European Journal of Radiology最新文献_第10页

Non-traumatic brachial plexopathy identification from routine MRIs: Retrospective studies with deep learning networks 从常规磁共振成像中识别非创伤性臂丛神经病：深度学习网络的回顾性研究

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-18 DOI: 10.1016/j.ejrad.2024.111744

Weiguo Cao , Benjamin M. Howe , Sumana Ramanathan , Nicholas G. Rhodes , Panagiotis Korfiatis , Kimberly K. Amrami , Robert J. Spinner , Timothy L. Kline

Purpose

This study aims to seek an optimized deep learning model for differentiating non-traumatic brachial plexopathy from routine MRI scans.

Materials and methods

This retrospective study collected patients through the electronic medical records (EMR) or pathological reports at Mayo Clinic and underwent BP MRI from January 2002 to December 2022. Using sagittal T1, fluid-sensitive and post-gadolinium images, a radiology panel selected BP’s region of interest (ROI) to form 3 dimensional volumes for this study. We designed six deep learning schemes to conduct BP abnormality differentiation across three MRI sequences. Utilizing five prestigious deep learning networks as the backbone, we trained and validated these models by nested five-fold cross-validation schemes. Furthermore, we defined a ’method score’ derived from the radar charts as a quantitative indicator as the guidance of the preference of the best model.

Results

This study selected 196 patients from initial 267 candidates. A total of 256 BP MRI series were compiled from them, comprising 123 normal and 133 abnormal series. The abnormal series included 4 sub-categories, et al. breast cancer (22.5 %), lymphoma (27.1 %), inflammatory conditions (33.1 %) and others (17.2 %). The best-performing model was produced by feature merging mode with triple MRI joint strategy (AUC, 92.2 %; accuracy, 89.5 %) exceeding the multiple channel merging mode (AUC, 89.6 %; accuracy, 89.0 %), solo channel volume mode (AUC, 89.2 %; accuracy, 86.7 %) and the remaining. Evaluated by method score (maximum 2.37), the feature merging mode with backbone of VGG16 yielded the highest score of 1.75 under the triple MRI joint strategy.

Conclusion

Deployment of deep learning models across sagittal T1, fluid-sensitive and post-gadolinium MRI sequences demonstrated great potential for brachial plexopathy diagnosis. Our findings indicate that utilizing feature merging mode and multiple MRI joint strategy may offer satisfied deep learning model for BP abnormalities than solo-sequence analysis.

目的：本研究旨在寻求一种优化的深度学习模型，用于从常规 MRI 扫描中区分非创伤性臂丛神经病：这项回顾性研究通过梅奥诊所的电子病历（EMR）或病理报告收集了2002年1月至2022年12月期间接受臂丛神经核磁共振检查的患者。放射学小组利用矢状面 T1、液敏和钆后图像选择 BP 的感兴趣区（ROI），形成本研究的三维卷。我们设计了六种深度学习方案，在三种磁共振成像序列中进行 BP 异常分化。利用五个著名的深度学习网络作为骨干，我们通过嵌套的五倍交叉验证方案对这些模型进行了训练和验证。此外，我们还定义了从雷达图中得出的 "方法得分"，作为指导优选最佳模型的量化指标：本研究从最初的 267 名候选者中选出了 196 名患者。结果：这项研究从最初的 267 名候选者中选出了 196 名患者，并从中整理出 256 个 BP MRI 序列，包括 123 个正常序列和 133 个异常序列。异常序列包括 4 个子类别，即乳腺癌（22.5%）、淋巴瘤（27.1%）、炎症（33.1%）和其他（17.2%）。采用三重磁共振成像联合策略的特征合并模式（AUC，92.2%；准确率，89.5%）产生的模型表现最佳，超过了多通道合并模式（AUC，89.6%；准确率，89.0%）、单通道容积模式（AUC，89.2%；准确率，86.7%）和其他模式。根据方法得分（最高 2.37 分）进行评估，在三重磁共振成像联合策略下，以 VGG16 为骨干的特征合并模式得分最高，为 1.75 分：在矢状T1、流体敏感和钆后磁共振成像序列中部署深度学习模型，显示了臂丛神经病诊断的巨大潜力。我们的研究结果表明，与单序列分析相比，利用特征合并模式和多重 MRI 联合策略可为臂丛神经异常提供满意的深度学习模型。

{"title":"Non-traumatic brachial plexopathy identification from routine MRIs: Retrospective studies with deep learning networks","authors":"Weiguo Cao , Benjamin M. Howe , Sumana Ramanathan , Nicholas G. Rhodes , Panagiotis Korfiatis , Kimberly K. Amrami , Robert J. Spinner , Timothy L. Kline","doi":"10.1016/j.ejrad.2024.111744","DOIUrl":"10.1016/j.ejrad.2024.111744","url":null,"abstract":"<div><h3>Purpose</h3><div>This study aims to seek an optimized deep learning model for differentiating non-traumatic brachial plexopathy from routine MRI scans.</div></div><div><h3>Materials and methods</h3><div>This retrospective study collected patients through the electronic medical records (EMR) or pathological reports at Mayo Clinic and underwent BP MRI from January 2002 to December 2022. Using sagittal T1, fluid-sensitive and post-gadolinium images, a radiology panel selected BP’s region of interest (ROI) to form 3 dimensional volumes for this study. We designed six deep learning schemes to conduct BP abnormality differentiation across three MRI sequences. Utilizing five prestigious deep learning networks as the backbone, we trained and validated these models by nested five-fold cross-validation schemes. Furthermore, we defined a ’method score’ derived from the radar charts as a quantitative indicator as the guidance of the preference of the best model.</div></div><div><h3>Results</h3><div>This study selected 196 patients from initial 267 candidates. A total of 256 BP MRI series were compiled from them, comprising 123 normal and 133 abnormal series. The abnormal series included 4 sub-categories, et al. breast cancer (22.5 %), lymphoma (27.1 %), inflammatory conditions (33.1 %) and others (17.2 %). The best-performing model was produced by feature merging mode with triple MRI joint strategy (AUC, 92.2 %; accuracy, 89.5 %) exceeding the multiple channel merging mode (AUC, 89.6 %; accuracy, 89.0 %), solo channel volume mode (AUC, 89.2 %; accuracy, 86.7 %) and the remaining. Evaluated by method score (maximum 2.37), the feature merging mode with backbone of VGG16 yielded the highest score of 1.75 under the triple MRI joint strategy.</div></div><div><h3>Conclusion</h3><div>Deployment of deep learning models across sagittal T1, fluid-sensitive and post-gadolinium MRI sequences demonstrated great potential for brachial plexopathy diagnosis. Our findings indicate that utilizing feature merging mode and multiple MRI joint strategy may offer satisfied deep learning model for BP abnormalities than solo-sequence analysis.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111744"},"PeriodicalIF":3.2,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142389143","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}

引用次数: 0

Non-invasive imaging biomarkers in chronic liver disease 慢性肝病的无创成像生物标志物

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-18 DOI: 10.1016/j.ejrad.2024.111749

Cesare Maino , Federica Vernuccio , Roberto Cannella , Laura Cristoferi , Paolo Niccolò Franco , Marco Carbone , Francesco Cortese , Riccardo Faletti , Elisabetta De Bernardi , Riccardo Inchingolo , Marco Gatti , Davide Ippolito

