Search

• Global Context Feature Extraction (GCFE) module is proposed, which can extract multi-scale global contextual information. • A Dual Scale Feature Aggregation (DSFA) module is proposed to fuse the features from different scales. • A learnable Gabor convolution layer is designed to enhance its ability to capture the edge and contour information of object. • A Spatial Attention (SA) mechanism is designed to further enhance the spatial structure information of object. Prohibited Object Detection (POD) in X-ray images plays an important role in protecting public safety. Automatic and accurate POD is required to relieve the working pressure of security inspectors. However, the existing methods cannot obtain a satisfactory detection accuracy, and especially, the problem of object occlusion also has not been solved well. Therefore, in this paper, according to the specific characteristics of X-ray images as well as low-level and high-level features of Convolutional Neural Network (CNN), different feature enhancement strategies have been elaborately designed for occluded POD. First, a learnable Gabor convolutional layer is designed and embedded into the low layer of the network to enhance the network's capability to capture the edge and contour information of object. A Spatial Attention (SA) mechanism is then designed to weight the output features of the Gabor convolutional layer to enhance the spatial structure information of object and suppress the background noises simultaneously. For the high-level features, Global Context Feature Extraction (GCFE) module is proposed to extract multi-scale global contextual information of object. And, a Dual Scale Feature Aggregation (DSFA) module is proposed to fuse these global features with those of another layer. To verify the effectiveness of the proposed modules, they are embedded into typical one-stage and two-stage object detection frameworks, i.e., Faster R-CNN and YOLO v5L, obtaining POD-F and POD-Y methods, respectively. The proposed methods have been extensively evaluated on three publicly available benchmark datasets, namely SIXray, OPIXray and WIXray. The experimental results show that, compared with existing methods, the proposed POD-Y method can achieve a state-of-the-art detection accuracy. And POD-F can also achieve a competitive detection performance among the two-stage detection methods. 1 1 https://github.com/Open-my-paper-code/POD. [ABSTRACT FROM AUTHOR]

• We analyze testing protocols used for Covid-19 detection used in the literature. • We successfully classify COVID images whose lungs areas are turned to black. • We try using different datasets for training and testing. • We show that many highly cited paper on this topic use biased testing protocols. In this paper, we compare and evaluate different testing protocols used for automatic COVID-19 diagnosis from X-Ray images in the recent literature. We show that similar results can be obtained using X-Ray images that do not contain most of the lungs. We are able to remove the lungs from the images by turning to black the center of the X-Ray scan and training our classifiers only on the outer part of the images. Hence, we deduce that several testing protocols for the recognition are not fair and that the neural networks are learning patterns in the dataset that are not correlated to the presence of COVID-19. Finally, we show that creating a fair testing protocol is a challenging task, and we provide a method to measure how fair a specific testing protocol is. In the future research we suggest to check the fairness of a testing protocol using our tools and we encourage researchers to look for better techniques than the ones that we propose. [ABSTRACT FROM AUTHOR]

[Display omitted] • We structure the GDN and LRN with translational invariance as the primary consideration. • We implement the multi-task strategy to improve the GDN's generalization. • The proposed method has robust translational invariance. Recently, intelligent cephalometric landmark detection systems that employ deep convolutional neural networks (DCNNs) have been widely developed. In this paper, we concentrate on two challenging difficulties of DCNNs based cephalometric landmark detection. First, previous DCNN methods usually decreased the resolution of high-resolution cephalometric images. Hence, the trained DCNNs missed the detailed local features of original images. Moreover, the detected landmarks' locations were recovered using a fixed ratio, so location error was increased after recovery. Second, the performance of cephalometric landmark detection is dependent on DCNNs' translational invariance because the target landmarks' locations are influenced by any shift of the cephalometric images. However, modern DCNNs have limited translational invariance, and previous DCNN methods did not take translational invariance into consideration. In this paper, we developed the widely used two-stage framework to resolve the above two challenges. In the first stage, we train a global detection network, which receives the whole images as input, to generate candidate landmarks. In the second stage, we split the input image into patches and train a local refine network, which receives image patches as input, to refine the landmarks' locations. The proposed local refine network can produce accurate landmarks' locations because it entirely focuses on the detailed local features of high-resolution cephalometric images. The global detection network and local refine network are structured with translational invariance as their primary consideration. The proposed method achieved robust performance on a private dataset and the public Automatic Cephalometric X-ray Landmark Detection Challenge 2015 dataset. [ABSTRACT FROM AUTHOR]

