Search

This paper studies a tensor factor model that augments samples from multiple classes. The nuisance common patterns shared across classes are characterised by pervasive noises, and the patterns that distinguish different classes are represented by class‐specific components. Additionally, the pervasive component is modelled by the production of a low‐rank tensor latent factor and several factor loading matrices. This augmented tensor factor model can be expanded to a series of matrix variate tensor factor models and estimated using principal component analysis. The ranks of latent factors are estimated using a modified eigen‐ratio method. The proposed estimators have fast convergence rates and enjoy the blessing of dimensionality. The proposed factor model is applied to address the challenge of overlapping issues in image classification through a factor adjustment procedure. The procedure is shown to be powerful through synthetic experiments and an application to COVID‐19 pneumonia diagnosis from frontal chest X‐ray images. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Medical Imaging & Radiation Sciences is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this study, we present a novel approach for enhancing chest X-ray image classification (normal, Covid-19, edema, mass nodules, and pneumothorax) by combining contrastive learning and machine learning algorithms. A vast amount of unlabeled data was leveraged to learn representations so that data efficiency is improved as a means of addressing the limited availability of labeled data in X-ray images. Our approach involves training classification algorithms using the extracted features from a linear fine-tuned Momentum Contrast (MoCo) model. The MoCo architecture with a Resnet34, Resnet50, or Resnet101 backbone is trained to learn features from unlabeled data. Instead of only fine-tuning the linear classifier layer on the MoCopretrained model, we propose training nonlinear classifiers as substitutes for softmax in deep networks. The empirical results show that while the linear fine-tuned ImageNet-pretrained models achieved the highest accuracy of only 82.9% and the linear fine-tuned MoCo-pretrained models an increased highest accuracy of 84.8%, our proposed method offered a significant improvement and achieved the highest accuracy of 87.9%. [ABSTRACT FROM AUTHOR]

Pneumonia is a chest disease that occurs as a result of inflammation of the lung tissue. Although pneumonia can occur at any age, it is quite dangerous in people under the age of two and over the age of sixty-five. According to World Health Organization data, approximately 7% of all deaths in the world are due to pneumonia. Early diagnosis and treatment of the disease is an important factor in reducing mortality rates due to the disease. In the study, an effective convolutional neural network (ESA) model was proposed for pneumonia detection from three-dimensional (3D) chest x-ray images. The proposed model is designed using transfer learning approach with pre-trained ResNet. The performance of the model has now been increased by skipping block connections and some layers in the deep learning architecture. An open access dataset consisting of 5617 X-Ray images, labeled pneumonia and normal, was used in the analysis. The performance of the proposed method is compared with three different models. The first model is a simple ESA model. The second model is an ESA model in which some blocks are removed from the proposed model. The third model represents a widely used pre-trained network known as ResNet-18. According to the analysis, the accuracy, specificity, sensitivity, precision and F-1 score values of the proposed method are 98.42%, respectively; 97.52%; 99.35%; It was obtained as 97.47% and 98.90%. When the results obtained from the analyzes are examined, it is revealed that the proposed method is successful in detecting pneumonia from chest X-ray images. [ABSTRACT FROM AUTHOR]

Artificial Intelligence (AI) technology has attracted tremendous interest in the medical community, from image analysis to lesion diagnosis. However, progress in medical AI is hampered by a lack of available medical image datasets and labor‐intensive labeling processes. Here, it is demonstrated that a large number of annotated, realistic chest X‐ray images can be generated using a state‐of‐the‐art generative adversarial network (GAN) that exploits noise produced by stochastic in‐memory computing of memristor crossbar arrays. Memristors based on polymer film with high thermal resistance can increase the stochasticity of the tunneling distance for randomly ruptured conductive filaments via excessive Joule heating, thus generating true random numbers required for creating naturally diverse images in GAN. Using StyleGAN2‐adaptive discriminator augmentation (ADA), high‐quality chest X‐ray images with and without pneumothorax are successfully augmented while maintaining a good Frechet inception distance score. The results provide a cost‐effective solution for preparing privacy‐sensitive medical images and labeling to develop innovative medical AI algorithms. [ABSTRACT FROM AUTHOR]

Medical professionals in thoracic medicine routinely analyze chest X-ray images, often comparing pairs of images taken at different times to detect lesions or anomalies in patients. This research aims to design a computer-aided diagnosis system that enhances the efficiency of thoracic physicians in comparing and diagnosing X-ray images, ultimately reducing misjudgments. The proposed system encompasses four key components: segmentation, alignment, comparison, and classification of lung X-ray images. Utilizing a public NIH Chest X-ray14 dataset and a local dataset gathered by the Chiayi Christian Hospital in Taiwan, the efficacy of both the traditional methods and deep-learning methods were compared. Experimental results indicate that, in both the segmentation and alignment stages, the deep-learning method outperforms the traditional method, achieving higher average IoU, detection rates, and significantly reduced processing time. In the comparison stage, we designed nonlinear transfer functions to highlight the differences between pre- and post-images through heat maps. In the classification stage, single-input and dual-input network architectures were proposed. The inclusion of difference information in single-input networks enhances AUC by approximately 1%, and dual-input networks achieve a 1.2–1.4% AUC increase, underscoring the importance of difference images in lung disease identification and classification based on chest X-ray images. While the proposed system is still in its early stages and far from clinical application, the results demonstrate potential steps forward in the development of a comprehensive computer-aided diagnostic system for comparative analysis of chest X-ray images. [ABSTRACT FROM AUTHOR]