Chronic liver disease (CLD) is a global and worldwide clinical challenge, considering that different underlying liver entities can lead to hepatic dysfunction. In the past, blood tests and clinical evaluation were the main noninvasive tools used to detect, diagnose and follow-up patients with CLD; in case of clinical suspicion of CLD or unclear diagnosis, liver biopsy has been considered as the reference standard to rule out different chronic liver conditions. Nowadays, noninvasive tests have gained a central role in the clinical pathway. Particularly, liver stiffness measurement (LSM) and cross-sectional imaging techniques can provide transversal information to clinicians, helping them to correctly manage, treat and follow patients during time. Cross-sectional imaging techniques, namely computed tomography (CT) and magnetic resonance imaging (MRI), have plenty of potential. Both techniques allow to compute the liver surface nodularity (LSN), associated with CLDs and risk of decompensation. MRI can also help quantify fatty liver infiltration, mainly with the proton density fat fraction (PDFF) sequences, and detect and quantify fibrosis, especially thanks to elastography (MRE). Advanced techniques, such as intravoxel incoherent motion (IVIM), T1- and T2- mapping are promising tools for detecting fibrosis deposition. Furthermore, the injection of hepatobiliary contrast agents has gained an important role not only in liver lesion characterization but also in assessing liver function, especially in CLDs. Finally, the broad development of radiomics signatures, applied to CT and MR, can be considered the next future approach to CLDs. The aim of this review is to provide a comprehensive overview of the current advancements and applications of both invasive and noninvasive imaging techniques in the evaluation and management of CLD.

慢性肝病（CLD）是一项全球性、世界性的临床挑战，因为不同的潜在肝脏实体都可能导致肝功能异常。过去，血液化验和临床评估是用于检测、诊断和随访慢性肝病患者的主要无创工具；在临床怀疑慢性肝病或诊断不明确的情况下，肝活检一直被视为排除不同慢性肝病的参考标准。如今，无创检测已在临床路径中占据重要地位。尤其是肝脏硬度测量（LSM）和横断面成像技术可以为临床医生提供横向信息，帮助他们正确管理、治疗和随访患者。横断面成像技术，即计算机断层扫描（CT）和磁共振成像（MRI），具有很大的潜力。这两种技术都可以计算肝表面结节度（LSN），这与慢性肝病和肝功能失代偿风险有关。核磁共振成像还能帮助量化脂肪肝浸润，主要是通过质子密度脂肪分数（PDFF）序列，以及检测和量化肝纤维化，特别是通过弹性成像（MRE）。先进的技术，如体内非相干运动（IVIM）、T1-和T2-映射，是检测纤维化沉积的有前途的工具。此外，注射肝胆造影剂不仅在肝脏病变定性方面，而且在评估肝功能方面，尤其是在慢性肝病方面，都发挥了重要作用。最后，应用于 CT 和 MR 的放射组学特征的广泛发展可被视为治疗 CLD 的下一个未来方法。本综述旨在全面概述目前有创和无创成像技术在评估和管理慢性肝病方面的进展和应用。

{"title":"Non-invasive imaging biomarkers in chronic liver disease","authors":"Cesare Maino , Federica Vernuccio , Roberto Cannella , Laura Cristoferi , Paolo Niccolò Franco , Marco Carbone , Francesco Cortese , Riccardo Faletti , Elisabetta De Bernardi , Riccardo Inchingolo , Marco Gatti , Davide Ippolito","doi":"10.1016/j.ejrad.2024.111749","DOIUrl":"10.1016/j.ejrad.2024.111749","url":null,"abstract":"<div><div>Chronic liver disease (CLD) is a global and worldwide clinical challenge, considering that different underlying liver entities can lead to hepatic dysfunction. In the past, blood tests and clinical evaluation were the main noninvasive tools used to detect, diagnose and follow-up patients with CLD; in case of clinical suspicion of CLD or unclear diagnosis, liver biopsy has been considered as the reference standard to rule out different chronic liver conditions. Nowadays, noninvasive tests have gained a central role in the clinical pathway. Particularly, liver stiffness measurement (LSM) and cross-sectional imaging techniques can provide transversal information to clinicians, helping them to correctly manage, treat and follow patients during time. Cross-sectional imaging techniques, namely computed tomography (CT) and magnetic resonance imaging (MRI), have plenty of potential. Both techniques allow to compute the liver surface nodularity (LSN), associated with CLDs and risk of decompensation. MRI can also help quantify fatty liver infiltration, mainly with the proton density fat fraction (PDFF) sequences, and detect and quantify fibrosis, especially thanks to elastography (MRE). Advanced techniques, such as intravoxel incoherent motion (IVIM), T1- and T2- mapping are promising tools for detecting fibrosis deposition. Furthermore, the injection of hepatobiliary contrast agents has gained an important role not only in liver lesion characterization but also in assessing liver function, especially in CLDs. Finally, the broad development of radiomics signatures, applied to CT and MR, can be considered the next future approach to CLDs. The aim of this review is to provide a comprehensive overview of the current advancements and applications of both invasive and noninvasive imaging techniques in the evaluation and management of CLD.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111749"},"PeriodicalIF":3.2,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142312252","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}

引用次数: 0

Can the preoperative CT-based deep learning radiomics model predict histologic grade and prognosis of chondrosarcoma? 基于术前 CT 的深度学习放射组学模型能否预测软骨肉瘤的组织学分级和预后？

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-17 DOI: 10.1016/j.ejrad.2024.111719

Pei Nie , Xia Zhao , Jinlong Ma , Yicong Wang , Ben Li , Xiaoli Li , Qiyuan Li , Yanmei Wang , Yuchao Xu , Zhengjun Dai , Jie Wu , Ning Wang , Guangjie Yang , Dapeng Hao , Tengbo Yu

Background and purpose

Computed tomography (CT) and biopsy may be insufficient for preoperative evaluation of the grade and outcome of patients with chondrosarcoma. The aim of this study was to develop and validate a CT-based deep learning radiomics model (DLRM) for predicting histologic grade and prognosis in chondrosarcoma (CS).

Methods

A multicenter 211 (training cohort/ test cohort, 127/84) CS patients were enrolled. Radiomics signature (RS), deep learning signature (DLS), and DLRM incorporating radiomics and deep learning features were developed for predicting the grade. Kaplan-Meier survival analysis was used to assess the association of the model-predicted grade with recurrence-free survival (RFS). Model performance was evaluated with the area under the receiver operating characteristic curve (AUC) and the Harrell’s concordance index (C-index).

Results

The DLRM (AUC, 0.879; 95 % confidence interval [CI], 0.802–0.956) outperformed (z = 2.773, P=0.006) the RS (AUC, 0.715;95 % CI, 0.606–0.825) in predicting grade in the test cohort. RFS showed significant differences (log-rank test, P<0.05) between low-grade and high-grade patients stratified by DLRM. The DLRM achieved a higher C-index (0.805; 95 % CI, 0.694–0.916) than the RS (0.692, 95 % CI, 0.540–0.844) did in predicting RFS for CS patients in the test cohort.

Conclusion

The DLRM can accurately predict the histologic grade and prognosis in CS.

