European Journal of Radiology最新文献_第6页

Computed tomography description of abdominal complications due to fishbone ingestion 鱼骨摄入引起的腹部并发症的计算机断层扫描描述。

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-02-07 DOI: 10.1016/j.ejrad.2026.112720

Aya Ben Zitoun , Aida Tankoano , Alolia Aboikoni , Magaly Zappa , Lorenzo Garzelli

Purpose: Inadvertent ingestion of fishbone can lead to bowel perforation. Numerous cases reports have been published; therefore, we aimed to describe the localization and type of complication in a dedicated cohort.

Material and Methods: From 2009 to 2023, we queried our local informatic department (Cayenne Hospital, French Guiana) to search electronic patient records (ICD-10 code T18: Foreign body in alimentary tract) and we reviewed each case to identify abdominal complication caused by fishbone ingestion. Patients with complication occurring above the stomach without a computed tomography available (CT) were excluded. Clinical, biological, imaging and treatment related data were collected. The overall raw incidence was estimated.

Results: Forty-two patients (mean age: 60 years old ± 15; female 36%) were included for analysis from May 2014 to June 2023. Abdominal pain was present in 41/42 patients (98%) and inflammatory biological syndrome was found in 34/42 patients (81%). On CT, perforation of the bowel wall was seen in 41/42 patients (98%), intra-abdominal collection in 17/42 patients (40%) and pneumoperitoneum in 2/42 patients (5%). The ileum (11/42, 26%) was the most common complication site, followed by the liver (6/42, 14%), and the right colon and stomach (5/42, 11% each). Treatment included surgery for 23/42 patients (55%), antibiotics alone for 14/42 patients (33%), endoscopic removal for 4/42 patients (10%), and percutaneous drainage for 1/42 patient (2%). The estimated incidence was 2.9 per 100,000 person-years.

Conclusion: We report the largest analysis of bowel perforation secondary to fishbone ingestion, providing an exhaustive spectrum of localization and type.

用途：误食鱼骨可导致肠穿孔。已发表了许多病例报告；因此，我们的目的是在一个专门的队列中描述并发症的定位和类型。材料与方法：2009年至2023年，我们查询当地信息科（法属圭亚那卡宴医院）检索电子病历（ICD-10代码T18：消化道异物），并对每例病例进行回顾性分析，以确定摄入鱼骨引起的腹部并发症。排除了发生在胃上方且没有CT检查的并发症患者。收集临床、生物学、影像学及治疗相关资料。估计总的原始发病率。结果：2014年5月至2023年6月共纳入42例患者，平均年龄60±15岁，女性36%。42例患者中有41例（98%）出现腹痛，34例（81%）出现炎症性生物学综合征。在CT上，41/42例患者（98%）可见肠壁穿孔，17/42例患者（40%）可见腹腔内收集，2/42例患者（5%）可见气腹。回肠（11/42,26%）是最常见的并发症部位，其次是肝脏（6/42,14%）、右结肠和胃（5/42，各占11%）。治疗包括手术23/42例（55%），单独使用抗生素14/42例（33%），内镜下切除4/42例（10%），经皮引流1/42例（2%）。估计发病率为每10万人年2.9例。结论：我们报告了最大的分析继发于鱼骨摄入肠穿孔，提供了详尽的定位和类型。

{"title":"Computed tomography description of abdominal complications due to fishbone ingestion","authors":"Aya Ben Zitoun , Aida Tankoano , Alolia Aboikoni , Magaly Zappa , Lorenzo Garzelli","doi":"10.1016/j.ejrad.2026.112720","DOIUrl":"10.1016/j.ejrad.2026.112720","url":null,"abstract":"<div><div>Purpose: Inadvertent ingestion of fishbone can lead to bowel perforation. Numerous cases reports have been published; therefore, we aimed to describe the localization and type of complication in a dedicated cohort.</div><div>Material and Methods: From 2009 to 2023, we queried our local informatic department (Cayenne Hospital, French Guiana) to search electronic patient records (ICD-10 code T18: Foreign body in alimentary tract) and we reviewed each case to identify abdominal complication caused by fishbone ingestion. Patients with complication occurring above the stomach without a computed tomography available (CT) were excluded. Clinical, biological, imaging and treatment related data were collected. The overall raw incidence was estimated.</div><div>Results: Forty-two patients (mean age: 60 years old ± 15; female 36%) were included for analysis from May 2014 to June 2023. Abdominal pain was present in 41/42 patients (98%) and inflammatory biological syndrome was found in 34/42 patients (81%). On CT, perforation of the bowel wall was seen in 41/42 patients (98%), intra-abdominal collection in 17/42 patients (40%) and pneumoperitoneum in 2/42 patients (5%). The ileum (11/42, 26%) was the most common complication site, followed by the liver (6/42, 14%), and the right colon and stomach (5/42, 11% each). Treatment included surgery for 23/42 patients (55%), antibiotics alone for 14/42 patients (33%), endoscopic removal for 4/42 patients (10%), and percutaneous drainage for 1/42 patient (2%). The estimated incidence was 2.9 per 100,000 person-years.</div><div>Conclusion: We report the largest analysis of bowel perforation secondary to fishbone ingestion, providing an exhaustive spectrum of localization and type.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112720"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146178338","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

A novel and general spatiotemporal diagnostic model: Intratumoral outflow and peritumoral inflow for the differentiation and stratification of breast tumor 一个新的和通用的时空诊断模型：肿瘤内流出和肿瘤周围流入用于乳腺肿瘤的分化和分层

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2025-12-20 DOI: 10.1016/j.ejrad.2025.112622

Zhanao Meng , Chenyi Zhou , Hui Xie , Ting Chen , Chunhua Wu , Wenxuan Li , Wenjie Tang , Yanling Wang

Objective

A novel Intratumoral Outflow and Peritumoral Inflow (IOPI) model based on spatiotemporal information was developed to quantify differences in blood perfusion and microcirculation. Its performance was compared with conventional indicators, and its clinical application for diagnosing benign and malignant breast tumors, as well as its potential for risk stratification, was validated in a two-center study.