Pneumonia is a common and challenging disease to treat. Diagnosis of pneumonia is performed by analysing chest X-ray images with a specialist doctor today. This situation can create an excessive workload for doctors and prolong the diagnosis time. Performing early and accurate diagnosis of pneumonia using pre-trained deep learning models, which are a subcategory of the deep learning method, can be extremely beneficial. Using computeraided diagnosis systems increases the accuracy of pneumonia diagnosis and thanks to these systems, doctors have an idea about the disease before diagnosis. In this study chest X-Ray images were classified as healthy or pneumonia using pre-trained deep learning methods. The histogram equalisation image processing method was used to improve image quality and the mask region-based convolutional neural network pre-trained method was used to segment the chest region. Alexnet, ResNet18 and VGG16 pre-trained models were used for image classification as healthy and pneumonia. ResNet18 showed outstanding performance in this study. According to the performance metrics of accuracy (0.983), recall (0.994) and F1-score (0.987), success rates were achieved by using the ResNet18 model. This study has shown that deep learning models can achieve high success rates in pneumonia diagnosis. [ABSTRACT FROM AUTHOR]

The amount of labeled samples always plays a significant role in deep learning. When there are insufficient training samples, deep learning usually struggles to carry out its task successfully. The problem becomes more prominent in that labeled data are expensive to obtain. Therefore, we want to address the problem by introducing a method that utilizes unlabeled samples to improve object detection performance. We propose a framework that utilizes a pseudo‐labeling approach in semi‐supervised learning and applies it to disease diagnosis of chest X‐ray images. Our objective is to train a model that works as a labeling expert and perform the labeling task for improving the object (disease area) detection model by these newly labeled samples. In addition, we also apply an ensemble technique to our framework to reduce the model bias and improve the generalization ability for pseudo‐labeling, which results in improvements in object detection performance. Our proposed method improves the mean average precision up to 5.32, compared with the method without the pseudo‐labeling. Additionally, the mean average precision further increases by up to 19.11 when applying ensemble learning. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Pneumonia and tuberculosis are the major public health problems worldwide. These diseases affect the lungs, and if they are not diagnosed properly in time, they can become a fatal health problem. Chest x-ray images are widely used to detect and diagnose Pneumonia and Tuberculosis disease. Detection of Pneumonia and Tuberculosis from chest x-ray images is difficult and requires experience due to the similar pathological features of the diseases. Sometimes a misdiagnosis of the disease occurs due to this similarity. Several researchers used deep learning and machine learning techniques to solve this misdiagnosis problem. However, these studies used the chest x-ray images only to develop Pneumonia and Tuberculosis disease detection models. But using the chest x-ray images alone cannot necessarily lead to accurate disease detection and classification. In the traditional or manual approach, medical records are required to support and correctly interpret the chest x-ray images in the appropriate clinical context. This study develops a multi-input Pneumonia and Tuberculosis detection model using chest x-ray images and medical records to follow the clinical procedure. The study applied a Convolutional Neural Network for the chest x-ray image data and a Multilayer perceptron for the medical record data to develop the models. We implemented feature-level concatenation to join the output feature vectors from the Convolutional Neural Network and a Multilayer perceptron for the development of the disease detection model. For the purpose of comparison, we also developed image-only and medical record-only models. Consequently, the image-only model gives an accuracy of 92.68%, the medical record-only model results in 98.72% accuracy, and the combined model accuracy is improved to 99.61%. In general, the study shows that the fusion of the chest x-ray and the medical records leads to better accuracy and is more similar to the clinical approach. [ABSTRACT FROM AUTHOR]

Biological fingerprints extracted from clinical images can be used for patient identity verification to determine misfiled clinical images in picture archiving and communication systems. However, such methods have not been incorporated into clinical use, and their performance can degrade with variability in the clinical images. Deep learning can be used to improve the performance of these methods. A novel method is proposed to automatically identify individuals among examined patients using posteroanterior (PA) and anteroposterior (AP) chest X-ray images. The proposed method uses deep metric learning based on a deep convolutional neural network (DCNN) to overcome the extreme classification requirements for patient validation and identification. It was trained on the NIH chest X-ray dataset (ChestX-ray8) in three steps: preprocessing, DCNN feature extraction with an EfficientNetV2-S backbone, and classification with deep metric learning. The proposed method was evaluated using two public datasets and two clinical chest X-ray image datasets containing data from patients undergoing screening and hospital care. A 1280-dimensional feature extractor pretrained for 300 epochs performed the best with an area under the receiver operating characteristic curve of 0.9894, an equal error rate of 0.0269, and a top-1 accuracy of 0.839 on the PadChest dataset containing both PA and AP view positions. The findings of this study provide considerable insights into the development of automated patient identification to reduce the possibility of medical malpractice due to human errors. [ABSTRACT FROM AUTHOR]

Image recognition techniques can recognize abnormal medical images, directly contributing to achieving high-quality diagnosis. In particular, with the development of deep learning technology, the recognition accuracy of abnormalities has steadily increased. However, there are usually no adequate learning instances for chest X-ray images, which directly leads to the failure of high-quality recognition. To solve this problem, we proposed a multi-weight-based limited learning instance model for chest X-ray image recognition. First, an optimized saliency detection model directly deleted the unsatisfactory learning instances, especially for learning instances without obvious significance. Second, multi-scale decomposition and sparse representation were combined to calculate the weights of different learning instances. Third, a multi-weight-based cost function was constructed to achieve high-quality recognition results by considering learning instances from multiple cases. Finally, according to the experimental database, we carried out experiments in which our method could achieve satisfactory recognition accuracy while using limited learning instances. More importantly, the economic value of this method cannot be underestimated considering that modern technology has become an important way to promote economic development. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has underscored the urgent need for rapid and accurate diagnosis facilitated by artificial intelligence (AI), particularly in computer-aided diagnosis using medical imaging. However, this context presents two notable challenges: high diagnostic accuracy demand and limited availability of medical data for training AI models. To address these issues, we proposed the implementation of a Masked AutoEncoder (MAE), an innovative self-supervised learning approach, for classifying 2D Chest X-ray images. Our approach involved performing imaging reconstruction using a Vision Transformer (ViT) model as the feature encoder, paired with a custom-defined decoder. Additionally, we fine-tuned the pretrained ViT encoder using a labeled medical dataset, serving as the backbone. To evaluate our approach, we conducted a comparative analysis of three distinct training methods: training from scratch, transfer learning, and MAE-based training, all employing COVID-19 chest X-ray images. The results demonstrate that MAE-based training produces superior performance, achieving an accuracy of 0.985 and an AUC of 0.9957. We explored the mask ratio influence on MAE and found ratio = 0.4 shows the best performance. Furthermore, we illustrate that MAE exhibits remarkable efficiency when applied to labeled data, delivering comparable performance to utilizing only 30% of the original training dataset. Overall, our findings highlight the significant performance enhancement achieved by using MAE, particularly when working with limited datasets. This approach holds profound implications for future disease diagnosis, especially in scenarios where imaging information is scarce. [ABSTRACT FROM AUTHOR]