背景和目的计算机断层扫描（CT）和活检可能不足以对软骨肉瘤患者的分级和预后进行术前评估。本研究旨在开发和验证一种基于 CT 的深度学习放射组学模型（DLRM），用于预测软骨肉瘤（CS）的组织学分级和预后。开发了用于预测分级的放射组学特征（RS）、深度学习特征（DLS）和包含放射组学和深度学习特征的DLRM。Kaplan-Meier 生存分析用于评估模型预测的分级与无复发生存期（RFS）之间的关系。结果 DLRM（AUC，0.879；95% 置信区间[CI]，0.802-0.956）在预测测试队列中的分级方面优于 RS（AUC，0.715；95% 置信区间[CI]，0.606-0.825）（z = 2.773，P=0.006）。按 DLRM 分层的低分级和高级别患者的 RFS 有明显差异（对数秩检验，P<0.05）。在预测测试队列中 CS 患者的 RFS 时，DLRM 的 C 指数（0.805；95 % CI，0.694-0.916）高于 RS（0.692，95 % CI，0.540-0.844）。

{"title":"Can the preoperative CT-based deep learning radiomics model predict histologic grade and prognosis of chondrosarcoma?","authors":"Pei Nie , Xia Zhao , Jinlong Ma , Yicong Wang , Ben Li , Xiaoli Li , Qiyuan Li , Yanmei Wang , Yuchao Xu , Zhengjun Dai , Jie Wu , Ning Wang , Guangjie Yang , Dapeng Hao , Tengbo Yu","doi":"10.1016/j.ejrad.2024.111719","DOIUrl":"10.1016/j.ejrad.2024.111719","url":null,"abstract":"<div><h3>Background and purpose</h3><p>Computed tomography (CT) and biopsy may be insufficient for preoperative evaluation of the grade and outcome of patients with chondrosarcoma. The aim of this study was to develop and validate a CT-based deep learning radiomics model (DLRM) for predicting histologic grade and prognosis in chondrosarcoma (CS).</p></div><div><h3>Methods</h3><p>A multicenter 211 (training cohort/ test cohort, 127/84) CS patients were enrolled. Radiomics signature (RS), deep learning signature (DLS), and DLRM incorporating radiomics and deep learning features were developed for predicting the grade. Kaplan-Meier survival analysis was used to assess the association of the model-predicted grade with recurrence-free survival (RFS). Model performance was evaluated with the area under the receiver operating characteristic curve (AUC) and the Harrell’s concordance index (C-index).</p></div><div><h3>Results</h3><p>The DLRM (AUC, 0.879; 95 % confidence interval [CI], 0.802–0.956) outperformed (z = 2.773, P=0.006) the RS (AUC, 0.715;95 % CI, 0.606–0.825) in predicting grade in the test cohort. RFS showed significant differences (log-rank test, P<0.05) between low-grade and high-grade patients stratified by DLRM. The DLRM achieved a higher C-index (0.805; 95 % CI, 0.694–0.916) than the RS (0.692, 95 % CI, 0.540–0.844) did in predicting RFS for CS patients in the test cohort.</p></div><div><h3>Conclusion</h3><p>The DLRM can accurately predict the histologic grade and prognosis in CS.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111719"},"PeriodicalIF":3.2,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142272582","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}

引用次数: 0

Accuracy of radiomics-Based models in distinguishing between ruptured and unruptured intracranial aneurysms: A systematic review and meta-Analysis 基于放射组学的模型在区分颅内动脉瘤破裂和未破裂方面的准确性：系统回顾和元分析

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-16 DOI: 10.1016/j.ejrad.2024.111739

Ahmadreza Sohrabi-Ashlaghi, Narges Azizi, Hedayat Abbastabar, Madjid Shakiba, Jayran Zebardast, Kavous Firouznia

Introduction

Intracranial aneurysms (IAs) pose a severe health risk due to the potential for subarachnoid hemorrhage upon rupture. This study aims to conduct a systematic review and meta-analysis on the accuracy of radiomics features derived from computed tomography angiography (CTA) in differentiating ruptured from unruptured IAs.

Materials and Methods

A systematic search was performed across multiple databases for articles published up to January 2024. Observational studies analyzing CTA using radiomics features were included. The area under the curve (AUC) for classifying ruptured vs. unruptured IAs was pooled using a random-effects model. Subgroup analyses were conducted based on the use of radiomics-only features versus radiomics plus additional image-based features, as well as the type of filters used for image processing.

Results

Six studies with 4,408 patients were included. The overall pooled AUC for radiomics features in differentiating ruptured from unruptured IAs was 0.86 (95% CI: 0.84–0.88). The AUC was 0.85 (95% CI: 0.82–0.88) for studies using only radiomics features and 0.87 (95% CI: 0.83–0.91) for studies incorporating radiomics plus additional image-based features. Subgroup analysis based on filter type showed an AUC of 0.87 (95% CI: 0.83–0.90) for original filters and 0.86 (95% CI: 0.81–0.90) for studies using additional filters.

Conclusion

Radiomics-based models demonstrate very good diagnostic accuracy in classifying ruptured and unruptured IAs, with AUC values exceeding 0.8. This highlights the potential of radiomics as a useful tool in the non-invasive assessment of aneurysm rupture risk, particularly in the management of patients with multiple aneurysms.

导言颅内动脉瘤（IAs）破裂后可能导致蛛网膜下腔出血，对健康构成严重威胁。本研究旨在对计算机断层扫描血管造影（CTA）得出的放射组学特征在区分破裂和未破裂的颅内动脉瘤方面的准确性进行系统回顾和荟萃分析。纳入了利用放射组学特征对 CTA 进行分析的观察性研究。采用随机效应模型对破裂与未破裂IA的分类曲线下面积（AUC）进行了汇总。根据仅使用放射组学特征与放射组学加其他图像特征以及图像处理所用过滤器的类型进行了分组分析。放射组学特征在区分破裂和未破裂IA方面的总体AUC为0.86（95% CI：0.84-0.88）。仅使用放射组学特征的研究的AUC为0.85（95% CI：0.82-0.88），结合放射组学和其他图像特征的研究的AUC为0.87（95% CI：0.83-0.91）。基于滤波器类型的分组分析显示，原始滤波器的 AUC 为 0.87（95% CI：0.83-0.90），使用附加滤波器的研究的 AUC 为 0.86（95% CI：0.81-0.90）。这凸显了放射组学作为无创评估动脉瘤破裂风险的有用工具的潜力，尤其是在管理多发性动脉瘤患者时。

