Journal of X-Ray Science and Technology最新文献

{"title":"A multimodal similarity-aware and knowledge-driven pre-training approach for reliable pneumoconiosis diagnosis.","authors":"Xueting Ren, Guohua Ji, Surong Chu, Shinichi Yoshida, Juanjuan Zhao, Baoping Jia, Yan Qiang","doi":"10.1177/08953996241296400","DOIUrl":"10.1177/08953996241296400","url":null,"abstract":"<p><strong>Background: </strong>Pneumoconiosis staging is challenging due to the low clarity of X-ray images and the small, diffuse nature of the lesions. Additionally, the scarcity of annotated data makes it difficult to develop accurate staging models. Although clinical text reports provide valuable contextual information, existing works primarily focus on designing multimodal image-text contrastive learning tasks, neglecting the high similarity of pneumoconiosis imaging representations. This results in inadequate extraction of fine-grained multimodal information and underutilization of domain knowledge, limiting their application in medical tasks.</p><p><strong>Objective: </strong>The study aims to address the limitations of current multimodal methods by proposing a new approach that improves the precision of pneumoconiosis diagnosis and staging through enhanced fine-grained learning and better utilization of domain knowledge.</p><p><strong>Methods: </strong>The proposed <b>M</b>ultimodal <b>S</b>imilarity-aware and <b>K</b>nowledge-driven <b>P</b>re-<b>T</b>raining (MSK-PT) approach involves two stages. In the first stage, we deeply analyze the similar features of pneumoconiosis images and use a similarity-aware modality alignment strategy to explore the fine-grained representations and associated disturbances of pneumoconiosis lesions between images and texts, guiding the model to match more appropriate feature representations. In the second stage, we utilize data-associated features and pre-stored domain knowledge features as priors and constraints to guide the downstream model in the visual domain without annotations. To address potential erroneous labels generated by model predictions, we further introduce an uncertainty threshold strategy to mitigate the negative impact of imperfect prediction labels and enhance model interpretability.</p><p><strong>Results: </strong>We collected and created the pneumoconiosis chest X-ray (PneumoCXR) dataset to evaluate our proposed MSK-PT method. The experimental results show that our method achieved a classification accuracy of 81.73%, outperforming the state-of-the-art algorithms by 2.53%.</p><p><strong>Conclusions: </strong>MSK-PT showed diagnostic performance that matches or exceeds the average radiologist's level, even with limited labeled data, highlighting the method's effectiveness and robustness.</p>","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":" ","pages":"229-248"},"PeriodicalIF":1.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143460306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced swin transformer based tuberculosis classification with segmentation using chest X-ray.","authors":"P Visu, V Sathiya, P Ajitha, R Surendran","doi":"10.1177/08953996241300018","DOIUrl":"10.1177/08953996241300018","url":null,"abstract":"<p><strong>Background: </strong>Tuberculosis disease is the disease that causes significant morbidity and mortality worldwide. Thus, early detection of the disease is crucial for proper treatment and controlling the spread of Tuberculosis disease. Chest X-ray imaging is one of the most widely used diagnostic tools for detecting the Tuberculosis, which is time-consuming, and prone to errors. Nowadays, deep learning model provides the automated classification of medical images with promising outcome.</p><p><strong>Objective: </strong>Thus, this research introduced a deep learning based segmentation and classification model. Initially, the Adaptive Gaussian Filtering based pre-processing and data augmentation is performed to remove artefacts and biased outcome. Then, Attention UNet (A_UNet) based segmentation is proposed for segmenting the required region of Chest X-ray.</p><p><strong>Methods: </strong>Using the segmented outcome, Enhanced Swin Transformer (EnSTrans) model based Tuberculosis classification model is designed with Residual Pyramid Network based Multi-layer perceptron (MLP) layer for enhancing the classification accuracy.</p><p><strong>Results: </strong>Enhanced Lotus Effect Optimization (EnLeO) Algorithm is employed for the loss function optimization of the EnSTrans model.</p><p><strong>Conclusions: </strong>The proposed methods acquired the Accuracy, Recall, Precision, F-score, and Specificity of 99.0576%, 98.9459%, 99.145%, 98.96%, and 99.152% respectively.</p>","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":" ","pages":"167-186"},"PeriodicalIF":1.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143460441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiomics and deep learning features of pericoronary adipose tissue on non-contrast computerized tomography for predicting non-calcified plaques.","authors":"Junli Yu, Yan Ding, Li Wang, Shunxin Hu, Ning Dong, Jiangnan Sheng, Yingna Ren, Ziyue Wang","doi":"10.1177/08953996241292476","DOIUrl":"10.1177/08953996241292476","url":null,"abstract":"<p><strong>Background: </strong>Inflammation of coronary arterial plaque is considered a key factor in the development of coronary heart disease. Early the plaque detection and timely treatment of the atherosclerosis could effectively reduce the risk of cardiovascular events. However, there is no study combining radiomics with deep learning techniques to predict non-calcified plaques (NCP) in coronary artery at present.