In order to solve the problem that the existing cross-modal methods of chest X-ray images and diagnostic reports focus on global information alignment, which ignore the fine-grained semantic association between chest X-ray images and diagnostic reports, resulting in low retrieval accuracy and poor matching degree, this paper proposed a method named CDTCR (chest X-ray images and diagnostic reports for twin-towers cross-modal retrieval) Specifically, for fine-grained semantic representation, it proposed an image encoder composed of residual network to learn the fine-grained features of the image and a BERT model composed of Transformer to learn the fine-grained semantic features of the diagnostic report. In order to solve the problem of fine-grained semantic association, it also designed an information alignment strategy of two different granularity modes for the global image to sentence and local region to phrase, which solved the problem of insufficient fine-grained semantic association between different modes. The experimental results on a large-scale medical dataset MIMIC-CXR show that CDTCR has higher retrieval accuracy and better interpretation than the existing cross-modal retrieval methods. [ABSTRACT FROM AUTHOR]

To identify COVID-19 chest X-Ray (CXR) images accurately and quickly, a lightweight model WCF-MobileNetV3 based on weighted channel filter (WCF) is proposed. The model uses MobileNetV3-small as the backbone network. Aiming at the problem that the difference between CXR image samples is small and it is difficult to extract distinguishing features, the WCF module is proposed. First, it extracts the weights of the high-dimensional and low-dimensional channel features from the input feature map. And then, it generates the high-dimensional and low-dimensional feature channel masks by weighted random sampling and fuses the weights. Next, it uses the masks to filter the fused weights. Finally, it assigns the weights to the input feature map to achieve channel-wise feature enhancement. The experimental results on the Chest-X-Ray Image dataset and the COVID-19 Chest X-Ray Image Repository dataset show that the accuracy, precision, and sensitivity of WCF-MobileNeteV3 for COVID-19 recognition are 97.93%, 98.64%, and 97.19%, respectively. Compared with other COVID-19 identification algorithms, WCF-MobileNetV3 can identify COVID-19 CXR images accurately and efficiently with better recognition performance. [ABSTRACT FROM AUTHOR]

Abstract Background Nontuberculous mycobacterial lung disease (NTM-LD) and Mycobacterium tuberculosis lung disease (MTB-LD) have similar clinical characteristics. Therefore, NTM-LD is sometimes incorrectly diagnosed with MTB-LD and treated incorrectly. To solve these difficulties, we aimed to distinguish the two diseases in chest X-ray images using deep learning technology, which has been used in various fields recently. Methods We retrospectively collected chest X-ray images from 3314 patients infected with Mycobacterium tuberculosis (MTB) or nontuberculosis mycobacterium (NTM). After selecting the data according to the diagnostic criteria, various experiments were conducted to create the optimal deep learning model. A performance comparison was performed with the radiologist. Additionally, the model performance was verified using newly collected MTB-LD and NTM-LD patient data. Results Among the implemented deep learning models, the ensemble model combining EfficientNet B4 and ResNet 50 performed the best in the test data. Also, the ensemble model outperformed the radiologist on all evaluation metrics. In addition, the accuracy of the ensemble model was 0.85 for MTB-LD and 0.78 for NTM-LD on an additional validation dataset consisting of newly collected patients. Conclusions In previous studies, it was known that it was difficult to distinguish between MTB-LD and NTM-LD in chest X-ray images, but we have successfully distinguished the two diseases using deep learning methods. This study has the potential to aid clinical decisions if the two diseases need to be differentiated.

Medical professionals in thoracic medicine routinely analyze chest X-ray images, often comparing pairs of images taken at different times to detect lesions or anomalies in patients. This research aims to design a computer-aided diagnosis system that enhances the efficiency of thoracic physicians in comparing and diagnosing X-ray images, ultimately reducing misjudgments. The proposed system encompasses four key components: segmentation, alignment, comparison, and classification of lung X-ray images. Utilizing a public NIH Chest X-ray14 dataset and a local dataset gathered by the Chiayi Christian Hospital in Taiwan, the efficacy of both the traditional methods and deep-learning methods were compared. Experimental results indicate that, in both the segmentation and alignment stages, the deep-learning method outperforms the traditional method, achieving higher average IoU, detection rates, and significantly reduced processing time. In the comparison stage, we designed nonlinear transfer functions to highlight the differences between pre- and post-images through heat maps. In the classification stage, single-input and dual-input network architectures were proposed. The inclusion of difference information in single-input networks enhances AUC by approximately 1%, and dual-input networks achieve a 1.2–1.4% AUC increase, underscoring the importance of difference images in lung disease identification and classification based on chest X-ray images. While the proposed system is still in its early stages and far from clinical application, the results demonstrate potential steps forward in the development of a comprehensive computer-aided diagnostic system for comparative analysis of chest X-ray images.