{"title":"Accuracy of radiomics-Based models in distinguishing between ruptured and unruptured intracranial aneurysms: A systematic review and meta-Analysis","authors":"Ahmadreza Sohrabi-Ashlaghi, Narges Azizi, Hedayat Abbastabar, Madjid Shakiba, Jayran Zebardast, Kavous Firouznia","doi":"10.1016/j.ejrad.2024.111739","DOIUrl":"10.1016/j.ejrad.2024.111739","url":null,"abstract":"<div><h3>Introduction</h3><p>Intracranial aneurysms (IAs) pose a severe health risk due to the potential for subarachnoid hemorrhage upon rupture. This study aims to conduct a systematic review and <em>meta</em>-analysis on the accuracy of radiomics features derived from computed tomography angiography (CTA) in differentiating ruptured from unruptured IAs.</p></div><div><h3>Materials and Methods</h3><p>A systematic search was performed across multiple databases for articles published up to January 2024. Observational studies analyzing CTA using radiomics features were included. The area under the curve (AUC) for classifying ruptured vs. unruptured IAs was pooled using a random-effects model. Subgroup analyses were conducted based on the use of radiomics-only features versus radiomics plus additional image-based features, as well as the type of filters used for image processing.</p></div><div><h3>Results</h3><p>Six studies with 4,408 patients were included. The overall pooled AUC for radiomics features in differentiating ruptured from unruptured IAs was 0.86 (95% CI: 0.84–0.88). The AUC was 0.85 (95% CI: 0.82–0.88) for studies using only radiomics features and 0.87 (95% CI: 0.83–0.91) for studies incorporating radiomics plus additional image-based features. Subgroup analysis based on filter type showed an AUC of 0.87 (95% CI: 0.83–0.90) for original filters and 0.86 (95% CI: 0.81–0.90) for studies using additional filters.</p></div><div><h3>Conclusion</h3><p>Radiomics-based models demonstrate very good diagnostic accuracy in classifying ruptured and unruptured IAs, with AUC values exceeding 0.8. This highlights the potential of radiomics as a useful tool in the non-invasive assessment of aneurysm rupture risk, particularly in the management of patients with multiple aneurysms.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111739"},"PeriodicalIF":3.2,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142238549","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}

引用次数: 0

MRI quantitative T1 and T2 mapping of the renal cortex: Assessment of normal values and potential usefulness for renal masses at 3 T 肾皮质的磁共振成像定量 T1 和 T2 映像：在 3 T 下评估正常值和肾脏肿块的潜在用途

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-15 DOI: 10.1016/j.ejrad.2024.111741

Remy Henry , Thibaut Goetsch , Laurent Brandhuber , Aissam Labani , Sébastien Moliére , Mickael Ohana , Catherine Roy

Purpose

The purpose of this monocentric retrospective study consisted in exploring the potential improvement of the assessment of renal masses on MRI by using the T1 (T1m) and T2 (T2m) mapping relaxation times.

Materials and Methods

We recorded the renal cortex values of 125 patients with normal kidneys (reference group) and 75 patients with renal masses on a clinical 3 T MR unit using T1m and T2m sequences.

For the quantitative evaluation, measurements were performed by delineating ROIs on T1m and T2m sequences in renal cortex of the reference group and in renal masses.

Interobserver agreement for the qualitative analysis of image quality was assessed using quadratic Cohen’s weighted kappa statistics (k).

Student’s paired t-test and non-parametric Kruskal-Wallis test were used to compare our datasets in terms of T1m and T2m values.

Results

For the cohort of reference group, mean renal cortex T1m and T2m values were 1,529 ± 83 ms and 98 ± 7 ms, respectively. No statistically significant differences were found for T1m and T2m in the reference group regardless of age, gender or eGRF categories.

For the group with renal masses, mean T1m and T2m values were 1,667 ± 87 ms and 105 ± 8 ms; 1,621 ± 96 ms and 117 ± 6 ms, and 1,530 ± 62 ms and 85 ± 4 ms for renal cell carcinomas, angiomyolipomas, and oncocytomas, respectively. For T1m values, there was no significant difference (p = 0.37) among the three types of renal masses. Among histological subtypes we have found: RCC versus angiomyolipoma (p = 0.25), RCC versus oncocytoma (p = 0.15), and oncocytoma versus angiomyolipoma (p = 0.47).

However, we have found a statistically significant difference for the T2m value (p = 0.0005). Among histological subtypes, only T2m values were statistically significant for each combination: RCC versus angiomyolipoma (p = 0.012), RCC versus oncocytoma (p = 0.0002), and oncocytoma versus angiomyolipoma (p = 0.003).

Conclusion

As this is the largest normal patient cohort, the T1m and T2m values recorded could be proposed as reference values and can play a role in the differential diagnosis between benign and malignant renal tumoral masses.

材料和方法我们在临床 3 T MR 设备上使用 T1m 和 T2m 序列记录了 125 名正常肾脏患者（参照组）和 75 名肾脏肿块患者的肾皮质值。在定量评估中，通过在参照组和肾脏肿块的 T1m 和 T2m 序列上划分 ROI 进行测量。采用学生配对 t 检验和非参数 Kruskal-Wallis 检验比较数据集的 T1m 和 T2m 值。对于肾肿块组，肾细胞癌、血管脂肪瘤和肿瘤细胞瘤的 T1m 和 T2m 平均值分别为 1,667 ± 87 ms 和 105 ± 8 ms；1,621 ± 96 ms 和 117 ± 6 ms；1,530 ± 62 ms 和 85 ± 4 ms。至于 T1m 值，三种类型的肾肿块之间没有显著差异（p = 0.37）。在组织学亚型中，我们发现然而，我们发现 T2m 值的差异具有统计学意义（p = 0.0005）。在组织学亚型中，只有 T2m 值对每种组合具有统计学意义：结论由于这是最大的正常患者队列，记录的 T1m 和 T2m 值可作为参考值，并可在良性和恶性肾肿瘤肿块的鉴别诊断中发挥作用。

{"title":"MRI quantitative T1 and T2 mapping of the renal cortex: Assessment of normal values and potential usefulness for renal masses at 3 T","authors":"Remy Henry , Thibaut Goetsch , Laurent Brandhuber , Aissam Labani , Sébastien Moliére , Mickael Ohana , Catherine Roy","doi":"10.1016/j.ejrad.2024.111741","DOIUrl":"10.1016/j.ejrad.2024.111741","url":null,"abstract":"<div><h3>Purpose</h3><p>The purpose of this monocentric retrospective study consisted in exploring the potential improvement of the assessment of renal masses on MRI by using the T1 (T1m) and T2 (T2m) mapping relaxation times.</p></div><div><h3>Materials and Methods</h3><p>We recorded the renal cortex values of 125 patients with normal kidneys (reference group) and 75 patients with renal masses on a clinical 3 T MR unit using T1m and T2m sequences.</p><p>For the quantitative evaluation, measurements were performed by delineating ROIs on T1m and T2m sequences in renal cortex of the reference group and in renal masses.</p><p>Interobserver agreement for the qualitative analysis of image quality was assessed using quadratic Cohen’s weighted kappa statistics (k).</p><p>Student’s paired <em>t</em>-test and non-parametric Kruskal-Wallis test were used to compare our datasets in terms of T1m and T2m values.</p></div><div><h3>Results</h3><p>For the cohort of reference group, mean renal cortex T1m and T2m values were 1,529 ± 83 ms and 98 ± 7 ms, respectively. No statistically significant differences were found for T1m and T2m in the reference group regardless of age, gender or eGRF categories.</p><p>For the group with renal masses, mean T1m and T2m values were 1,667 ± 87 ms and 105 ± 8 ms; 1,621 ± 96 ms and 117 ± 6 ms, and 1,530 ± 62 ms and 85 ± 4 ms for renal cell carcinomas, angiomyolipomas, and oncocytomas, respectively. For T1m values, there was no significant difference (p = 0.37) among the three types of renal masses. Among histological subtypes we have found: RCC versus angiomyolipoma (p = 0.25), RCC versus oncocytoma (p = 0.15), and oncocytoma versus angiomyolipoma (p = 0.47).</p><p>However, we have found a statistically significant difference for the T2m value (p = 0.0005). Among histological subtypes, only T2m values were statistically significant for each combination: RCC versus angiomyolipoma (p = 0.012), RCC versus oncocytoma (p = 0.0002), and oncocytoma versus angiomyolipoma (p = 0.003).</p></div><div><h3>Conclusion</h3><p>As this is the largest normal patient cohort, the T1m and T2m values recorded could be proposed as reference values and can play a role in the differential diagnosis between benign and malignant renal tumoral masses.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111741"},"PeriodicalIF":3.2,"publicationDate":"2024-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0720048X24004571/pdfft?md5=7a3b676fe67f748d221deeacfa5ebc35&pid=1-s2.0-S0720048X24004571-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142272599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Transjugular intrahepatic portosystemic shunt (TIPS) with variceal embolization reduces rebleeding risk for patients with portal pressure gradient over 12 mmHg: A long-term follow-up study 经颈静脉肝内门体分流术（TIPS）与静脉曲张栓塞术可降低门脉压力梯度超过 12 mmHg 患者的再出血风险：长期随访研究