• A light weight CORONA-NET for diagnosing COVID-19 from chest X-ray images. • Deep feature extraction is performed by CNN, reduced yet strengthened by DWT. • The dataset included 3000 X-rays obtained locally through hospital in Denmark. • The method achieves excellent performance in comparison with the existing methods. • The proposed system has a 99.56 percent accuracy and 99.667 percent specificity. Since December 2019, COVID-19 has posed the most serious threat to living beings. With the advancement of vaccination programs around the globe, the need to quickly diagnose COVID-19 in general with little logistics is fore important. As a consequence, the fastest diagnostic option to stop COVID-19 from spreading, especially among senior patients, should be the development of an automated detection system. This study aims to provide a lightweight deep learning method that incorporates a convolutional neural network (CNN), discrete wavelet transform (DWT), and a long short-term memory (LSTM), called CORONA-NET for diagnosing COVID-19 from chest X-ray images. In this system, deep feature extraction is performed by CNN, the feature vector is reduced yet strengthened by DWT, and the extracted feature is detected by LSTM for prediction. The dataset included 3000 X-rays, 1000 of which were COVID-19 obtained locally. Within minutes of the test, the proposed test platform's prototype can accurately detect COVID-19 patients. The proposed method achieves state-of-the-art performance in comparison with the existing deep learning methods. We hope that the suggested method will hasten clinical diagnosis and may be used for patients in remote areas where clinical labs are not easily accessible due to a lack of resources, location, or other factors. [ABSTRACT FROM AUTHOR]

Hospital patients can have catheters and lines inserted during the course of their admission to give medicines for the treatment of medical issues, especially the central venous catheter (CVC). However, malposition of CVC will lead to many complications, even death. Clinicians always detect the malposition based on position detection of CVC tip via X-ray images. To reduce the workload of the clinicians and the percentage of malposition occurrence, we propose an automatic catheter tip detection framework based on a convolutional neural network (CNN). The proposed framework contains three essential components which are modified HRNet, segmentation supervision module, and deconvolution module. The modified HRNet can retain high-resolution features from start to end, ensuring the maintenance of precise information from the X-ray images. The segmentation supervision module can alleviate the presence of other line-like structures such as the skeleton as well as other tubes and catheters used for treatment. In addition, the deconvolution module can further increase the feature resolution on the top of the highest-resolution feature maps in the modified HRNet to get a higher-resolution heatmap of the catheter tip. A public CVC Dataset is utilized to evaluate the performance of the proposed framework. The results show that the proposed algorithm offering a mean Pixel Error of 4.11 outperforms three comparative methods (Ma's method, SRPE method, and LCM method). It is demonstrated to be a promising solution to precisely detect the tip position of the catheter in X-ray images. • An automatic catheter tip detection framework based on X-ray images is proposed. • The high-resolution features are maintained from start to end for tip detection. • The intermediate outputs show the catheter and the potential positions for the tip. • The framework shows the potential of machine Explainability. [ABSTRACT FROM AUTHOR]