Materials and methods

A retrospective analysis of 159 patients (80 malignant and 79 benign lesions) undergoing dynamic contrast-enhanced MRI was conducted. Training set was from Center 1, testing set was from Center 2. Conventional kinetic parameters (PE and SER) and interpreted MRI features (ADC and BI-RADS) were extracted. Hemodynamics models (IOPI, IIO, PIO and IPIO) and combined models (ADC + IOPI, BIRADS + IOPI) were constructed by logistic regression. Classification performance was assessed by AUC, sensitivity, specificity and accuracy.

Results

In the training cohort, the proposed IOPI model achieved an AUC of 0.911, with specificity of 95.1 %, sensitivity of 82.0 % and accuracy of 89.5 %. In external testing cohort, the AUC was 0.802, significantly outperformed conven tional indicators and others models (p < 0.05). The BIRADS combined IOPI model achieved the highest diagnostic performance in classifying malignancy of breast lesions, with the AUC values of 0.940 and sensitivity of 92.0 % in training cohort. In the testing cohort, the ADC combined IOPI model achieved the highest diagnostic performance with the AUC of 0.878, accuracy of 84.2 % and specificity of 88.9 %. The combined models also showed comparable performance in stratifying invasive grades and predicting Ki-67 expression (exceptional specificity of 90.9 % in ADC combined IOPI).

Conclusion

IOPI serves as a valuable tool that highlights the additional potential of spatiotemporal kinetic information for improving breast tumor diagnosis. The combined model which integrates time and spatial concepts, exhibited strong diagnostic performance for improved breast cancer diagnosis and risk stratification.

目的建立一种基于时空信息的肿瘤内流出和肿瘤周围流入（IOPI）模型，以量化血液灌注和微循环的差异。将其性能与常规指标进行比较，并通过双中心研究验证其在乳腺良恶性肿瘤诊断中的临床应用及风险分层潜力。材料与方法对159例（恶性80例，良性79例）行动态增强MRI检查的患者进行回顾性分析。训练集来自中心1，测试集来自中心2。提取常规动力学参数（PE和SER）和解释MRI特征（ADC和BI-RADS）。通过logistic回归建立血流动力学模型（IOPI、IIO、PIO和IPIO）和组合模型（ADC + IOPI、BIRADS + IOPI）。通过AUC、敏感性、特异性和准确性评价分类效果。结果在训练队列中，所提出的IOPI模型的AUC为0.911，特异性为95.1%，敏感性为82.0%，准确性为89.5%。在外部测试队列中，AUC为0.802，显著优于常规指标和其他模型（p < 0.05）。BIRADS联合IOPI模型对乳腺病变恶性分类的诊断效能最高，在训练队列中AUC值为0.940，灵敏度为92.0%。在检测队列中，ADC联合IOPI模型的诊断性能最高，AUC为0.878，准确率为84.2%，特异性为88.9%。联合模型在浸润分级和预测Ki-67表达方面也显示出相当的性能（ADC联合IOPI的异常特异性为90.9%）。结论iopi是一种有价值的工具，它突出了时空动力学信息在提高乳腺肿瘤诊断方面的附加潜力。该模型融合了时间和空间概念，在提高乳腺癌诊断和风险分层方面表现出较强的诊断性能。

{"title":"A novel and general spatiotemporal diagnostic model: Intratumoral outflow and peritumoral inflow for the differentiation and stratification of breast tumor","authors":"Zhanao Meng , Chenyi Zhou , Hui Xie , Ting Chen , Chunhua Wu , Wenxuan Li , Wenjie Tang , Yanling Wang","doi":"10.1016/j.ejrad.2025.112622","DOIUrl":"10.1016/j.ejrad.2025.112622","url":null,"abstract":"<div><h3>Objective</h3><div>A novel Intratumoral Outflow and Peritumoral Inflow (IOPI) model based on spatiotemporal information was developed to quantify differences in blood perfusion and microcirculation. Its performance was compared with conventional indicators, and its clinical application for diagnosing benign and malignant breast tumors, as well as its potential for risk stratification, was validated in a two-center study.</div></div><div><h3>Materials and methods</h3><div>A retrospective analysis of 159 patients (80 malignant and 79 benign lesions) undergoing dynamic contrast-enhanced MRI was conducted. Training set was from Center 1, testing set was from Center 2. Conventional kinetic parameters (PE and SER) and interpreted MRI features (ADC and BI-RADS) were extracted. Hemodynamics models (IOPI, IIO, PIO and IPIO) and combined models (ADC + IOPI, BIRADS + IOPI) were constructed by logistic regression. Classification performance was assessed by AUC, sensitivity, specificity and accuracy.</div></div><div><h3>Results</h3><div>In the training cohort, the proposed IOPI model achieved an AUC of 0.911, with specificity of 95.1 %, sensitivity of 82.0 % and accuracy of 89.5 %. In external testing cohort, the AUC was 0.802, significantly outperformed conven tional indicators and others models (<em>p</em> < 0.05). The BIRADS combined IOPI model achieved the highest diagnostic performance in classifying malignancy of breast lesions, with the AUC values of 0.940 and sensitivity of 92.0 % in training cohort. In the testing cohort, the ADC combined IOPI model achieved the highest diagnostic performance with the AUC of 0.878, accuracy of 84.2 % and specificity of 88.9 %. The combined models also showed comparable performance in stratifying invasive grades and predicting Ki-67 expression (exceptional specificity of 90.9 % in ADC combined IOPI).</div></div><div><h3>Conclusion</h3><div>IOPI serves as a valuable tool that highlights the additional potential of spatiotemporal kinetic information for improving breast tumor diagnosis. The combined model which integrates time and spatial concepts, exhibited strong diagnostic performance for improved breast cancer diagnosis and risk stratification.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112622"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074486","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

Utilizing baseline multiregional MRI radiomics for prediction of tumor deposition and prognosis following neoadjuvant therapy in resectable rectal cancer 利用基线多区域MRI放射组学预测可切除直肠癌新辅助治疗后的肿瘤沉积和预后

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-01-14 DOI: 10.1016/j.ejrad.2026.112676

Bingjie Wu , Lingwei Wang , Yang Wang , Fan Liu , Xujie Gao , Wenpeng Wang , Bohan Xiao , Ying Liu

Objective

To investigate whether pre-treatment T2WI-based multiregional radiomics can predict the probability of post-treatment tumor deposit (TD) and prognostic outcomes in patients with resectable rectal cancer after neoadjuvant therapy.