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-15 DOI: 10.1016/j.ejrad.2024.111740

Yaowei Bai , Jiacheng Liu , Wenlong Wu , Binqian Zhou , Bo Sun , Wei Yao , Xiaoming Liu , Hu Zhao , Yusheng Guo , Xin Jiang , Bin Liang , Lian Yang , Chuansheng Zheng

Objective

The consensus on whether Transjugular intrahepatic portosystemic shunt (TIPS) should be combined with variceal embolization in the treatment of portal hypertension-induced bleeding has not yet been reached. This study aimed to compare the difference in rebleeding incidence between TIPS and TIPS combined with variceal embolization and to analyze the optimal population for variceal embolization.

Methods

Clinical data of 721 patients undergoing TIPS were retrospectively collected. Patients were divided into two groups: TIPS alone (n = 155) and TIPS with embolization (TIPS+E, n = 251). Kaplan-Meier (KM) curves were used to analyze prognostic differences between the two groups, and subgroup analysis was conducted based on post-TIPS portal pressure gradient (PPG) exceeding 12 mmHg.

Results

After TIPS placement, the mean PPG significantly decreased for all patients. A total of 51 patients (12.6 %) experienced rebleeding, with 24 cases (15.9 %) in the TIPS group and 27 cases (10.6 %) in the TIPS+E group. There was no significant difference in cumulative rebleeding incidence between the TIPS+E and TIPS groups. In the subgroup with post-TIPS PPG greater than 12 mmHg, the cumulative rebleeding incidence was significantly lower in the TIPS+E group compared to the TIPS group (HR = 0.47, 95 %CI = 0.24–0.93, Log rank P = 0.026). No significant difference was found in patients with a post-TIPS PPG less than 12 mmHg.

Conclusion

For patients with post-TIPS PPG exceeding 12 mmHg, simultaneous variceal embolization with TIPS placement significantly reduces the risk of rebleeding.

目的关于经颈静脉肝内门体分流术（TIPS）是否应与静脉曲张栓塞术联合治疗门静脉高压引起的出血，目前尚未达成共识。本研究旨在比较 TIPS 和 TIPS 联合静脉曲张栓塞术在再出血发生率上的差异，并分析静脉曲张栓塞术的最佳适用人群。患者分为两组：方法回顾性收集了 721 名接受 TIPS 的患者的临床数据，将患者分为两组：单纯 TIPS 组（155 人）和 TIPS 加栓塞组（TIPS+E，251 人）。采用 Kaplan-Meier (KM) 曲线分析两组患者的预后差异，并根据 TIPS 术后门静脉压力梯度（PPG）超过 12 mmHg 进行亚组分析。共有 51 例患者（12.6%）出现再出血，其中 TIPS 组 24 例（15.9%），TIPS+E 组 27 例（10.6%）。TIPS+E 组和 TIPS 组的累积再出血发生率没有明显差异。在TIPS后PPG大于12 mmHg的亚组中，TIPS+E组的累积再出血发生率明显低于TIPS组（HR = 0.47, 95 %CI = 0.24-0.93, Log rank P = 0.026）。结论对于 TIPS 后 PPG 超过 12 mmHg 的患者，同时进行静脉曲张栓塞术和 TIPS 置入术可大大降低再出血的风险。

{"title":"Transjugular intrahepatic portosystemic shunt (TIPS) with variceal embolization reduces rebleeding risk for patients with portal pressure gradient over 12 mmHg: A long-term follow-up study","authors":"Yaowei Bai , Jiacheng Liu , Wenlong Wu , Binqian Zhou , Bo Sun , Wei Yao , Xiaoming Liu , Hu Zhao , Yusheng Guo , Xin Jiang , Bin Liang , Lian Yang , Chuansheng Zheng","doi":"10.1016/j.ejrad.2024.111740","DOIUrl":"10.1016/j.ejrad.2024.111740","url":null,"abstract":"<div><h3>Objective</h3><p>The consensus on whether Transjugular intrahepatic portosystemic shunt (TIPS) should be combined with variceal embolization in the treatment of portal hypertension-induced bleeding has not yet been reached. This study aimed to compare the difference in rebleeding incidence between TIPS and TIPS combined with variceal embolization and to analyze the optimal population for variceal embolization.</p></div><div><h3>Methods</h3><p>Clinical data of 721 patients undergoing TIPS were retrospectively collected. Patients were divided into two groups: TIPS alone (n = 155) and TIPS with embolization (TIPS+E, n = 251). Kaplan-Meier (KM) curves were used to analyze prognostic differences between the two groups, and subgroup analysis was conducted based on post-TIPS portal pressure gradient (PPG) exceeding 12 mmHg.</p></div><div><h3>Results</h3><p>After TIPS placement, the mean PPG significantly decreased for all patients. A total of 51 patients (12.6 %) experienced rebleeding, with 24 cases (15.9 %) in the TIPS group and 27 cases (10.6 %) in the TIPS+E group. There was no significant difference in cumulative rebleeding incidence between the TIPS+E and TIPS groups. In the subgroup with post-TIPS PPG greater than 12 mmHg, the cumulative rebleeding incidence was significantly lower in the TIPS+E group compared to the TIPS group (HR = 0.47, 95 %CI = 0.24–0.93, Log rank P = 0.026). No significant difference was found in patients with a post-TIPS PPG less than 12 mmHg.</p></div><div><h3>Conclusion</h3><p>For patients with post-TIPS PPG exceeding 12 mmHg, simultaneous variceal embolization with TIPS placement significantly reduces the risk of rebleeding.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111740"},"PeriodicalIF":3.2,"publicationDate":"2024-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142238550","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}

引用次数: 0

Clinical diagnostic model for predicting indolent or aggressive lymphoma based on clinical information and ultrasound features of superficial lymph nodes 根据浅表淋巴结的临床信息和超声波特征预测轻度或侵袭性淋巴瘤的临床诊断模型

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-15 DOI: 10.1016/j.ejrad.2024.111738

Huifang Weng , Huisen Hu , Yanan Zhao , Yongyuan Xu , Panpan Chen , Pintong Huang

Purpose

The aim of this study was to develop a diagnostic model for predicting indolent lymphoma or aggressive lymphoma using clinical information and ultrasound characteristics of superficial lymph nodes.