Although the widespread use of digital imaging has enabled real-time image display, images in chest X-ray examinations can be confirmed by the radiologist's eyes. Considering the development of deep learning (DL) technology, its application will make it possible to immediately determine the need for a retake, which is expected to further improve examination throughput. In this study, we developed software for evaluating chest X-ray images to determine whether a repeat radiographic examination is necessary, based on the combined application of DL technologies, and evaluated its accuracy. The target population was 4809 chest images from a public database. Three classification models (CLMs) for lung field defects, obstacle shadows, and the location of obstacle shadows and a semantic segmentation model (SSM) for the lung field regions were developed using a fivefold cross validation. The CLM was evaluated using the overall accuracy in the confusion matrix, the SSM was evaluated using the mean intersection over union (mIoU), and the DL technology-combined software was evaluated using the total response time on this software (RT) per image for each model. The results of each CLM with respect to lung field defects, obstacle shadows, and obstacle shadow location were 89.8%, 91.7%, and 91.2%, respectively. The mIoU of the SSM was 0.920, and the software RT was 3.64 × 10−2 s. These results indicate that the software can immediately and accurately determine whether a chest image needs to be re-scanned. [ABSTRACT FROM AUTHOR]

Background: Nontuberculous mycobacterial lung disease (NTM-LD) and Mycobacterium tuberculosis lung disease (MTB-LD) have similar clinical characteristics. Therefore, NTM-LD is sometimes incorrectly diagnosed with MTB-LD and treated incorrectly. To solve these difficulties, we aimed to distinguish the two diseases in chest X-ray images using deep learning technology, which has been used in various fields recently. Methods: We retrospectively collected chest X-ray images from 3314 patients infected with Mycobacterium tuberculosis (MTB) or nontuberculosis mycobacterium (NTM). After selecting the data according to the diagnostic criteria, various experiments were conducted to create the optimal deep learning model. A performance comparison was performed with the radiologist. Additionally, the model performance was verified using newly collected MTB-LD and NTM-LD patient data. Results: Among the implemented deep learning models, the ensemble model combining EfficientNet B4 and ResNet 50 performed the best in the test data. Also, the ensemble model outperformed the radiologist on all evaluation metrics. In addition, the accuracy of the ensemble model was 0.85 for MTB-LD and 0.78 for NTM-LD on an additional validation dataset consisting of newly collected patients. Conclusions: In previous studies, it was known that it was difficult to distinguish between MTB-LD and NTM-LD in chest X-ray images, but we have successfully distinguished the two diseases using deep learning methods. This study has the potential to aid clinical decisions if the two diseases need to be differentiated. [ABSTRACT FROM AUTHOR]

In daily healthcare work routines, automatic diagnosis systems are essential. Human errors are very likely when working in those dangerous environments with a heavy workload and stress. One of the medical procedures where mistakes are risky and can result in severe complications if not caught in time is the task of positioning tubes and catheters. A type of tube is inserted for a patient as part of the tube placement procedure. The position of the installed tube is then determined by screening the patient. Waiting for a radiologist to confirm the diagnosis will delay the tube adjustment. Indeed, more complications may result from the tube adjustment delay or any potential diagnostic mistakes. Through this work, we propose a framework for diagnosis and validation for in-time tube placement error detection. The framework analyzes the chest X-ray right after the tube is inserted and generates a segmentation mask along with classification values for possible errors. Our proposed framework is founded on a customized U-net model that provides competitive segmentation results (dice coefficient of 94,5%) compared to the original U-Net model version. Moreover, the proposed framework is optimized to support deployment on production-edge mobile devices with 75% fewer training parameters. [ABSTRACT FROM AUTHOR]

Cardiothoracic diseases are a serious threat to human health and chest X-ray image is a great reference in diagnosis and treatment. At present, it has been a research hot-spot how to recognize chest X-ray image automatically and exactly by the computer vision technology, and many scholars have gotten the excited research achievements. While both emphysema and cardiomegaly often are associated, and the symptom of them are very similar, so the X-ray images discrimination for them led easily to misdiagnosis too. Therefore, some efforts are still expected to develop a higher precision and better performance deep learning model to recognize efficiently the two diseases. In this work, we construct an adaptive cross-transfer residual neural network (ACRnet) to identify emphysema, cardiomegaly and normal. We cross-transfer the information extracted by the residual block and adaptive structure to different levels in ACRnet, and the method avoids the reduction of the adaptive function by residual structure and improves the recognition performance of the model. To evaluate the recognition ability of ACRnet, four neural networks VGG16, InceptionV2, ResNet101 and CliqueNet are used for comparison. The results show that ACRnet has better recognition ability than other networks. In addition, we use the deep convolution generative adversarial network (DCGAN) to expand the original dataset and ACRnet's recognition ability is greatly improved.

Coronavirus disease, frequently referred to as COVID-19, is a contagious and transmittable disease produced by the SARS-CoV-2 virus. The only solution to tackle this virus and reduce its spread is early diagnosis. Pathogenic laboratory tests such as the polymerase chain reaction (PCR) process take a long time. Also, they regularly produce incorrect results. However, they are still considered the critical standard for detecting the virus. Hence, there is a solid need to evolve computer-assisted diagnosis systems capable of providing quick and low-cost testing in areas where traditional testing procedures are not feasible. This study focuses on COVID-19 detection using X-ray images. The prime objective is to introduce a computer-assisted diagnosis (CAD) system to differentiate COVID-19 from healthy and pneumonia cases using X-ray image sequences. This work utilizes standard transfer-learning techniques for COVID-19 detection. It proposes the master–slave architecture using the most state-of-the-art Densenet201 and Squeezenet1_0 techniques for classifying the COVID-19 virus in chest X-ray image sequences. This paper compares the proposed models with other standard transfer-learning approaches for COVID-19. The performance metrics demonstrate that the proposed approach outperforms standard transfer-learning approaches. This research also fine-tunes hyperparameters and predicts the optimized learning rate to achieve the highest accuracy in the model. After fine-tuning the learning rate, the DenseNet201 model retrieves an accuracy of 83.33%, while the fastest model is SqueezeNet1_0, which retrieves an accuracy of 80%. [ABSTRACT FROM AUTHOR]