</p><p><strong>Objective: </strong>To investigate the value of combination of radiomics and deep learning features based on non-contrast computerized tomography (CT) scans of pericoronary adipose tissue (PCAT), integrating with clinical risk factors of patients, in identifying coronary inflammation and predicting the presence of NCP.</p><p><strong>Methods: </strong>The clinical and imaging data of 353 patients were analyzed. The region of interest (ROI) of PCAT was manually outlined on non-contrast CT scan images, like coronary CT calcium score sequential images, then the radiomics and deep learning features in ROIs were extracted respectively. In training set (Center 1), after performing feature selection, radiomics and deep learning feature models were established, meanwhile, clinical models were built. Finally, combined models were developed out via integrating clinical, radiomics, and deep learning features. The predictive performance of the four feature model groups (clinical, radiomics, deep learning, and three combination) was assessed by seven different machine learning models through generation of receiver operating characteristic curves (ROC) and the calculation of area under the curve (AUC), sensitivity, specificity, and accuracy. Furthermore, the predictive performance of each model was validated in an external validation set (Center 2).</p><p><strong>Results: </strong>For the single model comparation, eXtreme Gradient Boosting (XGBoost) showed the best performance among the clinical model group in the validation set. And Random Forest (RF) exhibited the best indicative performance not only among the radiomics feature group but also in the deep learning feature model group. What's more, among the combined model group, RF still displayed the best predictive performance, with the value of AUC, sensitivity, specificity, and accuracy in the validation set are 0.963, 0.857, 0.929, and 0.905.</p><p><strong>Conclusion: </strong>The RF model in the combined model group based on non-contrast CT scan PCAT can predict the presence of NCP more accurately and has the potential for preliminary screening of the NCP.</p>","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":" ","pages":"96-108"},"PeriodicalIF":1.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143460277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MRI classification and discrimination of spinal schwannoma and meningioma based on deep learning.","authors":"Yidan Liu, Zhenhua Zhou, Yuanjun Wang","doi":"10.1177/08953996241289745","DOIUrl":"10.1177/08953996241289745","url":null,"abstract":"<p><strong>Backgroud: </strong>Schwannoma (SCH) and meningiomas (MEN) are the two most common primary spinal cord tumors. Differentiating between them preoperatively remains a clinical challenge due to the substantial overlap in their clinical presentation and imaging characteristics.</p><p><strong>Objective: </strong>The objective of this study is to facilitate early diagnosis of patients and reduce clinician stress by constructing a deep learning-based classification model for automatic diagnosis of schwannoma and meningiomas using magnetic resonance images (MRI).</p><p><strong>Methods: </strong>We retrospectively collected MRI images of 74 patients with pathologically confirmed schwannoma and meningiomas from 2015 to 2020 at a local hosipital, and constructed a CNN model based on the PyTorch's deep learning framework for the discrimination between the two. First, a modified feature fusion CNN model (ResNet34-SKConv) was trained by introducing a selective convolutional kernel module into the original CNN model. The introduction of the selective convolutional kernel module enhances the network's focus on tumor features and effectively improves the network's performance. Finally, the trained model was used to process all the MRI image slices to achieve the classification of SCH and MEN patients by the voting prediction method.</p><p><strong>Results: </strong>Using the 5-fold cross-validation method, this new ResNet34-SKConv model achieves a classification accuracy of 92.32%, a specificity of 95.87%, and a F1-score of 93.54, respectively.</p><p><strong>Conclusion: </strong>This study demonstrated that a classification model using a deep learning network can be effective in achieving differential diagnosis of SCH and MEN. Thus, the new method has great potential for developing new computer-aided diagnosis and applications with future clinical practice.</p>","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":" ","pages":"26-36"},"PeriodicalIF":1.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143460311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Selecting projection views based on error equidistribution for computed tomography.","authors":"Yinghui Zhang, Xing Zhao, Ke Chen, Hongwei Li","doi":"10.1177/08953996241289267","DOIUrl":"10.1177/08953996241289267","url":null,"abstract":"<p><strong>Background: </strong>Nonuniform sampling is a useful technique to optimize the acquisition of projections with a limited budget. Existing methods for selecting important projection views have limitations, such as relying on blueprint images or excessive computing resources.