Method

Patients with confirmed pathological lymphoma subtypes who had undergone ultrasound and contrast-enhanced ultrasound examinations were enrolled. Clinical and ultrasound imaging features were retrospectively analysed and compared to the pathological results, which were considered the gold standard for diagnosis. Two diagnostic models were developed: a clinical model (Model-C) using clinical data only, and a combined model (Model-US) integrating ultrasound features into the clinical model. The efficacy of these models in differentiating between indolent and aggressive lymphoma was compared.

Results

In total, 236 consecutive patients were enrolled, including 78 patients with indolent lymphomas and 158 patients with aggressive lymphomas. Receiver operating characteristic (ROC) curve analysis revealed that the areas under the curves of Model-C and Model-US were 0.78 (95 % confidence interval: 0.72–0.84) and 0.87 (95 % confidence interval: 0.82–0.92), respectively (p < 0.001). Model-US was further evaluated for calibration and is presented as a nomogram.

Conclusions

The diagnostic model incorporated clinical and ultrasound characteristics and offered a noninvasive method for assessing lymphoma with good discrimination and calibration.

目的本研究的目的是利用浅表淋巴结的临床信息和超声波特征，建立一个预测淋巴瘤是非淋巴瘤还是侵袭性淋巴瘤的诊断模型。对临床和超声成像特征进行回顾性分析，并与病理结果进行比较，病理结果被认为是诊断的金标准。研究人员开发了两种诊断模型：一种是仅使用临床数据的临床模型（模型-C），另一种是将超声特征纳入临床模型的组合模型（模型-US）。结果总共有 236 名连续患者被纳入研究，其中包括 78 名惰性淋巴瘤患者和 158 名侵袭性淋巴瘤患者。接收者操作特征曲线（ROC）分析显示，模型-C 和模型-US 的曲线下面积分别为 0.78（95% 置信区间：0.72-0.84）和 0.87（95% 置信区间：0.82-0.92）（p < 0.001）。结论该诊断模型结合了临床和超声特征，为评估淋巴瘤提供了一种无创方法，具有良好的区分度和校准性。

{"title":"Clinical diagnostic model for predicting indolent or aggressive lymphoma based on clinical information and ultrasound features of superficial lymph nodes","authors":"Huifang Weng , Huisen Hu , Yanan Zhao , Yongyuan Xu , Panpan Chen , Pintong Huang","doi":"10.1016/j.ejrad.2024.111738","DOIUrl":"10.1016/j.ejrad.2024.111738","url":null,"abstract":"<div><h3>Purpose</h3><p>The aim of this study was to develop a diagnostic model for predicting indolent lymphoma or aggressive lymphoma using clinical information and ultrasound characteristics of superficial lymph nodes.</p></div><div><h3>Method</h3><p>Patients with confirmed pathological lymphoma subtypes who had undergone ultrasound and contrast-enhanced ultrasound examinations were enrolled. Clinical and ultrasound imaging features were retrospectively analysed and compared to the pathological results, which were considered the gold standard for diagnosis. Two diagnostic models were developed: a clinical model (Model-C) using clinical data only, and a combined model (Model-US) integrating ultrasound features into the clinical model. The efficacy of these models in differentiating between indolent and aggressive lymphoma was compared.</p></div><div><h3>Results</h3><p>In total, 236 consecutive patients were enrolled, including 78 patients with indolent lymphomas and 158 patients with aggressive lymphomas. Receiver operating characteristic (ROC) curve analysis revealed that the areas under the curves of Model-C and Model-US were 0.78 (95 % confidence interval: 0.72–0.84) and 0.87 (95 % confidence interval: 0.82–0.92), respectively (p < 0.001). Model-US was further evaluated for calibration and is presented as a nomogram.</p></div><div><h3>Conclusions</h3><p>The diagnostic model incorporated clinical and ultrasound characteristics and offered a noninvasive method for assessing lymphoma with good discrimination and calibration.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111738"},"PeriodicalIF":3.2,"publicationDate":"2024-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142238548","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}

引用次数: 0

Clinical feasibility of deep learning–accelerated single-shot turbo spin echo sequence with enhanced denoising for pancreas MRI at 3 Tesla 深度学习加速单发涡轮自旋回波序列与增强型去噪在 3 特斯拉胰腺磁共振成像中的临床可行性

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-15 DOI: 10.1016/j.ejrad.2024.111737

Jeong Woo Kim , Bit Na Park , Dominik Nickel , Mun Young Paek , Chang Hee Lee

Purpose

To assess the feasibility of the single-shot turbo spin echo sequence using deep learning-based reconstruction (DLR) (HASTE_DL) with enhanced denoising for pancreas MRI.

Methods

Patients who underwent pancreas MRI from March to April 2021 were included. Four T2-weighted images (non-FS conventional HASTE vs. HASTE_DL with enhanced denoising and FS HASTE_DL with enhanced denoising vs. HASTE_DL) were acquired. Two readers independently assessed the image quality parameters of the two non-FS image sets using a 4-point Likert scale. The signal-to-noise ratio (SNR) of the cystic lesions and pancreatic parenchyma and the contrast-to-noise ratio between the cystic lesion and pancreatic parenchyma were calculated for all four image sets. The size of the largest cystic lesion and the diameter of pancreatic duct were measured.

Results

A total of 63 patients were included, 48 (76.2 %) of whom had 136 pancreatic cystic lesion(s). The acquisition times of conventional HASTE and HASTE_DL were 69 and 18 sec, respectively. All image quality parameters except artifacts for reader 2 were significantly better for HASTE_DL with enhanced denoising. Those images also received scores for overall image quality that were significantly higher than those for the conventional HASTE (3.26 ± 0.54 vs. 2.47 ± 0.56, p < 0.001). The SNR of the pancreatic cystic lesion and pancreatic parenchyma was significantly higher in the HASTE_DL with enhanced denoising (p < 0.001 for both). Inter-reader variability for measuring the pancreatic cyst size (ICC, 0.999 vs. 0.995; 95 % LoA, −0.13481 to 0.14743 vs. −0.24097 to 0.27404) and duct diameter (ICC, 0.994 vs. 0.969; 95 % LoA, −0.11684 to 0.36026 vs. −0.45544 to 0.44664) was lower in HASTE_DL with enhanced denoising than in the conventional HASTE.

Conclusion

HASTE_DL with enhanced denoising could be useful for reducing the acquisition time of pancreas MRI while improving the image quality for the evaluation of pancreatic cystic lesions.