Materials and methods

This retrospective study included 159 patients with pathologically confirmed rectal cancer who received neoadjuvant therapy and then underwent surgery from March 2013 to March 2024. Radiomics features were extracted from the intratumoral region, a 3-mm-region straddling the tumor margin, and peritumoral 3 mm region on pre-treatment T2WI images. Clinical-radiomics nomogram was developed based on the most predictive radiomics signatures and clinical risk factors. Prognostic model for 5-year recurrence-free survival (RFS) was constructed by Cox regression analysis.

Results

The nomogram integrating clinical risk factors (Tumor distance to anal margin and MRI-reported extramural vascular invasion (EMVI)) with an intra-straddle 3 mm radiomics signature score (radscore) demonstrated optimal predictive performance with area under the receiver operating characteristic curve (AUC) of 0.953 (95% CI: 0.877–0.988), 0.810 (95% CI: 0.629–0.928) and 0.952 (95% CI: 0.857–0.992) in the training cohort, validation cohort and test cohort, respectively. The prognostic model constructed by intra-straddle 3 mm radscore (hazard ratio [HR] = 3.60, 95% CI: 1.59–8.16) and MRI-reported EMVI (HR = 6.07, 95% CI: 2.51–14.63) showed good performance for predicting 5‑year RFS with AUC of 0.827 (95% CI: 0.772–0.890) in the entire cohort.

Conclusion

The nomogram, incorporating pre-treatment MRI-based intra-straddle 3 mm radscore along with clinical risk factors, facilitates noninvasive assessment of the likelihood of TD positivity following neoadjuvant therapy, and has the power to predict 5-year RFS in patients with resectable rectal cancer.

目的探讨t2wi多区域放射组学对可切除直肠癌患者新辅助治疗后肿瘤沉积（TD）发生概率及预后的预测作用。材料与方法回顾性研究2013年3月至2024年3月，159例经病理证实的直肠癌患者接受新辅助治疗后行手术治疗。在治疗前T2WI图像上提取肿瘤内区域、横跨肿瘤边缘的3mm区域和肿瘤周围3mm区域的放射组学特征。临床放射组学图是基于最具预测性的放射组学特征和临床危险因素而开发的。采用Cox回归分析建立5年无复发生存期（RFS）预后模型。结果将临床危险因素（肿瘤到肛门边缘的距离和mri报告的外血管侵犯（EMVI））与跨骑3 mm放射组学特征评分（radscore）相结合的nomogram预测效果最佳，训练队列、验证队列和测试队列的受试者工作特征曲线下面积（AUC）分别为0.953 （95% CI: 0.877 ~ 0.988）、0.810 （95% CI: 0.629 ~ 0.928）和0.952 （95% CI: 0.857 ~ 0.992）。由跨骑3 mm radscore（风险比[HR] = 3.60, 95% CI: 1.59-8.16）和mri报告的EMVI（风险比[HR] = 6.07, 95% CI: 2.51-14.63）构建的预后模型在预测整个队列的5年RFS方面表现良好，AUC为0.827 （95% CI: 0.772-0.890）。该成像结合了治疗前基于mri的跨骑3mm radscore以及临床危险因素，有助于对新辅助治疗后TD阳性可能性进行无创评估，并具有预测可切除直肠癌患者5年RFS的能力。

{"title":"Utilizing baseline multiregional MRI radiomics for prediction of tumor deposition and prognosis following neoadjuvant therapy in resectable rectal cancer","authors":"Bingjie Wu , Lingwei Wang , Yang Wang , Fan Liu , Xujie Gao , Wenpeng Wang , Bohan Xiao , Ying Liu","doi":"10.1016/j.ejrad.2026.112676","DOIUrl":"10.1016/j.ejrad.2026.112676","url":null,"abstract":"<div><h3>Objective</h3><div>To investigate whether pre-treatment T2WI-based multiregional radiomics can predict the probability of post-treatment tumor deposit (TD) and prognostic outcomes in patients with resectable rectal cancer after neoadjuvant therapy.</div></div><div><h3>Materials and methods</h3><div>This retrospective study included 159 patients with pathologically confirmed rectal cancer who received neoadjuvant therapy and then underwent surgery from March 2013 to March 2024. Radiomics features were extracted from the intratumoral region, a 3-mm-region straddling the tumor margin, and peritumoral 3 mm region on pre-treatment T2WI images. Clinical-radiomics nomogram was developed based on the most predictive radiomics signatures and clinical risk factors. Prognostic model for 5-year recurrence-free survival (RFS) was constructed by Cox regression analysis.</div></div><div><h3>Results</h3><div>The nomogram integrating clinical risk factors (Tumor distance to anal margin and MRI-reported extramural vascular invasion (EMVI)) with an intra-straddle 3 mm radiomics signature score (radscore) demonstrated optimal predictive performance with area under the receiver operating characteristic curve (AUC) of 0.953 (95% CI: 0.877–0.988), 0.810 (95% CI: 0.629–0.928) and 0.952 (95% CI: 0.857–0.992) in the training cohort, validation cohort and test cohort, respectively. The prognostic model constructed by intra-straddle 3 mm radscore (hazard ratio [HR] = 3.60, 95% CI: 1.59–8.16) and MRI-reported EMVI (HR = 6.07, 95% CI: 2.51–14.63) showed good performance for predicting 5‑year RFS with AUC of 0.827 (95% CI: 0.772–0.890) in the entire cohort.</div></div><div><h3>Conclusion</h3><div>The nomogram, incorporating pre-treatment MRI-based intra-straddle 3 mm radscore along with clinical risk factors, facilitates noninvasive assessment of the likelihood of TD positivity following neoadjuvant therapy, and has the power to predict 5-year RFS in patients with resectable rectal cancer.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112676"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074619","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

Association between cognitive status and structural brain changes in Alzheimer’s disease: Clinical implication of lightweight deep learning-aided diagnosis 阿尔茨海默病认知状态与大脑结构变化之间的关系：轻量级深度学习辅助诊断的临床意义

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-01-14 DOI: 10.1016/j.ejrad.2026.112678

Po-Hsuan Hsieh , Ya-Fang Chen , Ta-Fu Chen , Wen-Chau Wu , for the Alzheimer’s Disease Neuroimaging Initiative

Background

The complex brain changes involved in Alzheimer’s disease (AD) development constitute a high-dimensional nonlinear feature space where deep learning (DL) classification/diagnosis may be advantageous over classical non-learning methods. However, the practicality of DL remains under debate among healthcare professionals, largely because many models are computationally expensive and operate without explicit interpretability. This study aimed to construct a lightweight DL model to disclose the association between cognitive status and structural brain changes in AD.