</p><p><strong>Methods: </strong>We aim to develop a simple nonuniform sampling method for selecting informative projection views suitable for practical CT applications. The proposed algorithm is inspired by two key observations: projection errors contain angle-specific information, and adding views around error peaks effectively reduces errors and improves reconstruction. Given a budget and an initial view set, the proposed method involves: estimating projection errors based on current set of projection views, adding more projection views based on error equidistribution to smooth out errors, and final image reconstruction based on the new set of projection views. This process can be recursive, and the initial view can be obtained uniformly or from a prior for greater efficiency.</p><p><strong>Results: </strong>Comparison with popular view selection algorithms using simulated and real data demonstrates consistently superior performance in identifying optimal views and generating high-quality reconstructions. Notably, the new algorithm performs well in both PSNR and SSIM metrics while being computationally efficient, enhancing its practicality for CT optimization.</p><p><strong>Conclusions: </strong>A projection view selection algorithm based on error equidistribution is proposed, offering superior reconstruction quality and efficiency over existing methods. It is ready for real CT applications to optimize dose utilization.</p>","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":" ","pages":"249-269"},"PeriodicalIF":1.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143459999","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive guide to content-based image retrieval algorithms with visualsift ensembling.","authors":"C Ramesh Babu Durai,R Sathesh Raaj,Sindhu Chandra Sekharan,V S Nishok","doi":"10.3233/xst-240189","DOIUrl":"https://doi.org/10.3233/xst-240189","url":null,"abstract":"BACKGROUNDContent-based image retrieval (CBIR) systems are vital for managing the large volumes of data produced by medical imaging technologies. They enable efficient retrieval of relevant medical images from extensive databases, supporting clinical diagnosis, treatment planning, and medical research.OBJECTIVEThis study aims to enhance CBIR systems' effectiveness in medical image analysis by introducing the VisualSift Ensembling Integration with Attention Mechanisms (VEIAM). VEIAM seeks to improve diagnostic accuracy and retrieval efficiency by integrating robust feature extraction with dynamic attention mechanisms.METHODSVEIAM combines Scale-Invariant Feature Transform (SIFT) with selective attention mechanisms to emphasize crucial regions within medical images dynamically. Implemented in Python, the model integrates seamlessly into existing medical image analysis workflows, providing a robust and accessible tool for clinicians and researchers.RESULTSThe proposed VEIAM model demonstrated an impressive accuracy of 97.34% in classifying and retrieving medical images. This performance indicates VEIAM's capability to discern subtle patterns and textures critical for accurate diagnostics.CONCLUSIONSBy merging SIFT-based feature extraction with attention processes, VEIAM offers a discriminatively powerful approach to medical image analysis. Its high accuracy and efficiency in retrieving relevant medical images make it a promising tool for enhancing diagnostic processes and supporting medical research in CBIR systems.","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":"79 1","pages":""},"PeriodicalIF":3.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142258308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Photothermal effect in X-ray images for computed tomography of metallic parts: Stainless steel spheres","authors":"V. Moock, Darien E. Arce Chávez, Crescencio García-Segundo, L. Ruiz-Huerta","doi":"10.3233/xst-230260","DOIUrl":"https://doi.org/10.3233/xst-230260","url":null,"abstract":"BACKGROUND: The environmental impact on industrial X-ray tomography systems has gained its attention in terms of image precision and metrology over recent years, yet is still complex due to the variety of applications. OBJECTIVE: The current study explores the photothermal repercussions of the overall radiation exposure time. It shows the emerging dimensional uncertainty when measuring a stainless steel sphere by means of circular tomography scans. METHODS: The authors develop a novel frame difference method for X-ray radiographies to evaluate the spatial changes induced in the projected absorption maps on the X-ray panel. The object of interest has a simple geometry for the purpose of proof of concept. The dominant source of the observed radial uncertainty is the photothermal effect due to high-energy X-ray scattering at the metal workpiece. Thermal variations are monitored by an infrared camera within the industrial tomography system, which confines that heat in the industrial grade X-ray system. RESULTS: The authors demonstrate that dense industrial computed tomography programs with major X-ray power notably affect the uncertainty of digital dimensional measurements. The registered temperature variations are consistent with dimensional changes in radiographies and hence form a source of error that might result in visible artifacts within the 3D image reconstruction. CONCLUSIONS: This contribution is of fundamental value to reach the balance between the number of projections and radial uncertainty tolerance when performing analysis with X-ray dimensional exploration in precision measurements with industrial