目的评估使用基于深度学习的重建（DLR）（HASTEDL）和增强去噪的单发涡轮自旋回波序列用于胰腺磁共振成像的可行性。方法纳入 2021 年 3 月至 4 月期间接受胰腺磁共振成像的患者。采集了四幅T2加权图像（非FS传统HASTE与增强去噪的HASTEDL对比，FS增强去噪的HASTEDL与HASTEDL对比）。两名阅读者使用 4 点李克特量表独立评估两组非 FS 图像的图像质量参数。计算所有四组图像中囊性病变和胰腺实质的信噪比（SNR）以及囊性病变和胰腺实质的对比度-噪声比。结果共纳入 63 例患者，其中 48 例（76.2%）有 136 个胰腺囊性病变。传统 HASTE 和 HASTEDL 的采集时间分别为 69 秒和 18 秒。除了读者2的伪影之外，所有图像质量参数都明显优于增强去噪的HASTEDL。这些图像的整体图像质量得分也明显高于传统 HASTE（3.26 ± 0.54 vs. 2.47 ± 0.56，p < 0.001）。在增强去噪的 HASTEDL 中，胰腺囊性病变和胰腺实质的 SNR 明显更高（两者均为 p <0.001）。测量胰腺囊肿大小（ICC，0.999 vs. 0.995; 95 % LoA, -0.13481 to 0.14743 vs. -0.24097 to 0.27404）和胰管直径（ICC，0.994 vs. 0.969; 95 % LoA, -0.11684 to 0.36026 vs. -0.45544 to 0.结论增强去噪的 HASTEDL 可以缩短胰腺 MRI 的采集时间，同时提高胰腺囊性病变的图像质量。

{"title":"Clinical feasibility of deep learning–accelerated single-shot turbo spin echo sequence with enhanced denoising for pancreas MRI at 3 Tesla","authors":"Jeong Woo Kim , Bit Na Park , Dominik Nickel , Mun Young Paek , Chang Hee Lee","doi":"10.1016/j.ejrad.2024.111737","DOIUrl":"10.1016/j.ejrad.2024.111737","url":null,"abstract":"<div><h3>Purpose</h3><p>To assess the feasibility of the single-shot turbo spin echo sequence using deep learning-based reconstruction (DLR) (HASTE<sub>DL</sub>) with enhanced denoising for pancreas MRI.</p></div><div><h3>Methods</h3><p>Patients who underwent pancreas MRI from March to April 2021 were included. Four T2-weighted images (non-FS conventional HASTE vs. HASTE<sub>DL</sub> with enhanced denoising and FS HASTE<sub>DL</sub> with enhanced denoising vs. HASTE<sub>DL</sub>) were acquired. Two readers independently assessed the image quality parameters of the two non-FS image sets using a 4-point Likert scale. The signal-to-noise ratio (SNR) of the cystic lesions and pancreatic parenchyma and the contrast-to-noise ratio between the cystic lesion and pancreatic parenchyma were calculated for all four image sets. The size of the largest cystic lesion and the diameter of pancreatic duct were measured.</p></div><div><h3>Results</h3><p>A total of 63 patients were included, 48 (76.2 %) of whom had 136 pancreatic cystic lesion(s). The acquisition times of conventional HASTE and HASTE<sub>DL</sub> were 69 and 18 sec, respectively. All image quality parameters except artifacts for reader 2 were significantly better for HASTE<sub>DL</sub> with enhanced denoising. Those images also received scores for overall image quality that were significantly higher than those for the conventional HASTE (3.26 ± 0.54 vs. 2.47 ± 0.56, p < 0.001). The SNR of the pancreatic cystic lesion and pancreatic parenchyma was significantly higher in the HASTE<sub>DL</sub> with enhanced denoising (p < 0.001 for both). Inter-reader variability for measuring the pancreatic cyst size (ICC, 0.999 vs. 0.995; 95 % LoA, −0.13481 to 0.14743 vs. −0.24097 to 0.27404) and duct diameter (ICC, 0.994 vs. 0.969; 95 % LoA, −0.11684 to 0.36026 vs. −0.45544 to 0.44664) was lower in HASTE<sub>DL</sub> with enhanced denoising than in the conventional HASTE.</p></div><div><h3>Conclusion</h3><p>HASTE<sub>DL</sub> with enhanced denoising could be useful for reducing the acquisition time of pancreas MRI while improving the image quality for the evaluation of pancreatic cystic lesions.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111737"},"PeriodicalIF":3.2,"publicationDate":"2024-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142272598","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}

引用次数: 0

Quantifying the reproducibility and longitudinal repeatability of radiomics features in magnetic resonance Image-Guide accelerator Imaging: A phantom study 量化磁共振成像导引加速器成像中放射组学特征的可重复性和纵向可重复性：模型研究

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-14 DOI: 10.1016/j.ejrad.2024.111735

Hang Yu , Bin Tang , Yuchuan Fu , Weige Wei , Yisong He , Guyu Dai , Qing Xiao

Objective

This study aimed to quantitatively evaluate the inter-platform reproducibility and longitudinal acquisition repeatability of MRI radiomics features in Fluid-Attenuated Inversion Recovery (FLAIR), T2-weighted (T2W), and T1-weighted (T1W) sequences on MR-Linac systems using an American College of Radiology (ACR) phantom.

Materials and Methods

This study used two MR-Linac systems (A and B) in different cancer centers. The ACR phantom was scanned on system A daily for 30 consecutive days to evaluate longitudinal repeatability. Additionally, retest data were collected after repositioning the phantom. Inter-platform reproducibility was assessed by conducting scans under identical conditions using system B. Regions of interest were delineated on the T1W sequence from system A and mapped to other sequences via rigid registration. Intra-observer and inter-observer comparisons were conducted. Repeatability and reproducibility were assessed using the intraclass correlation coefficient (ICC) and coefficient of variation (CV). Robust radiomics features were identified based on ICC>0.9 and CV<10 %.

Results

Analysis showed that a higher proportion of radiomics features derived from longitudinal FLAIR sequence (51.65 %) met robustness criteria compared to T2W (48.35 %) and T1W (43.96 %). Additionally, more inter-platform features from the FLAIR sequence (62.64 %) were robust compared to T2W (42.86 %) and T1W (39.56 %). Test-retest and intra-observer repeatability were excellent across all sequences, with a median ICC of 0.99 and CV<5%. However, inter-observer reproducibility was inferior, especially for the T1W sequence.

Conclusions

Different sequences show variations in repeatability and reproducibility. The FLAIR sequence demonstrated advantages in both longitudinal repeatability and inter-platform reproducibility. Caution is warranted when interpreting data, particularly in longitudinal or multiplatform radiomics studies.

本研究旨在使用美国放射学会（ACR）模型，定量评估 MR-Linac 系统上流体衰减反转恢复（FLAIR）、T2 加权（T2W）和 T1 加权（T1W）序列中 MRI 放射组学特征的平台间可重复性和纵向采集可重复性。连续 30 天每天在 A 系统上扫描 ACR 模型，以评估纵向可重复性。此外，在重新定位模型后还收集了重测数据。通过使用系统 B 在相同条件下进行扫描，评估平台间的可重复性。在系统 A 的 T1W 序列上划定感兴趣区，并通过刚性配准映射到其他序列。进行观察者内部和观察者之间的比较。使用类内相关系数（ICC）和变异系数（CV）评估重复性和再现性。结果分析表明，与 T2W（48.35%）和 T1W（43.96%）相比，纵向 FLAIR 序列得出的放射组学特征中符合稳健性标准的比例更高（51.65%）。此外，与 T2W（42.86%）和 T1W（39.56%）相比，来自 FLAIR 序列的更多平台间特征（62.64%）具有稳健性。所有序列的测试重复性和观察者内部重复性都很好，中位 ICC 为 0.99，CV<5%。结论不同序列的重复性和再现性存在差异。FLAIR序列在纵向可重复性和平台间可重复性方面均表现出优势。在解释数据时，尤其是在纵向或多平台放射组学研究中，必须谨慎。