Methods

By using the data obtained from the Alzheimer’s Disease Neuroimaging Initiative database, 418 AD patients and 418 age-matched cognitively normal (CN) subjects were included for DL model construction based on their T1-weighted magnetic resonance images at baseline visit. A lightweight design was achieved by incorporating group convolution, global pooling, and efficient channel attention.

Results

The accuracy rate of our model was 90.6 %, competitive with previous models built with up-to-ten times more parameters. The occlusion maps showed that the medial temporal area and thalamus accounted the most for our model’s differentiation between AD and CN, in line with current knowledge of the pathological trajectory. Hierarchical regression further revealed that the logit of the DL model output explained a significant amount of variance in the mini mental state examination score, above and beyond the clinical indices including age, sex, and education duration (R² change = 0.341, F(1, 91) = 57.623, p < 0.001).

Conclusions

Lightweight DL can be clinically practicable for AD diagnosis by focusing on pathologically interpretable structural changes and offering image-based assessment of cognitive status.

背景：阿尔茨海默病（AD）发展过程中涉及的复杂大脑变化构成了一个高维非线性特征空间，深度学习（DL）分类/诊断可能比经典的非学习方法更有优势。然而，深度学习的实用性在医疗保健专业人员中仍然存在争议，主要是因为许多模型在计算上很昂贵，并且没有明确的可解释性。本研究旨在构建轻量级DL模型，揭示AD患者认知状态与大脑结构变化之间的关系。方法：利用从阿尔茨海默病神经影像学倡议数据库获得的数据，根据基线就诊时的t1加权磁共振图像，纳入418例AD患者和418例年龄匹配的认知正常（CN）受试者进行DL模型构建。通过结合群体卷积、全局池化和有效的信道关注，实现了轻量级设计。结果：该模型的准确率为90.6%，优于以往参数增加10倍以上的模型。闭塞图显示内侧颞区和丘脑是我们的模型区分AD和CN的主要原因，这与目前对病理轨迹的了解一致。层次回归进一步显示，DL模型输出的logit解释了迷你精神状态检查分数的显著差异，超出了年龄、性别和受教育程度等临床指标（R2变化= 0.341，F(1,91) = 57.623, p）。结论：轻量级DL可以通过关注病理可解释的结构变化和基于图像的认知状态评估来诊断AD。

{"title":"Association between cognitive status and structural brain changes in Alzheimer’s disease: Clinical implication of lightweight deep learning-aided diagnosis","authors":"Po-Hsuan Hsieh , Ya-Fang Chen , Ta-Fu Chen , Wen-Chau Wu , for the Alzheimer’s Disease Neuroimaging Initiative","doi":"10.1016/j.ejrad.2026.112678","DOIUrl":"10.1016/j.ejrad.2026.112678","url":null,"abstract":"<div><h3>Background</h3><div>The complex brain changes involved in Alzheimer’s disease (AD) development constitute a high-dimensional nonlinear feature space where deep learning (DL) classification/diagnosis may be advantageous over classical non-learning methods. However, the practicality of DL remains under debate among healthcare professionals, largely because many models are computationally expensive and operate without explicit interpretability. This study aimed to construct a lightweight DL model to disclose the association between cognitive status and structural brain changes in AD.</div></div><div><h3>Methods</h3><div>By using the data obtained from the Alzheimer’s Disease Neuroimaging Initiative database, 418 AD patients and 418 age-matched cognitively normal (CN) subjects were included for DL model construction based on their T1-weighted magnetic resonance images at baseline visit. A lightweight design was achieved by incorporating group convolution, global pooling, and efficient channel attention.</div></div><div><h3>Results</h3><div>The accuracy rate of our model was 90.6 %, competitive with previous models built with up-to-ten times more parameters. The occlusion maps showed that the medial temporal area and thalamus accounted the most for our model’s differentiation between AD and CN, in line with current knowledge of the pathological trajectory. Hierarchical regression further revealed that the logit of the DL model output explained a significant amount of variance in the mini mental state examination score, above and beyond the clinical indices including age, sex, and education duration (<em>R</em><sup>2</sup> change = 0.341, <em>F</em>(1, 91) = 57.623, <em>p</em> < 0.001).</div></div><div><h3>Conclusions</h3><div>Lightweight DL can be clinically practicable for AD diagnosis by focusing on pathologically interpretable structural changes and offering image-based assessment of cognitive status.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112678"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146009423","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

Dual-energy computed tomography angiography for lenticulostriate arteries: Stepwise optimization of virtual monochromatic imaging and kernel setting 双能计算机断层血管造影透镜状纹状动脉：逐步优化的虚拟单色成像和核设置

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-01-26 DOI: 10.1016/j.ejrad.2026.112694

Yangyang Yin , Qiuxia Wang , Zhihan Xu , Xudong Ai , Huan Liu , Haoyue Shao , Linhan Zhai , Shen Gui , Zhiwen Yang , Yu Chen , Baoyi Wang , Jing Zhang

Background

The lenticulostriate arteries (LSAs) are critical perforating vessels, but conventional CTA poorly depicts them. Dual-Energy CT offers Virtual Monochromatic Imaging and kernel adjustment, yet the optimal strategy for LSAs visualization remains unclear.

Materials and Methods

This retrospective study included 556 patients (59 [46–72] years; 310 males [55.8%]) who underwent DE-CTA between May 2023 to July 2024. Images were reconstructed Virtual Monochromatic Imaging at 40–90 KeV and multiple convolution kernels (Qr40, Hv36, Hv40, Hv44, and Hv49). Step 1 compared image quality across KeV levels, and Step 2 optimized kernels on the basis of the best KeV and compared these optimized reconstructions again. Quantitative metrics included CT value, SNR, CNR, edge rise distance, and edge rise slope; qualitative assessment and visible LSAs branch counts were performed.