Purpose: Many artificial intelligence-based computational procedures are developed to diagnose COVID-19 infection from chest X-ray (CXR) images, as diagnosis by CXR imaging is less time consuming and economically cheap compared to other detection procedures. Due to unavailability of skilled computer professionals and high computer architectural resource, majority of the employed methods are difficult to implement in rural and poor economic settings. Majority of such reports are devoid of codes and ignores related diseases (pneumonia). The absence of codes makes limitation in applying them widely. Hence, validation testing followed by evidence-based medical practice is difficult. The present work was aimed to develop a simple method that requires a less computational expertise and minimal level of computer resource, but with statistical inference. Materials and Methods: A Fast Fourier Transform-based (FFT) method was developed with GNU Octave, a free and open-source platform. This was employed to the images of CXR for further analysis. For statistical inference, two variables, i.e., the highest peak and number of peaks in the FFT distribution plot were considered. Results: The comparison of mean values among different groups (normal, COVID-19, viral, and bacterial pneumonia [BP]) showed statistical significance, especially when compared to normal, except between viral and BP groups. Conclusion: Parametric statistical inference from our result showed high level of significance (P < 0.001). This is comparable to the available artificial intelligence-based methods (where accuracy is about 94%). Developed method is easy, availability with codes, and requires a minimal level of computer resource and can be tested with a small sample size in different demography, and hence, be implemented in a poor socioeconomic setting.

Chest X-ray image is an important clinical diagnostic reference to lung diseases that is a serious threat to human health. At present, with the rapid development of computer vision and deep learning technology, many scholars have carried out the fruitful research on how to build a valid model for chest X-ray images recognition of lung diseases. While some efforts are still expected to improve the performance of the recognition model and enhance the interpretability of the recognition results. In this paper, we construct a multi-scale adaptive residual neural network (MARnet) to identify chest X-ray images of lung diseases. To make the model better extract image features, we cross-transfer the information extracted by residual block and the information extracted by adaptive structure to different layer, avoiding the reduction effect of residual structure on adaptive function. We compare MARnet with some classical neural networks, and the results show that MARnet achieves accuracy (ACC) of 83.3% and the area under ROC curve (AUC) of 0.97 in the identification of 4 kinds of typical lung X-ray images including nodules, atelectasis, normal and infection, which are higher than those of other methods. Moreover, to avoid the randomness of the train-test-split method, 5-fold cross-validation method is used to verify the generalization ability of the MARnet model and the results are satisfactory. Finally, the technique called Gradient-weighted Class Activation Mapping (Grad-CAM), is adopted to display significantly the discriminative regions of the images in the form of the heat map, which provides an explainable and more direct clinical diagnostic reference to lung diseases.

Early detection of pneumoconiosis by routine health screening of workers in the mining industry is critical for preventing the progression of this incurable disease. Automated pneumoconiosis classification in chest X-ray images is challenging due to the low contrast of opacities, inter-class similarity, intra-class variation and the existence of artifacts. Compared to traditional methods, convolutional neural networks have shown significant improvement in pneumoconiosis classification tasks, however, accurate classification remains challenging due to mainly the inability to focus on semantically meaningful lesion opacities. Most existing networks focus on high level abstract information and ignore low level detailed object information. Different from natural images where an object occupies large space, the classification of pneumoconiosis depends on the density of small opacities inside the lung. To address this issue, we propose a novel two-stage adaptive multi-scale feature pyramid network called AMFP-Net for the diagnosis of pneumoconiosis from chest X-rays. The proposed model consists of 1) an adaptive multi-scale context block to extract rich contextual and discriminative information and 2) a weighted feature fusion module to effectively combine low level detailed and high level global semantic information. This two-stage network first segments the lungs to focus more on relevant regions by excluding irrelevant parts of the image, and then utilises the segmented lungs to classify pneumoconiosis into different categories. Extensive experiments on public and private datasets demonstrate that the proposed approach can outperform state-of-the-art methods for both segmentation and classification. • Automated pneumoconiosis classification in chest X-ray images is challenging. • We propose an adaptive multi-scale context module, to extract rich contextual features adaptively. • We introduce weighted feature fusion to aggregate multi-level features in an effective way. • We develop a two-stage Network to capture local object-level and global image-level features and integrate the above-mentioned two methods. • Our proposed methods can be easily plugged into many existing DCNN architectures to improve the performance. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has underscored the urgent need for rapid and accurate diagnosis facilitated by artificial intelligence (AI), particularly in computer-aided diagnosis using medical imaging. However, this context presents two notable challenges: high diagnostic accuracy demand and limited availability of medical data for training AI models. To address these issues, we proposed the implementation of a Masked AutoEncoder (MAE), an innovative self-supervised learning approach, for classifying 2D Chest X-ray images. Our approach involved performing imaging reconstruction using a Vision Transformer (ViT) model as the feature encoder, paired with a custom-defined decoder. Additionally, we fine-tuned the pretrained ViT encoder using a labeled medical dataset, serving as the backbone. To evaluate our approach, we conducted a comparative analysis of three distinct training methods: training from scratch, transfer learning, and MAE-based training, all employing COVID-19 chest X-ray images. The results demonstrate that MAE-based training produces superior performance, achieving an accuracy of 0.985 and an AUC of 0.9957. We explored the mask ratio influence on MAE and found ratio = 0.4 shows the best performance. Furthermore, we illustrate that MAE exhibits remarkable efficiency when applied to labeled data, delivering comparable performance to utilizing only 30% of the original training dataset. Overall, our findings highlight the significant performance enhancement achieved by using MAE, particularly when working with limited datasets. This approach holds profound implications for future disease diagnosis, especially in scenarios where imaging information is scarce.