COVID-19 quickly became a global pandemic after only four months of its first detection. It is crucial to detect this disease as soon as possible to decrease its spread. The use of chest X-ray (CXR) images became an effective screening strategy, complementary to the reverse transcription-polymerase chain reaction (RT-PCR). Convolutional neural networks (CNNs) are often used for automatic image classification and they can be very useful in CXR diagnostics. In this paper, 21 different CNN architectures are tested and compared in the task of identifying COVID-19 in CXR images. They were applied to the COVIDx8B dataset, a large COVID-19 dataset with 16,352 CXR images coming from patients of at least 51 countries. Ensembles of CNNs were also employed and they showed better efficacy than individual instances. The best individual CNN instance results were achieved by DenseNet169, with an accuracy of 98.15% and an F1 score of 98.12%. These were further increased to 99.25% and 99.24%, respectively, through an ensemble with five instances of DenseNet169. These results are higher than those obtained in recent works using the same dataset. • This paper compares 21 CNN architectures for COVID-19 detection in CXR images. • DenseNet169 achieved an accuracy of 98.15% and an F1 score of 98.12%. • An ensemble of DenseNet169 instances increased the F1 score to 99.24%. • The training was repeated five times for each model to get more reliable results. [ABSTRACT FROM AUTHOR]

• Hierarchical Bezier-based Learning improves residual networks to detect stenosis. • A high-quality image patch generator for X-ray coronary angiography analysis. • The generative model reaches a Fréchet Inception Distance of up to 84.0886. • Self-attention mechanisms maximize neural network performance up to 7%. • Notable stenosis detection accuracy, precision, F1-score, and specificity of 0.8934, 0.9031, 0.8880, 0.9111, respectively, were reached. [Display omitted] Background and Objective: Automatic detection of stenosis on X-ray Coronary Angiography (XCA) images may help diagnose early coronary artery disease. Stenosis is manifested by a buildup of plaque in the arteries, decreasing the blood flow to the heart, increasing the risk of a heart attack. Convolutional Neural Networks (CNNs) have been successfully applied to identify pathological, regular, and featured tissues on rich and diverse medical image datasets. Nevertheless, CNNs find operative and performing limitations while working with small and poorly diversified databases. Transfer learning from large natural image datasets (such as ImageNet) has become a de-facto method to improve neural networks performance in the medical image domain. Methods: This paper proposes a novel Hierarchical Bezier-based Generative Model (HBGM) to improve the CNNs training process to detect stenosis. Herein, artificial image patches are generated to enlarge the original database, speeding up network convergence. The artificial dataset consists of 10,000 images containing 50% stenosis and 50% non-stenosis cases. Besides, a reliable Fréchet Inception Distance (FID) is used to evaluate the generated data quantitatively. Therefore, by using the proposed framework, the network is pre-trained with the artificial datasets and subsequently fine-tuned using the real XCA training dataset. The real dataset consists of 250 XCA image patches, selecting 125 images for stenosis and the remainder for non-stenosis cases. Furthermore, a Convolutional Block Attention Module (CBAM) was included in the network architecture as a self-attention mechanism to improve the efficiency of the network. Results: The results showed that the pre-trained networks using the proposed generative model outperformed the results concerning training from scratch. Particularly, an accuracy, precision, sensitivity, and F1-score of 0.8934, 0.9031, 0.8746, 0.8880, 0.9111, respectively, were achieved. The generated artificial dataset obtains a mean FID of 84.0886, with more realistic visual XCA images. Conclusions: Different ResNet architectures for stenosis detection have been evaluated, including attention modules into the network. Numerical results demonstrated that by using the HBGM is obtained a higher performance than training from scratch, even outperforming the ImageNet pre-trained models. [ABSTRACT FROM AUTHOR]