tomography.","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":"43 36","pages":""},"PeriodicalIF":3.0,"publicationDate":"2024-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139442549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An adaptive weighted ensemble learning network for diabetic retinopathy classification","authors":"Panpan Wu, Yue Qu, Ziping Zhao, Yue Cui, Yurou Xu, Peng An, Hengyong Yu","doi":"10.3233/xst-230252","DOIUrl":"https://doi.org/10.3233/xst-230252","url":null,"abstract":"Diabetic retinopathy (DR) is one of the leading causes of blindness. However, because the data distribution of classes is not always balanced, it is challenging for automated early DR detection using deep learning techniques. In this paper, we propose an adaptive weighted ensemble learning method for DR detection based on optical coherence tomography (OCT) images. Specifically, we develop an ensemble learning model based on three advanced deep learning models for higher performance. To better utilize the cues implied in these base models, a novel decision fusion scheme is proposed based on the Bayesian theory in terms of the key evaluation indicators, to dynamically adjust the weighting distribution of base models to alleviate the negative effects potentially caused by the problem of unbalanced data size. Extensive experiments are performed on two public datasets to verify the effectiveness of the proposed method. A quadratic weighted kappa of 0.8487 and an accuracy of 0.9343 on the DRAC2022 dataset, and a quadratic weighted kappa of 0.9007 and an accuracy of 0.8956 on the APTOS2019 dataset are obtained, respectively. The results demonstrate that our method has the ability to enhance the ovearall performance of DR detection on OCT images.","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":"58 7","pages":""},"PeriodicalIF":3.0,"publicationDate":"2024-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139449386","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The investigation of dose rate and photon beam energy dependence of optimized PASSAG polymer gel dosimeter using magnetic resonance imaging","authors":"Bo Liu, Shaima Haithem Zaki, Eduardo García, Amanda Bonilla, D. Thabit, Aya Hussein Adab","doi":"10.3233/xst-230282","DOIUrl":"https://doi.org/10.3233/xst-230282","url":null,"abstract":"OBJECTIVE: It seems that dose rate (DR) and photon beam energy (PBE) may influence the sensitivity and response of polymer gel dosimeters. In the current project, the sensitivity and response dependence of optimized PASSAG gel dosimeter (OPGD) on DR and PBE were assessed. MATERIALS AND METHODS: We fabricated the OPGD and the gel samples were irradiated with various DRs and PBEs. Then, the sensitivity and response (R 2) of OPGD were obtained by MRI at various doses and post-irradiation times. RESULTS: Our analysis showed that the sensitivity and response of OPGD are not affected by the evaluated DRs and PBEs. It was also found that the dose resolution values of OPGD ranged from 9 to 33 cGy and 12 to 34 cGy for the evaluated DRs and PBEs, respectively. Additionally, the data demonstrated that the sensitivity and response dependence of OPGD on DR and PBE do not vary over various times after the irradiation. CONCLUSIONS: The findings of this research project revealed that the sensitivity and response dependence of OPGD are independent of DR and PBE.","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":"56 20","pages":""},"PeriodicalIF":3.0,"publicationDate":"2024-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139449477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Semi-supervised segmentation of metal-artifact contaminated industrial CT images using improved CycleGAN","authors":"Shi Bo Jiang, Yue Wen Sun, Shuo Xu, Hua Xia Zhang, Zhi Fang Wu","doi":"10.3233/xst-230233","DOIUrl":"https://doi.org/10.3233/xst-230233","url":null,"abstract":"Accurate segmentation of industrial CT images is of great significance in industrial fields such as quality inspection and defect analysis. However, reconstruction of industrial CT images often suffers from typical metal artifacts caused by factors like beam hardening, scattering, statistical noise, and partial volume effects. Traditional segmentation methods are difficult to achieve precise segmentation of CT images mainly due to the presence of these metal artifacts. Furthermore, acquiring paired CT image data required by fully supervised networks proves to be extremely challenging. To address these issues, this paper introduces an improved CycleGAN approach for achieving semi-supervised segmentation of industrial CT images. This method not only eliminates the need for removing metal artifacts and noise, but also enables the direct conversion of metal artifact-contaminated images into segmented images without the requirement of paired data. The average values of quantitative assessment of image segmentation performance can reach 0.96645 for Dice Similarity Coefficient(Dice) and 0.93718 for Intersection over Union(IoU). In comparison to traditional segmentation methods, it presents significant improvements in both quantitative metrics and visual quality, provides valuable insights for further research.","PeriodicalId":49948,"journal":{"name":"Journal of X-Ray Science and Technology","volume":"53 12","pages":""},"PeriodicalIF":3.0,"publicationDate":"2024-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139449293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}