Purpose: Many artificial intelligence‑based computational procedures are developed to diagnose COVID‑19 infection from chest X‑ray (CXR) images, as diagnosis by CXR imaging is less time consuming and economically cheap compared to other detection procedures. Due to unavailability of skilled computer professionals and high computer architectural resource, majority of the employed methods are difficult to implement in rural and poor economic settings. Majority of such reports are devoid of codes and ignores related diseases (pneumonia). The absence of codes makes limitation in applying them widely. Hence, validation testing followed by evidence‑based medical practice is difficult. The present work was aimed to develop a simple method that requires a less computational expertise and minimal level of computer resource, but with statistical inference. Materials and Methods: A Fast Fourier Transform‑based (FFT) method was developed with GNU Octave, a free and open‑source platform. This was employed to the images of CXR for further analysis. For statistical inference, two variables, i.e., the highest peak and number of peaks in the FFT distribution plot were considered. Results: The comparison of mean values among different groups (normal, COVID‑19, viral, and bacterial pneumonia [BP]) showed statistical significance, especially when compared to normal, except between viral and BP groups. Conclusion: Parametric statistical inference from our result showed high level of significance (P < 0.001). This is comparable to the available artificial intelligence‑based methods (where accuracy is about 94%). Developed method is easy, availability with codes, and requires a minimal level of computer resource and can be tested with a small sample size in different demography, and hence, be implemented in a poor socioeconomic setting. [ABSTRACT FROM AUTHOR]

Nowadays, the global COVID-19 situation is still serious, and the new mutant virus Delta has already spread all over the world. The chest X-ray is one of the most common radiological examinations for screening catheters and diagnosis of many lung diseases, which plays an important role in assisting clinical diagnosis during the outbreak. This study considers the problem of multi-label catheters and thorax disease classification on chest X-ray images based on computer vision. Therefore, we propose a new variant of pyramid vision Transformer for multi-label chest X-ray image classification, named MXT, which can capture both short and long-range visual information through self-attention. Especially, downsampling spatial reduction attention can reduce the resource consumption of using Transformer. Meanwhile, multi-layer overlap patch (MLOP) embedding is used to tokenize images and dynamic position feed forward with zero paddings can encode position instead of adding a positional mask. Furthermore, class token Transformer block and multi-label attention (MLA) are utilized to offer more effective processing of multi-label classification. We evaluate our MXT on Chest X-ray14 dataset which has 14 disease pathologies and Catheter dataset containing 11 types of catheter placement. Each image is labeled one or more categories. Compared with some state-of-the-art baselines, our MXT can yield the highest mean AUC score of 83.0% on the Chest X-ray14 dataset and 94.6% on the Catheter dataset. According to the ablation study, we can obtain the following results: (1) The proposed MLOP embedding has a better performance than overlap patch (OP) embedding layer and non-overlap patch (N-OP) embedding layer that the mean AUC score is improved 0.6% and 0.4%, respectively. (2) Our demonstrate dynamic position feed forward can replace the traditional position mask which can learn the position information, and the mean AUC increased by 0.6%. (3) The mean AUC score by the designed MLA is more 0.2% and 0.6% than using the class token and calculating the mean scores of all tokens. The comprehensive experiments on two datasets demonstrate the effectiveness of the proposed method for multi-label chest X-ray image classification. Hence, our MXT can assist radiologists in diagnoses of lung diseases and check the placement of catheters, which can reduce the work pressure of medical staff. [ABSTRACT FROM AUTHOR]

Although the widespread use of digital imaging has enabled real-time image display, images in chest X-ray examinations can be confirmed by the radiologist’s eyes. Considering the development of deep learning (DL) technology, its application will make it possible to immediately determine the need for a retake, which is expected to further improve examination throughput. In this study, we developed software for evaluating chest X-ray images to determine whether a repeat radiographic examination is necessary, based on the combined application of DL technologies, and evaluated its accuracy. The target population was 4809 chest images from a public database. Three classification models (CLMs) for lung field defects, obstacle shadows, and the location of obstacle shadows and a semantic segmentation model (SSM) for the lung field regions were developed using a fivefold cross validation. The CLM was evaluated using the overall accuracy in the confusion matrix, the SSM was evaluated using the mean intersection over union (mIoU), and the DL technology-combined software was evaluated using the total response time on this software (RT) per image for each model. The results of each CLM with respect to lung field defects, obstacle shadows, and obstacle shadow location were 89.8%, 91.7%, and 91.2%, respectively. The mIoU of the SSM was 0.920, and the software RT was 3.64 × 10−2 s. These results indicate that the software can immediately and accurately determine whether a chest image needs to be re-scanned.