Results

Step 1 demonstrated significant differences in CT value, and CNR across KeV levels (all p < 0.001), with 40 KeV yielding the highest CT value and CNR and the best subjective scores for artery branches visualization. Step 2 showed that sharper kernels significantly improved ERS and reduced ERD, with Hv49-40 KeV achieving the greatest edge sharpness (ERS, 10374.4 [7407.0–13341.9]; ERD, 0.10 [0.08–0.13]) and the highest number of visible LSAs branches (6.5 [2.5–10.5]; all adjusted p < . 001).

Conclusion

Hv49-40KeV on DE‑CTA of the head and neck substantially improves the visualization of arteries across different calibers, with a marked enhancement in the depiction of the LSAs, providing a solid technical foundation for the evaluation of LSAs with DE‑CT.

Relevance statement

This optimized Dual Energy-CT angiography reconstruction increases the utility of head and neck CTA by improving lenticulostriate artery visualization, thereby supporting routine evaluation of both large and small cerebral vessels in clinical practice.

背景：透镜状纹状动脉（LSAs）是重要的穿孔血管，但传统的CTA不能很好地描述它们。双能CT提供了虚拟单色成像和核调整，但LSAs可视化的最佳策略尚不清楚。材料与方法回顾性研究纳入556例于2023年5月至2024年7月行DE-CTA手术的患者（59例[46-72]岁，男性310例[55.8%]）。在40-90 KeV和多个卷积核（Qr40、Hv36、Hv40、Hv44和Hv49）下重建图像。步骤1比较不同KeV级别的图像质量，步骤2根据最佳KeV优化内核，并再次比较这些优化后的重建。定量指标包括CT值、信噪比、北比、边缘上升距离、边缘上升斜率；定性评估和可见LSAs分支计数。结果第1步显示了不同KeV水平的CT值和CNR的显著差异（均p <； 0.001）， 40 KeV产生最高的CT值和CNR，以及动脉分支可视化的最佳主观评分。步骤2显示，更锋利的核可以显著提高ERS并降低ERD， Hv49-40 KeV的边缘清晰度最大（ERS, 10374.4 [7407.0-13341.9]； ERD, 0.10[0.08-0.13]），可见lsa分支数最多(6.5[2.5-10.5]，均经过调整p <；001)。结论hv49 - 40kev在头颈部DE - CTA上显著改善了不同口径动脉的可视化，对lsa的描绘有明显增强，为DE - CT评价lsa提供了坚实的技术基础。该优化的双能量ct血管成像重建通过改善纹状体动脉的显像，提高了头颈部CTA的实用性，从而在临床实践中支持对大脑血管和小脑血管的常规评估。

{"title":"Dual-energy computed tomography angiography for lenticulostriate arteries: Stepwise optimization of virtual monochromatic imaging and kernel setting","authors":"Yangyang Yin , Qiuxia Wang , Zhihan Xu , Xudong Ai , Huan Liu , Haoyue Shao , Linhan Zhai , Shen Gui , Zhiwen Yang , Yu Chen , Baoyi Wang , Jing Zhang","doi":"10.1016/j.ejrad.2026.112694","DOIUrl":"10.1016/j.ejrad.2026.112694","url":null,"abstract":"<div><h3>Background</h3><div>The lenticulostriate arteries (LSAs) are critical perforating vessels, but conventional CTA poorly depicts them. Dual-Energy CT offers Virtual Monochromatic Imaging and kernel adjustment, yet the optimal strategy for LSAs visualization remains unclear.</div></div><div><h3>Materials and Methods</h3><div>This retrospective study included 556 patients (59 [46–72] years; 310 males [55.8%]) who underwent DE-CTA between May 2023 to July 2024. Images were reconstructed Virtual Monochromatic Imaging at 40–90 KeV and multiple convolution kernels (Qr40, Hv36, Hv40, Hv44, and Hv49). Step 1 compared image quality across KeV levels, and Step 2 optimized kernels on the basis of the best KeV and compared these optimized reconstructions again. Quantitative metrics included CT value, SNR, CNR, edge rise distance, and edge rise slope; qualitative assessment and visible LSAs branch counts were performed.</div></div><div><h3>Results</h3><div>Step 1 demonstrated significant differences in CT value, and CNR across KeV levels (all <em>p</em> < 0.001), with 40 KeV yielding the highest CT value and CNR and the best subjective scores for artery branches visualization. Step 2 showed that sharper kernels significantly improved ERS and reduced ERD, with Hv49-40 KeV achieving the greatest edge sharpness (ERS, 10374.4 [7407.0–13341.9]; ERD, 0.10 [0.08–0.13]) and the highest number of visible LSAs branches (6.5 [2.5–10.5]; all adjusted <em>p</em> < . 001).</div></div><div><h3>Conclusion</h3><div>Hv49-40KeV on DE‑CTA of the head and neck substantially improves the visualization of arteries across different calibers, with a marked enhancement in the depiction of the LSAs, providing a solid technical foundation for the evaluation of LSAs with DE‑CT.</div></div><div><h3>Relevance statement</h3><div>This optimized Dual Energy-CT angiography reconstruction increases the utility of head and neck CTA by improving lenticulostriate artery visualization, thereby supporting routine evaluation of both large and small cerebral vessels in clinical practice.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112694"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074525","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

Feasibility study of fast T2-weighted imaging with deep-learning reconstruction in volunteers and emergency patients with acute abdomen 基于深度学习重建的快速t2加权成像在志愿者和急腹症患者中的可行性研究

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-01-25 DOI: 10.1016/j.ejrad.2026.112697

Jia Xu , Liang Zhu , Wei Liu , Wenjing Liu , Yitong Lu , Jingjuan Liu , Chenxue Ma , Yifei Zhang , Xuan Wang , Feng Feng

Purpose

To evaluate the image quality and clinical utility of DLR-enhanced single-shot fast spin-echo (SSFSE) T2-weighted imaging (T2WI) for diagnosing acute abdominal conditions, compared to standard SSFSE and Periodically Rotated Overlapping Parallel Lines with Enhanced Reconstruction (PROPELLER) T2WI sequences.