Despite advances in Deep Learning, the Convolutional Neural Networks methods still manifest limitations in medical applications because datasets are usually restricted in the number of samples or include poorly contrasted images. Such a case is found in stenosis detection using X-rays coronary angiography. In this study, the emerging field of quantum computing is applied in the context of hybrid neural networks. So, a hybrid transfer-learning paradigm is used for stenosis detection, where a quantum network drives and improves the performance of a pre-trained classical network. An intermediate layer between the classical and quantum network post-processes the classical features by mapping them into a hypersphere of fixed radius through a hyperbolic tangent function. Next, these normalized features are processed in the quantum network, and through a SoftMax function, the class probabilities are obtained: stenosis and non-stenosis. Furthermore, a distributed variational quantum circuit is implemented to split the data into multiple quantum circuits within the quantum network, improving the training time without compromising the stenosis detection performance. The proposed method is evaluated on a small X-ray coronary angiography dataset containing 250 image patches (50%–50% of positive and negative stenosis cases). The hybrid classical–quantum network significantly outperformed the classical network. Evaluation results showed a boost concerning the classical transfer learning paradigm in the accuracy of 9%, recall of 20%, and F 1 -score of 11%, reaching 91.8033%, 94.9153%, and 91.8033%, respectively. • A quantum network boosts the performance of classical neural architecture. • An L 2 hyperbolic tangent layer bounds the features between classical and quantum stages. • The X-ray angiography images are analyzed to develop a robust stenosis detector system. • Transfer learning and quantum network substantially improved stenosis detection. • An efficient hybrid classical–quantum architecture is focused on stenosis detection. [ABSTRACT FROM AUTHOR]

• A deep learning method is proposed using different optimizers and transfer learning to classify Pneumonia patients. • Preparing a dataset of around 5300 X-ray images for pneumonia detection. • The proposed deep transfer learning method is trained on a benchmark open dataset of chest x-ray images. • Presenting the optimization results, precision, recall, accuracy, and F1-score for proposed method. • In proposed method achieve better accuracy in the detection rate than other techniques. Pneumonia is a disease that leads to the death of individuals within a short period since the flow of fluid in the lungs. Hence, initial diagnosis and drugs are very important to avoid the progress of the disease. This paper proposes a novel deep learning approach for automatic detection of pneumonia using deep transfer learning to simplify the detection process with improved accuracy. This work was aimed to preprocess the input chest X-ray images to identify the presence of pneumonia using U-Net architecture based segmentation and classifies the pneumonia as normal and abnormal (Bacteria, viral) using pre-trained on ImageNet dataset models such as ResNet50, InceptionV3, InceptionResNetV2. Besides, to extract the efficient features and improve accuracy of pre-trained models two optimizers, namely, Adam and Stochastic Gradient Descent (SGD) used and its performances are analyzed with batch sizes of 16 and 32. Based on the values obtained, the performances of undertaken pre-trained models are analyzed and compared with other Convolutional Neural Network (CNN) models such as DenseNet-169+SVM, VGG16, RetinaNet + Mask RCNN, VGG16 and Xception, Fully connected RCNN, etc using various measures. From the results observed that the proposed ResNet50 model work achieved 93.06% accuracy, 88.97 % precision rate, 96.78% Recall rate and 92.71% F1-score rate, which than is higher than the other models aforementioned. [ABSTRACT FROM AUTHOR]

X-ray flaw detection is a key link in the detection of internal defects in titanium alloy castings which are used for most important components in aeroengines. However, the existing manual defect detection methods from the X-ray images have common drawbacks such as unstable artificial recognition, misdetection, misjudgment, fails of quantitative analysis, huge workload, and low-quality inspection efficiency. To avoid these drawbacks, this paper proposes a new artificial intelligent (AI) method to detect and recognize the aerospace titanium casting defects from the X-ray images. It includes the target defect positioning method named as filtered selective search algorithm (FSS) and the defect classification method named as evenly distributed convolutional neural network (ED-CNN). In the target positioning step, through statistical analysis of defect characteristics, a filtered selective search algorithm is built with two filters (size and edge curvature). In this way, the FSS algorithm can position the defects with almost 100 % of accuracy, hence avoid missed detection and false detection. In the target classification step, an ED-CNN is constructed with a similar structure of the same number of layers in each feature extraction stage, and its entire architecture is evenly distributed. Compared with other three classic high-performance convolutional neural network models (AlexNet, VGG16 and VGG19), the ED-CNN model has the best performance. The ED-CNN model was tested with 324 targets from 50 original images, a classification accuracy of nearly 90 % was obtained for low density holes, porosity, linear defects, high density inclusions and casting structure. The FSS/ED-CNN method of two phases defect detection proposed in this paper can achieve accurate positioning and high accurate classification of typical defect targets, and is expected to solve the common drawbacks of "manual defect detection". The newly-proposed FSS/ED-CNN method has important research significance and engineering value. [ABSTRACT FROM AUTHOR]

