 2019 (COVID-19) detection. Objective(s): In this study, we propose the use of an artificial intelligence (AI) algorithm to automatically detect abnormalities associated with COVID-19 on chest radiographs. We aimed to evaluate the performance of the algorithm against the interpretation of radiologists to assess its utility as a COVID-19 triage tool. Material(s) and Method(s): The study was conducted in collaboration with Kaushalya Medical Trust Foundation Hospital, Thane, Maharashtra, between July and August 2020. We used a collection of public and private datasets to train our AI models. Specificity and sensitivity measures were used to assess the performance of the AI algorithm by comparing AI and radiology predictions using the result of the reverse transcriptase-polymerase chain reaction as reference. We also compared the existing open-source AI algorithms with our method using our private dataset to ascertain the reliability of our algorithm. Result(s): We evaluated 611 scans for semantic and non-semantic features. Our algorithm showed a sensitivity of 77.7% and a specificity of 75.4%. Our AI algorithm performed better than the radiologists who showed a sensitivity of 75.9% and specificity of 75.4%. The open-source model on the same dataset showed a large disparity in performance measures with a specificity of 46.5% and sensitivity of 91.8%, thus confirming the reliability of our approach. Conclusion(s): Our AI algorithm can aid radiologists in confirming the findings of COVID-19 pneumonia on chest radiography and identifying additional abnormalities and can be used as an assistive and complementary first-line COVID-19 triage tool.Copyright © Cancer Research, Statistics, and Treatment.

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.

LLDSNet: Limited labeled data segmentation network for COVID-19 CT lesion segmentation

With the COVID-19 outbreak in 2019, the world is facing a major crisis and people's health is at serious risk. Accurate segmentation of lesions in CT images can help doctors understand disease infections, prescribe the right medicine and control patients' conditions. Fast and accurate diagnosis not only can make the limited medical resources get reasonable allocation, but also can control the spread of disease, and computer-aided diagnosis can achieve this purpose, so this paper proposes a deep learning segmentation network LLDSNet based on a small amount of data, which is divided into two modules: contextual feature-aware module (CFAM) and shape edge detection module (SEDM). Due to the different morphology of lesions in different CT, lesions with dispersion, small lesion area and background area imbalance, lesion area and normal area boundary blurred, etc. The problem of lesion segmentation in COVID-19 poses a major challenge. The CFAM can effectively extract the overall and local features, and the SEDM can accurately find the edges of the lesion area to segment the lesions in this area. The hybrid loss function is used to avoid the class imbalance problem and improve the overall network performance. It is demonstrated that LLDSNet dice achieves 0.696 for a small number of data sets, and the best performance compared to five currently popular segmentation networks. © 2023 SPIE.

An Improved Soft Subspace Clustering Algorithm for Brain MR Image Segmentation

In recent years, the soft subspace clustering algorithm has shown good results for high-dimensional data, which can assign different weights to each cluster class and use weights to measure the contribution of each dimension in various features. The enhanced soft subspace clustering algorithm combines interclass separation and intraclass tightness information, which has strong results for image segmentation, but the clustering algorithm is vulnerable to noisy data and dependence on the initialized clustering center. However, the clustering algorithm is susceptible to the influence of noisy data and reliance on initialized clustering centers and falls into a local optimum;the clustering effect is poor for brain MR images with unclear boundaries and noise effects. To address these problems, a soft subspace clustering algorithm for brain MR images based on genetic algorithm optimization is proposed, which combines the generalized noise technique, relaxes the equational weight constraint in the objective function as the boundary constraint, and uses a genetic algorithm as a method to optimize the initialized clustering center. The genetic algorithm finds the best clustering center and reduces the algorithm's dependence on the initial clustering center. The experiment verifies the robustness of the algorithm, as well as the noise immunity in various ways and shows good results on the common dataset and the brain MR images provided by the Changshu First People's Hospital with specific high accuracy for clinical medicine.

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.

Multilevel Thresholding of Grayscale Complex Crop Images using Minimum Cross Entropy

Crop image segmentation plays a key step in the field of agriculture. The crop images present near the environs have complex backgrounds and their grayscale histogram is mostly multimodal. Hence, multilevel segmentation of grayscale crop images may be helpful for better analysis. This paper proposed multilevel thresholding of grayscale crop images incorporated with minimum cross entropy as an objective function. The time complexity of this technique increases with the threshold levels. Hence, the coronavirus herd immunity optimizer (CHIO) has been amalgamated with the objective function. This technique improves the image's accuracy. The CHIO is a humanbased algorithm that separates the foreground and background efficiently with multiple thresholds value. The simulation has been performed on grayscale crop images. It is. compared with bacterial foraging algorithm (BFO), and beta differential algorithm (BDE) to validate the accuracy. The results validates that the proposed method outperforms BFO and BDE for grayscale crop images in terms of fidelity parameters. The qualitative and quantitative results evidence the proficiency of suggested method. © 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.

Framework for biometric iris recognition in video, by deep learning and quality assessment of the iris-pupil region

In the current world scenario the influence of the COVID19 pandemic has reached universal proportions affecting almost all countries. In this sense, the need has arisen to wear gloves or to reduce direct contact with objects (such as sensors for capturing fingerprints or palm prints) as a sanitary measure to protect against the virus. In this new reality, it is necessary to have a biometric identification method that allows safe and rapid recognition of people at borders, or in quarantine controls, or in access to places of high biological risk, among others. In this scenario, iris biometric recognition has reached increasing relevance. This biometric modality avoids all the aforementioned inconveniences with proven high efficiency. However, there are still problems associated with the iris capturing and segmentation in real time that could affect the effectiveness of a System of this nature and that it is necessary to take into account. This work presents a framework for real time iris detection and segmentation in video as part of a biometric recognition system. Our proposal focuses on the stages of image capture, iris detection and segmentation in RGB video frames under controlled conditions (conditions of border and access controls, where people collaborate in the recognition process). The proposed framework is based on the direct detection of the iris-pupil region using the YOLO network, the evaluation of its quality and the semantic segmentation of iris by a Fully Convolutional Network. (FCN). The proposal of an evaluation step of the quality of the iris-pupil region reduce the passage to the system of images with problems of out of focus, blurring, occlusions, light changing and pose of the subject. For the evaluation of image quality, we propose a measure that combines parameters defined in ISO/IEC 19794-6 2005 and others derived from the systematization of the knowledge of the specialized literature. The experiments carried out in four different reference databases and an own video data set demonstrates the feasibility of its application under controlled conditions of border and access controls. The achieved results exceed or equal state-of-the-art methods under these working conditions.