Methods

This prospective single-institutional study enrolled 70 participants (35 healthy volunteers and 35 patients with acute abdominal pain). Abdominopelvic MRI were performed, including SSFSE and PROPELLER sequences. SSFSE images were reconstructed with and without deep-learning algorithm. Three radiologists independently evaluated image quality and target organ conditions. The Friedman test was used to compare image quality across sequences. The diagnostic performance for identifying disorders (e.g., cholecystitis, appendicitis, etc.) was assessed using the area under receiver operating characteristic curves (AUCs) and compared using the Delong method. Three pregnant women and one elderly patient who underwent SSFSE protocols only were also analyzed.

Results

SSFSE-DLR demonstrated significantly higher image quality and lower noise than SSFSE and PROPELLER across all imaging planes (p < 0.05). It exhibited fewer motion artifacts and superior clarity of the appendix, gallbladder, and common bile duct compared to PROPELLER (p < 0.05). SSFSE-DLR achieved higher diagnostic accuracy for common acute abdominopelvic disorders (AUCs: 0.977–1.0), compared to SSFSE and PROPELLER (AUCs: 0.887–1.00 and 0.585–0.953, respectively, p < 0.05). In vulnerable patients with fast protocol, SSFSE-DLR identified one appendicitis missed by ultrasound, enabling timely surgery.

Conclusion

SSFSE-DLR significantly improves image quality and diagnostic accuracy in healthy volunteers and patients with acute abdomen.

目的评价dlr增强单次快速自旋回波（SSFSE） t2加权成像（T2WI）诊断急腹症的图像质量和临床应用，并与标准SSFSE和周期性旋转重叠平行线增强重建（PROPELLER） T2WI序列进行比较。方法本前瞻性单机构研究纳入70名参与者（35名健康志愿者和35名急性腹痛患者）。进行了腹部骨盆MRI，包括SSFSE和PROPELLER序列。分别用深度学习算法和不使用深度学习算法重建SSFSE图像。三位放射科医生独立评估图像质量和靶器官状况。弗里德曼检验用于比较序列间的图像质量。采用受试者工作特征曲线下面积（auc）评估对疾病（如胆囊炎、阑尾炎等）的诊断性能，并采用Delong方法进行比较。还分析了仅接受SSFSE方案的3名孕妇和1名老年患者。结果SSFSE- dlr在各成像平面上的成像质量显著高于SSFSE和PROPELLER （p < 0.05）。与PROPELLER相比，它表现出更少的运动伪影和更清晰的阑尾、胆囊和胆总管（p < 0.05）。SSFSE- dlr对常见急性盆腔疾病的诊断准确率（auc: 0.977-1.0）高于SSFSE和PROPELLER （auc: 0.887-1.00和0.585-0.953,p < 0.05）。在快速方案的易感患者中，SSFSE-DLR识别出1例超声未发现的阑尾炎，及时手术。结论ssfse - dlr可显著提高健康志愿者和急腹症患者的图像质量和诊断准确率。

{"title":"Feasibility study of fast T2-weighted imaging with deep-learning reconstruction in volunteers and emergency patients with acute abdomen","authors":"Jia Xu , Liang Zhu , Wei Liu , Wenjing Liu , Yitong Lu , Jingjuan Liu , Chenxue Ma , Yifei Zhang , Xuan Wang , Feng Feng","doi":"10.1016/j.ejrad.2026.112697","DOIUrl":"10.1016/j.ejrad.2026.112697","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the image quality and clinical utility of DLR-enhanced single-shot fast spin-echo (SSFSE) T2-weighted imaging (T2WI) for diagnosing acute abdominal conditions, compared to standard SSFSE and Periodically Rotated Overlapping Parallel Lines with Enhanced Reconstruction (PROPELLER) T2WI sequences.</div></div><div><h3>Methods</h3><div>This prospective single-institutional study enrolled 70 participants (35 healthy volunteers and 35 patients with acute abdominal pain). Abdominopelvic MRI were performed, including SSFSE and PROPELLER sequences. SSFSE images were reconstructed with and without deep-learning algorithm. Three radiologists independently evaluated image quality and target organ conditions. The Friedman test was used to compare image quality across sequences. The diagnostic performance for identifying disorders (e.g., cholecystitis, appendicitis, etc.) was assessed using the area under receiver operating characteristic curves (AUCs) and compared using the Delong method. Three pregnant women and one elderly patient who underwent SSFSE protocols only were also analyzed.</div></div><div><h3>Results</h3><div>SSFSE-DLR demonstrated significantly higher image quality and lower noise than SSFSE and PROPELLER across all imaging planes (<em>p</em> < 0.05). It exhibited fewer motion artifacts and superior clarity of the appendix, gallbladder, and common bile duct compared to PROPELLER (<em>p</em> < 0.05). SSFSE-DLR achieved higher diagnostic accuracy for common acute abdominopelvic disorders (AUCs: 0.977–1.0), compared to SSFSE and PROPELLER (AUCs: 0.887–1.00 and 0.585–0.953, respectively, <em>p</em> < 0.05). In vulnerable patients with fast protocol, SSFSE-DLR identified one appendicitis missed by ultrasound, enabling timely surgery.</div></div><div><h3>Conclusion</h3><div>SSFSE-DLR significantly improves image quality and diagnostic accuracy in healthy volunteers and patients with acute abdomen.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112697"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074526","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

Enhancing liver diffusion-weighted imaging quality with correlation-weighted averaging: notable benefits in the left hepatic lobe 用相关加权平均增强肝脏弥散加权成像质量：在左肝叶有显著的改善

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-01-18 DOI: 10.1016/j.ejrad.2026.112680

Tetsuro Kaga , Yoshifumi Noda , Masashi Asano , Nobuyuki Kawai , Shingo Omata , Yukiko Takai , Satoshi Ido , Kimihiro Kajita , Abdelazim Elsayed Elhelaly , Hirohiko Imai , Hiroki Kato , Masayuki Matsuo

Purpose

To evaluate the feasibility of correlation-weighted averaging factor (CWAF) in liver diffusion-weighted imaging (DWI).