• A neural network is proposed for an edge detection model of noised X-ray images. • An image processing filter is embedded in the model for advanced detection. • X-ray image datasets with three levels of noise are built for training and test. • Defect image datasets are built in experiment stage for non-destructive testing. • The developed model presents superior noise resistance for defect detection. X-ray non-destructive testing (NDT) is a primary detection technology in industrial fields, providing an effective detection for fragile and complex structures without destructing components. In this study, we adopt the principle of convolutional neural network (CNN) and a Laplacian filter to propose an edge detection model with improved performance. By constructing X-ray image datasets with different noise levels, our proposed CNN model successfully detects fuzzy defects on noised X-ray images, and presents better structure similarity of the detected information compared to conventional edge detection algorithms, Canny and SUSAN. Additionally, the experiment results indicate that the noised training datasets effectively improves the model's capability of noise resistance in edge detection tasks. Furthermore, the quality of training images significantly affects the performance of the trained model. This study develops a robust edge detection algorithm for low-cost and noise-independent X-ray non-destructive testing technology, providing a meaningful reference in edge detection of industrial X-ray images. [ABSTRACT FROM AUTHOR]

Aluminum alloy castings have a high utilization rate in the automotive industry, and its quality directly affects the safety performance of the mechanical components. Hence, casting quality management is vital during the casting production process. This paper presents a weakly-supervised Convolutional Neural Network model to recognize defects based on casting X-ray images. These images are divided into two classes including defective and non-defective. Firstly, attention maps are generated to represent the defective parts by weakly-supervised learning for each image. Then mutual-channel loss combined with the cross-entropy loss function encourage the network to focus on discriminative features. Simultaneously, a novel data-augmentation methods guided by these attention maps is proposed to enlarge the dataset. The test accuracy achieves 95.5%, and the recall is 96.0%, which means our model is accurate and robust. The efficiency of the proposed approach is verified by comparing the state-of-art approaches and the ablation experiments. • A weakly supervised convolutional neural network for classification is proposed. • The defective castings with the X-ray casting images are classified automatically. • The attention-guided data augmentation enlarges the dataset effectively. • The mutual-channel loss focuses on subtle discriminative details in two classes. [ABSTRACT FROM AUTHOR]

• Deep learning-based pneumonia detection in x-ray images is done in this work. • Different models of deep learning and transfer learning are analysed in this work for the image classification application. • An extensive analysis is carried out in this work with several experimental results. A large number of children die due to pneumonia every year worldwide. An estimated 1.2 million episodes of pneumonia were reported in children up to 5 years of age, of which 880,000 died in 2016. Hence, pneumonia is a major cause of death amongst children, with high prevalence rate in South Asia and Sub-Saharan Africa. Even in a developed country like the United States, pneumonia is among the top 10 causes of deaths. Early detection and treatment of pneumonia can reduce mortality rates among children significantly in countries having a high prevalence. Hence, this paper presents Convolutional Neural Network models to detect pneumonia using x-ray images. Several Convolutional Neural Networks were trained to classify x-ray images into two classes viz., pneumonia and non-pneumonia, by changing various parameters, hyperparameters and number of convolutional layers. Six models have been mentioned in the paper. First and second models consist of two and three convolutional layers, respectively. The other four models are pre-trained models, which are VGG16, VGG19, ResNet50, and Inception-v3. The first and second models achieve a validation accuracy of 85.26% and 92.31% respectively. The accuracy of VGG16, VGG19, ResNet50 and Inception-v3 are 87.28%, 88.46%, 77.56% and 70.99% respectively. [ABSTRACT FROM AUTHOR]