The coronavirus (COVID-19) is a major global disaster of humankind, in the 21st century. COVID-19 initiates breathing infection, including pneumonia, common cold, sneezing, and coughing. Initial detection becomes crucial, to classify the virus and limit its spread. COVID-19 infection is similar to other types of pneumonia, and it may result in severe pneumonia, with bundles of illness onsets. This research is focused on identifying people affected by COVID-19 at a very early stage, through chest X-ray images. Chest X-ray classification is a beneficial method in the identification, follow up, and evaluation of treatment efficiency, for people with pneumonia. This research, also, considered chest X-ray classification as a basic method to evaluate the existence of lung irregularities in symptomatic patients, alleged for COVID-19 disease. The aim of this research is to classify COVID-19 samples from normal chest X-ray images and pneumonia-affected chest X-ray images of people, for early identification of the disease. This research will help people in diagnosing individuals for viruses and insisting that people receive proper treatment as well as preventive action, to stop the spread of the virus. To provide accurate classification of disease in patients' chest X-ray images, this research proposed a novel classification model, named 2dCNN-BiCuDNNLSTM, which combines two-dimensional Convolutional Neural Network (CNN) and a Bidirectional CUDA Deep Neural Network Long Short-Term Memory (BiCuDNNLSTM). Deep learning is known for identifying the patterns in available data that will be helpful in accurate classification of disease. The proposed model (2dCNN and BiCuDNNLSTM layers, with proper hyperparameters) can differentiate normal chest X-rays from viral pneumonia and COVID-19 ones, with high accuracy. A total of 6863 X-ray images (JPEG) (1000 COVID-19 patients, 3863 normal cases, and 2000 pneumonia patients) have been engaged, to examine the achievement of the suggested neural network; 80% of the images dataset for every group is received for proposed model training, 10% is accepted for validation, and 10% is accepted for testing. It is observed that the proposed model acquires the towering classification accuracy of 93%. The proposed network is used for predictive analysis, to prompt people regarding the risk of early detection of COVID-19. X-ray images help to classify people with COVID-19 variants and to indicate the severity of disease in the future. This study demonstrates the effectiveness of the proposed CUDA-enabled hybrid deep learning models, to classify the X-ray image data, with a high accuracy of detecting COVID-19. It reveals that the proposed model can be applicable in numerous virus classifications. The chest X-ray classification is a commonly available and reasonable approach, for diagnosing people with lower respiratory signs or suspected COVID-19. Therefore, it is demonstrated that the proposed model has an efficient and promising accomplishment for classifying COVID-19 through X-ray images. The proposed hybrid model can, efficiently, preserve the comprehensive characteristic facts of the image data, for more exceptional concluding classification results than an individual neural network. [ABSTRACT FROM AUTHOR]

In medical image detection, the dataset often has a situation od class-imbalanced, which causes the long-tail distribution problem in the dataset samples and the performance of the detection model becomes seriously. This paper aims to solve the over-fitting problem of the network when training with multiple class-imbalance dataset, improving the original loss function by re-weight method, and preprocessing the X-ray image by CLAHE algorithm to highlight the internal details of the image. Finally, ResNext50 is selected as the feature extraction network. By using covid-chestxray dataset as the dataset, this paper evaluates the Accuracy, Precision, Recall rate and F1 value of the model. The effectiveness of this method is verified in experiments. [ABSTRACT FROM AUTHOR]

Chest X-ray radiographic (CXR) imagery enables earlier and easier lung disease diagnosis. Therefore, in this paper, we propose a deep learning method using a transfer learning technique to classify lung diseases on CXR images to improve the efficiency and accuracy of computer-aided diagnostic systems' (CADs') diagnostic performance. Our proposed method is a one-step, end-to-end learning, which means that raw CXR images are directly inputted into a deep learning model (EfficientNet v2-M) to extract their meaningful features in identifying disease categories. We experimented using our proposed method on three classes of normal, pneumonia, and pneumothorax of the U.S. National Institutes of Health (NIH) data set, and achieved validation performances of loss = 0.6933, accuracy = 82.15%, sensitivity = 81.40%, and specificity = 91.65%. We also experimented on the Cheonan Soonchunhyang University Hospital (SCH) data set on four classes of normal, pneumonia, pneumothorax, and tuberculosis, and achieved validation performances of loss = 0.7658, accuracy = 82.20%, sensitivity = 81.40%, and specificity = 94.48%; testing accuracy of normal, pneumonia, pneumothorax, and tuberculosis classes was 63.60%, 82.30%, 82.80%, and 89.90%, respectively. [ABSTRACT FROM AUTHOR]

• A novel self-supervised multi-task learning framework for medical image analysis. • An effective explicit uniformity regularizer to promote representation uniformity. • The extraction of comprehensive contextual information from anatomical structures. • Extensive experiments on chest X-ray image classification and segmentation tasks. • Visualizations of results to enhance the interpretability. Deep learning is crucial for preliminary screening and diagnostic assistance based on medical image analysis. However, limited annotated data and complex anatomical structures challenge existing models as they struggle to capture anatomical context information effectively. In response, we propose a novel self-supervised multi-task learning framework (SSMT), which integrates two key modules: a discriminative-based module and a generative-based module. These modules collaborate through multiple proxy tasks, encouraging models to learn global discriminative representations and local fine-grained representations. Additionally, we introduce an efficient uniformity regularization to further enhance the learned representations. To demonstrate the effectiveness of SSMT, we conduct extensive experiments on six public Chest X-ray image datasets. Our results highlight that SSMT not only outperforms existing state-of-the-art methods but also achieves comparable performance to the supervised model in challenging downstream tasks. The ablation study demonstrates collaboration between the key components of SSMT, showcasing its potential for advancing medical image analysis. [ABSTRACT FROM AUTHOR]

• Introduce a learnable label embedding as queries to detect and match class-related features from the feature maps. • Spatial reduction attention is employed to simplify the complexity of global self-attention. • Dual-path attention is designed to establish a compact connection between visual features and their corresponding labels of thoracic diseases. • A classification attention block is used to balance two classification scores based on feature output and label output. • Achieve the highest average per-class AUC of 83.1% on the ChestX-ray14 datasets for multi-label classification of thoracic diseases, while yielding the highest ACC of 94.3% for multi-class classification. Manually and accurately detecting thoracic diseases from CXR images is a time-consuming task that requires experienced radiologists. Therefore, automated thoracic diseases classification has great significance. However, most existing methods solely leverage the feature maps extracted from CXR images to classify thoracic diseases, without effectively connecting the correlation between the local discriminative lesion features and their corresponding labels. To address this issue, we innovatively introduce a learnable label embedding as queries to detect and match class-related features from the feature maps, and then processed by a novel Transformer-based dual-path decoder (TransDD) to facilitate interaction. The proposed TransDD is comprised of three key components: spatial reduction attention (SRA), dual-path attention (DPA), and feature enhancement module (FEM). SRA is employed in simplifying the complexity of self-attention, while DPA is specifically designed to connect the explicit correlation between the features and labels. Moreover, FEM is used to boost the expressiveness of local features. Subsequently, the classification attention block is utilized to balance two classification scores based on the feature output and label output, respectively. The proposed TransDD-PVT attained SOTA performance on the ChestX-ray14 dataset, achieving a mean area under the receiver operating characteristic (AUC) of 83.1% across all 14 classes. Also, our method achieves 94.31% accuracy and 93.31% sensitivity on three-class classifications. Extensive experiments conducted on several datasets demonstrate the powerful ability of our TransDD to improve the performance of thoracic diseases classification. It can serve as a plug-and-play structure to improve the classification performance of both CNNs and recent Transformer-based backbones. [ABSTRACT FROM AUTHOR]