Materials and methods

This prospective study included 103 participants who underwent liver MRI. DWI were reconstructed using both original data (DWI_Original) and CWAF-corrected data (DWI_CWAF). Two radiologists independently assessed high-b DWI images for overall image quality, image noise, hepatic edge sharpness, and lesion conspicuity in the right and left lobes using five-point scales. Signal intensity ratio (SIR) and apparent diffusion coefficient (ADC) values were measured in four hepatic segments and in liver lesions, with lesion measurements analyzed separately for each lobe. These parameters were compared between the two image sets.

Results

The scores for overall image quality (P < 0.001), image noise (P < 0.001), and hepatic edge sharpness in the right lobe (P = 0.001) were higher in DWI_Original compared with DWI_CWAF. In contrast, hepatic edge sharpness (P < 0.001) and lesion conspicuity (P < 0.001) in the left lobe were superior in DWI_CWAF. Liver and lesion SIRs were higher in DWI_CWAF across all segments than in DWI_Original (P < 0.007). Liver ADC values were lower in DWI_CWAF than in DWI_Original in all segments (P < 0.001). Lesion ADC values were also lower in DWI_CWAF than in DWI_Original in the right lobe (P < 0.001) but were not different in the left lobe (P = 0.48).

Conclusion

CWAF improved hepatic edge sharpness and lesion conspicuity in the left lobe, although overall image quality was slightly reduced. ADC values were generally lower in DWI_CWAF than in DWI_Original.

目的探讨相关加权平均因子（CWAF）在肝脏弥散加权成像（DWI）中的可行性。材料和方法本前瞻性研究包括103名接受肝脏MRI检查的参与者。使用原始数据（DWIOriginal）和cwaf校正数据（dwwiwaf）重建DWI。两名放射科医生独立评估了高b DWI图像的整体图像质量、图像噪声、肝脏边缘清晰度以及左右叶病变的显著性。测量四个肝节段和肝病变的信号强度比（SIR）和表观扩散系数（ADC）值，并对每个肝叶的病变测量分别进行分析。将这些参数在两个图像集之间进行比较。结果DWIOriginal在整体图像质量（P < 0.001）、图像噪声（P < 0.001）和右叶肝边缘清晰度（P = 0.001）得分均高于dwwiwaf。相比之下，DWICWAF肝边缘清晰度（P < 0.001）和左叶病变显著性（P < 0.001）优于DWICWAF。dwiwaf各节段的肝脏和病变SIRs均高于DWIOriginal （P < 0.007）。肝ADC值在所有节段中均低于DWIOriginal组（P < 0.001）。右叶dwiwaf组病变ADC值也低于DWIOriginal组（P < 0.001），而左叶dwiwaf组病变ADC值差异无统计学意义（P = 0.48）。结论cwaf提高了肝脏边缘的清晰度和左叶病变的显著性，但整体图像质量略有下降。dwiwaf中的ADC值普遍低于DWIOriginal。

{"title":"Enhancing liver diffusion-weighted imaging quality with correlation-weighted averaging: notable benefits in the left hepatic lobe","authors":"Tetsuro Kaga , Yoshifumi Noda , Masashi Asano , Nobuyuki Kawai , Shingo Omata , Yukiko Takai , Satoshi Ido , Kimihiro Kajita , Abdelazim Elsayed Elhelaly , Hirohiko Imai , Hiroki Kato , Masayuki Matsuo","doi":"10.1016/j.ejrad.2026.112680","DOIUrl":"10.1016/j.ejrad.2026.112680","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the feasibility of correlation-weighted averaging factor (CWAF) in liver diffusion-weighted imaging (DWI).</div></div><div><h3>Materials and methods</h3><div>This prospective study included 103 participants who underwent liver MRI. DWI were reconstructed using both original data (DWI<sub>Original</sub>) and CWAF-corrected data (DWI<sub>CWAF</sub>). Two radiologists independently assessed high-<em>b</em> DWI images for overall image quality, image noise, hepatic edge sharpness, and lesion conspicuity in the right and left lobes using five-point scales. Signal intensity ratio (SIR) and apparent diffusion coefficient (ADC) values were measured in four hepatic segments and in liver lesions, with lesion measurements analyzed separately for each lobe. These parameters were compared between the two image sets.</div></div><div><h3>Results</h3><div>The scores for overall image quality (<em>P</em> < 0.001), image noise (<em>P</em> < 0.001), and hepatic edge sharpness in the right lobe (<em>P</em> = 0.001) were higher in DWI<sub>Original</sub> compared with DWI<sub>CWAF</sub>. In contrast, hepatic edge sharpness (<em>P</em> < 0.001) and lesion conspicuity (<em>P</em> < 0.001) in the left lobe were superior in DWI<sub>CWAF</sub>. Liver and lesion SIRs were higher in DWI<sub>CWAF</sub> across all segments than in DWI<sub>Original</sub> (<em>P</em> < 0.007). Liver ADC values were lower in DWI<sub>CWAF</sub> than in DWI<sub>Original</sub> in all segments (<em>P</em> < 0.001). Lesion ADC values were also lower in DWI<sub>CWAF</sub> than in DWI<sub>Original</sub> in the right lobe (<em>P</em> < 0.001) but were not different in the left lobe (<em>P</em> = 0.48).</div></div><div><h3>Conclusion</h3><div>CWAF improved hepatic edge sharpness and lesion conspicuity in the left lobe, although overall image quality was slightly reduced. ADC values were generally lower in DWI<sub>CWAF</sub> than in DWI<sub>Original</sub>.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112680"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074550","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

Deep learning outperformed radiomics based on MRI in the differentiation of sinonasal small round cell and non-small round cell malignant tumors 在鼻腔小圆细胞和非小圆细胞恶性肿瘤的鉴别上，深度学习优于基于MRI的放射组学

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-01-25 DOI: 10.1016/j.ejrad.2026.112700

Yuchen Wang , Xiaoxia Qu , Zheng Li , Dinggang Shen , Junfang Xian

Objective

Differentiating sinonasal small round cell malignant tumors (SRCMTs) from non-SRCMTs is challenging due to overlapping MRI features. This study aimed to compare the diagnostic performance of deep learning and radiomics models for preoperative MRI-based classification.