{"title":"Quantifying the reproducibility and longitudinal repeatability of radiomics features in magnetic resonance Image-Guide accelerator Imaging: A phantom study","authors":"Hang Yu , Bin Tang , Yuchuan Fu , Weige Wei , Yisong He , Guyu Dai , Qing Xiao","doi":"10.1016/j.ejrad.2024.111735","DOIUrl":"10.1016/j.ejrad.2024.111735","url":null,"abstract":"<div><h3>Objective</h3><p>This study aimed to quantitatively evaluate the inter-platform reproducibility and longitudinal acquisition repeatability of MRI radiomics features in Fluid-Attenuated Inversion Recovery (FLAIR), T2-weighted (T2W), and T1-weighted (T1W) sequences on MR-Linac systems using an American College of Radiology (ACR) phantom.</p></div><div><h3>Materials and Methods</h3><p>This study used two MR-Linac systems (A and B) in different cancer centers. The ACR phantom was scanned on system A daily for 30 consecutive days to evaluate longitudinal repeatability. Additionally, retest data were collected after repositioning the phantom. Inter-platform reproducibility was assessed by conducting scans under identical conditions using system B. Regions of interest were delineated on the T1W sequence from system A and mapped to other sequences via rigid registration. Intra-observer and inter-observer comparisons were conducted. Repeatability and reproducibility were assessed using the intraclass correlation coefficient (ICC) and coefficient of variation (CV). Robust radiomics features were identified based on ICC>0.9 and CV<10 %.</p></div><div><h3>Results</h3><p>Analysis showed that a higher proportion of radiomics features derived from longitudinal FLAIR sequence (51.65 %) met robustness criteria compared to T2W (48.35 %) and T1W (43.96 %). Additionally, more inter-platform features from the FLAIR sequence (62.64 %) were robust compared to T2W (42.86 %) and T1W (39.56 %). Test-retest and intra-observer repeatability were excellent across all sequences, with a median ICC of 0.99 and CV<5%. However, inter-observer reproducibility was inferior, especially for the T1W sequence.</p></div><div><h3>Conclusions</h3><p>Different sequences show variations in repeatability and reproducibility. The FLAIR sequence demonstrated advantages in both longitudinal repeatability and inter-platform reproducibility. Caution is warranted when interpreting data, particularly in longitudinal or multiplatform radiomics studies.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111735"},"PeriodicalIF":3.2,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142228897","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}

引用次数: 0

Imaging the pancreas with photon-counting CT – A review of normal pancreatic anatomy 用光子计数 CT 对胰腺进行成像 - 正常胰腺解剖回顾

IF 3.2 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2024-09-14 DOI: 10.1016/j.ejrad.2024.111736

Erik G.S. Brandt , Christoph F. Müller , Henrik Thomsen , Anders B. Rodell , Bulat Ibragimov , Michael B. Andersen

Purpose

Compared to conventional energy integrating detector CT, Photon-Counting CT (PCCT) has the advantage of increased spatial resolution. The pancreas is a highly complex organ anatomically. The increased spatial resolution of PCCT challenges radiologists’ knowledge of pancreatic anatomy. The purpose of this review was to review detailed macroscopic and microscopic anatomy of the pancreas in the context of current and future PCCT.

Method

This review is based on a literature review of all parts of pancreatic anatomy and a retrospective imaging review of PCCT scans from 20 consecutively included patients without pancreatic pathology (mean age 61.8 years, 11 female), scanned in the workup of pancreatic cancer with a contrast enhanced multiphase protocol. Two radiologists assessed the visibility of the main and accessory pancreatic ducts, side ducts, ampulla, major papilla, minor papilla, pancreatic arteries and veins, regional lymph nodes, coeliac ganglia, and coeliac plexus.

Results

The macroscopic anatomy of the pancreas was consistently visualized with PCCT. Visualization of detailed anatomy of the ductal system (including side ducts), papillae, arteries, vein, lymph nodes, and innervation was possible in 90% or more of patients with moderate to good interreader agreement.

Conclusion

PCCT scans of the pancreas visualizes previously unseen or inconsistently seen small anatomical structures consistently. Increased knowledge of pancreatic anatomy could have importance in imaging of pancreatic cancer and other pancreatic diseases.

目的与传统的能量积分探测器 CT 相比，光子计数 CT（PCCT）具有空间分辨率更高的优势。胰腺是一个解剖结构非常复杂的器官。PCCT 空间分辨率的提高对放射医师的胰腺解剖知识提出了挑战。本综述的目的是结合当前和未来的 PCCT 综述胰腺的详细宏观和微观解剖结构。方法本综述基于对胰腺解剖结构所有部分的文献综述，以及对连续纳入的 20 名无胰腺病变患者（平均年龄 61.8 岁，11 名女性）的 PCCT 扫描的回顾性成像综述，这些患者是在胰腺癌检查中使用对比度增强的多相方案进行扫描的。两名放射科医生评估了主胰管和附属胰管、侧胰管、安瓿、大乳头、小乳头、胰腺动脉和静脉、区域淋巴结、腹腔神经节和腹腔神经丛的可见度。90%或更多的患者可以看到胰腺导管系统（包括侧导管）、乳头、动脉、静脉、淋巴结和神经支配的详细解剖结构，读片者之间的一致性为中等到良好。胰腺解剖知识的增加对胰腺癌和其他胰腺疾病的成像具有重要意义。

{"title":"Imaging the pancreas with photon-counting CT – A review of normal pancreatic anatomy","authors":"Erik G.S. Brandt , Christoph F. Müller , Henrik Thomsen , Anders B. Rodell , Bulat Ibragimov , Michael B. Andersen","doi":"10.1016/j.ejrad.2024.111736","DOIUrl":"10.1016/j.ejrad.2024.111736","url":null,"abstract":"<div><h3>Purpose</h3><p>Compared to conventional energy integrating detector CT, Photon-Counting CT (PCCT) has the advantage of increased spatial resolution. The pancreas is a highly complex organ anatomically. The increased spatial resolution of PCCT challenges radiologists’ knowledge of pancreatic anatomy. The purpose of this review was to review detailed macroscopic and microscopic anatomy of the pancreas in the context of current and future PCCT.</p></div><div><h3>Method</h3><p>This review is based on a literature review of all parts of pancreatic anatomy and a retrospective imaging review of PCCT scans from 20 consecutively included patients without pancreatic pathology (mean age 61.8 years, 11 female), scanned in the workup of pancreatic cancer with a contrast enhanced multiphase protocol. Two radiologists assessed the visibility of the main and accessory pancreatic ducts, side ducts, ampulla, major papilla, minor papilla, pancreatic arteries and veins, regional lymph nodes, coeliac ganglia, and coeliac plexus.</p></div><div><h3>Results</h3><p>The macroscopic anatomy of the pancreas was consistently visualized with PCCT. Visualization of detailed anatomy of the ductal system (including side ducts), papillae, arteries, vein, lymph nodes, and innervation was possible in 90% or more of patients with moderate to good interreader agreement.</p></div><div><h3>Conclusion</h3><p>PCCT scans of the pancreas visualizes previously unseen or inconsistently seen small anatomical structures consistently. Increased knowledge of pancreatic anatomy could have importance in imaging of pancreatic cancer and other pancreatic diseases.</p></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"181 ","pages":"Article 111736"},"PeriodicalIF":3.2,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0720048X24004522/pdfft?md5=6cb66523b14b745452d0e45710831fbd&pid=1-s2.0-S0720048X24004522-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142272584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0