Early screening of COVID-19 is essential for pandemic control, and thus to relieve stress on the health care system. Lung segmentation from chest X-ray (CXR) is a promising method for early diagnoses of pulmonary diseases. Recently, deep learning has achieved great success in supervised lung segmentation. However, how to effectively utilize the lung region in screening COVID-19 still remains a challenge due to domain shift and lack of manual pixel-level annotations. We hereby propose a multi-appearance COVID-19 screening framework by using lung region priors derived from CXR images. Firstly, we propose a multi-scale adversarial domain adaptation network (MS-AdaNet) to boost the cross-domain lung segmentation task as the prior knowledge to the classification network. Then, we construct a multi-appearance network (MA-Net), which is composed of three sub-networks to realize multi-appearance feature extraction and fusion using lung region priors. At last, we can obtain prediction results from normal, viral pneumonia, and COVID-19 using the proposed MA-Net. We extend the proposed MS-AdaNet for lung segmentation task on three different public CXR datasets. The results suggest that the MS-AdaNet outperforms contrastive methods in cross-domain lung segmentation. Moreover, experiments reveal that the proposed MA-Net achieves accuracy of 98.83 $\%$ and F1-score of 98.71 $\%$ on COVID-19 screening. The results indicate that the proposed MA-Net can obtain significant performance on COVID-19 screening. [ABSTRACT FROM AUTHOR]

Coronavirus disease, frequently referred to as COVID-19, is a contagious and transmittable disease produced by the SARS-CoV-2 virus. The only solution to tackle this virus and reduce its spread is early diagnosis. Pathogenic laboratory tests such as the polymerase chain reaction (PCR) process take a long time. Also, they regularly produce incorrect results. However, they are still considered the critical standard for detecting the virus. Hence, there is a solid need to evolve computer-assisted diagnosis systems capable of providing quick and low-cost testing in areas where traditional testing procedures are not feasible. This study focuses on COVID-19 detection using X-ray images. The prime objective is to introduce a computer-assisted diagnosis (CAD) system to differentiate COVID-19 from healthy and pneumonia cases using X-ray image sequences. This work utilizes standard transfer-learning techniques for COVID-19 detection. It proposes the master–slave architecture using the most state-of-the-art Densenet201 and Squeezenet1_0 techniques for classifying the COVID-19 virus in chest X-ray image sequences. This paper compares the proposed models with other standard transfer-learning approaches for COVID-19. The performance metrics demonstrate that the proposed approach outperforms standard transfer-learning approaches. This research also fine-tunes hyperparameters and predicts the optimized learning rate to achieve the highest accuracy in the model. After fine-tuning the learning rate, the DenseNet201 model retrieves an accuracy of 83.33%, while the fastest model is SqueezeNet1_0, which retrieves an accuracy of 80%.

COVID-19 has led to a severe impact on the society and healthcare system, with early diagnosis and effective treatment becoming critical. The Chest X-ray (CXR) is the most time-saving and cost-effective tool for diagnosing COVID-19. However, manual diagnosis through human eyes is time-consuming and tends to introduce errors. With the challenge of a large number of infections and a shortage of medical resources, a fast and accurate diagnosis technique is required. Manual detection is time-consuming, depends on individual experience, and tends to easily introduce errors. Deep learning methods can be used to develop automated detection and computer-aided diagnosis. However, they require a large amount of data, which is not practical due to the limited annotated CXR images. In this research, SDViT, an approach based on transformers, is proposed for COVID-19 diagnosis through image classification. We propose three innovations, namely, self-supervised learning, detail correction path (DCP), and domain transfer, then add them to the ViT Transformer architecture. Based on experimental results, our proposed method achieves an accuracy of 95.2381%, which is better performance compared to well-established methods on the X-ray Image dataset, along with the highest precision (0.952310), recall (0.963964), and F1-score (0.958102). Extensive experiments show that our model achieves the best performance on the synthetic-covid-cxr dataset as well. The experimental results demonstrate the advantages of our design for the classification task of COVID-19 X-ray images.

Chest X-ray radiographic (CXR) imagery enables earlier and easier lung disease diagnosis. Therefore, in this paper, we propose a deep learning method using a transfer learning technique to classify lung diseases on CXR images to improve the efficiency and accuracy of computer-aided diagnostic systems’ (CADs’) diagnostic performance. Our proposed method is a one-step, end-to-end learning, which means that raw CXR images are directly inputted into a deep learning model (EfficientNet v2-M) to extract their meaningful features in identifying disease categories. We experimented using our proposed method on three classes of normal, pneumonia, and pneumothorax of the U.S. National Institutes of Health (NIH) data set, and achieved validation performances of loss = 0.6933, accuracy = 82.15%, sensitivity = 81.40%, and specificity = 91.65%. We also experimented on the Cheonan Soonchunhyang University Hospital (SCH) data set on four classes of normal, pneumonia, pneumothorax, and tuberculosis, and achieved validation performances of loss = 0.7658, accuracy = 82.20%, sensitivity = 81.40%, and specificity = 94.48%; testing accuracy of normal, pneumonia, pneumothorax, and tuberculosis classes was 63.60%, 82.30%, 82.80%, and 89.90%, respectively.