Methods

We retrospectively analyzed 325 patients with pathologically confirmed sinonasal malignancies (163 SRCMTs and 162 non-SRCMTs). Each patient underwent T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and contrast-enhanced T1-weighted imaging (CE-T1WI). Tumors were manually segmented. Five convolutional neural networks (CNNs)—ResNet-18, ResNet-34, ResNet-50, VGG13, and VGG16—were trained on each sequence. For radiomics, 1200 features were extracted per sequence, and multiple machine learning classifiers were trained. Model performance was assessed by area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The DeLong test was used to compare AUCs between models, with statistical significance set at P < 0.05.

Results

The CE-T1WI-based ResNet-34 model achieved the best performance, with an average AUC of 0.830, the accuracy of 0.755, sensitivity of 0.918, specificity of 0.592, PPV of 0.692, and NPV of 0.879. The corresponding CE-T1WI-based radiomics model using a support vector machine yielded an AUC of 0.758 (accuracy = 0.755, sensitivity = 0.840, specificity = 0.667, PPV = 0.724, NPV = 0.800). On the independent test cohort, ResNet-34 showed numerically higher discriminative performance than the radiomics model, although this difference did not reach statistical significance. For T1WI and T2WI, deep learning and radiomics models demonstrated broadly comparable performance.

Conclusions

A CE-T1WI-based ResNet-34 network provided high diagnostic efficacy, and in our cohort deep learning models achieved numerically higher comparable performance to MRI-based radiomics models for differentiating SRCMTs from non-SRCMTs.

目的鼻腔小圆细胞恶性肿瘤（SRCMTs）与非SRCMTs的MRI特征重叠，具有挑战性。本研究旨在比较深度学习和放射组学模型在术前mri分类中的诊断性能。方法回顾性分析325例经病理证实的鼻窦恶性肿瘤患者（163例SRCMTs和162例非SRCMTs）。每位患者均行t1加权成像（T1WI）、t2加权成像（T2WI）和对比增强t1加权成像（CE-T1WI）检查。手工分割肿瘤。在每个序列上训练5个卷积神经网络（cnn）——resnet -18、ResNet-34、ResNet-50、VGG13和vgg16。对于放射组学，每个序列提取1200个特征，并训练多个机器学习分类器。通过受试者工作特征曲线下面积（AUC）、准确性、敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV）评估模型的性能。模型间auc比较采用DeLong检验，差异有统计学意义P <； 0.05。结果基于ce - t1wi的ResNet-34模型表现最佳，平均AUC为0.830，准确率为0.755，灵敏度为0.918，特异性为0.592，PPV为0.692，NPV为0.879。相应的基于ce - t1wi的支持向量机放射组学模型的AUC为0.758（准确性= 0.755，灵敏度= 0.840，特异性= 0.667,PPV = 0.724, NPV = 0.800）。在独立测试队列中，ResNet-34在数值上表现出比放射组学模型更高的判别性能，尽管这种差异没有达到统计学意义。对于T1WI和T2WI，深度学习和放射组学模型表现出大致相当的性能。结论基于ce - t1wi的ResNet-34网络具有较高的诊断效能，在我们的队列中，深度学习模型在区分srcmt和非srcmt方面取得了比基于mri的放射组学模型更高的数值可比性能。

{"title":"Deep learning outperformed radiomics based on MRI in the differentiation of sinonasal small round cell and non-small round cell malignant tumors","authors":"Yuchen Wang , Xiaoxia Qu , Zheng Li , Dinggang Shen , Junfang Xian","doi":"10.1016/j.ejrad.2026.112700","DOIUrl":"10.1016/j.ejrad.2026.112700","url":null,"abstract":"<div><h3>Objective</h3><div>Differentiating sinonasal small round cell malignant tumors (SRCMTs) from non-SRCMTs is challenging due to overlapping MRI features. This study aimed to compare the diagnostic performance of deep learning and radiomics models for preoperative MRI-based classification.</div></div><div><h3>Methods</h3><div>We retrospectively analyzed 325 patients with pathologically confirmed sinonasal malignancies (163 SRCMTs and 162 non-SRCMTs). Each patient underwent T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and contrast-enhanced T1-weighted imaging (CE-T1WI). Tumors were manually segmented. Five convolutional neural networks (CNNs)—ResNet-18, ResNet-34, ResNet-50, VGG13, and VGG16—were trained on each sequence. For radiomics, 1200 features were extracted per sequence, and multiple machine learning classifiers were trained. Model performance was assessed by area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The DeLong test was used to compare AUCs between models, with statistical significance set at P < 0.05.</div></div><div><h3>Results</h3><div>The CE-T1WI-based ResNet-34 model achieved the best performance, with an average AUC of 0.830, the accuracy of 0.755, sensitivity of 0.918, specificity of 0.592, PPV of 0.692, and NPV of 0.879. The corresponding CE-T1WI-based radiomics model using a support vector machine yielded an AUC of 0.758 (accuracy = 0.755, sensitivity = 0.840, specificity = 0.667, PPV = 0.724, NPV = 0.800). On the independent test cohort, ResNet-34 showed numerically higher discriminative performance than the radiomics model, although this difference did not reach statistical significance. For T1WI and T2WI, deep learning and radiomics models demonstrated broadly comparable performance.</div></div><div><h3>Conclusions</h3><div>A CE-T1WI-based ResNet-34 network provided high diagnostic efficacy, and in our cohort deep learning models achieved numerically higher comparable performance to MRI-based radiomics models for differentiating SRCMTs from non-SRCMTs.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112700"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074551","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

Application of gastric ultrasound in evaluating the efficacy of preoperative neoadjuvant therapy for gastric carcinoma 胃超声在评价胃癌术前新辅助治疗效果中的应用

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2025-11-17 DOI: 10.1016/j.ejrad.2025.112546

Xiaohong Deng, Zhongshi Du, Zhougui Wu, Weiqin Huang, Yaoqin Wang, Lina Tang

Purpose

To evaluate the efficacy of gastric ultrasound in the neoadjuvant treatment of gastric carcinoma.

Methods

This retrospective study involved patients with locally advanced gastric carcinoma treated with preoperative neoadjuvant therapy (NAT) and surgery at our institute from April 2022 to June 2024. Ultrasound examinations were conducted at two time points: pre-NAT and within a week pre-surgery. We calculated changes in ultrasound measurements, including length and thickness, as well as the rate of these changes. The tumor regression grade (TRG) obtained from surgical pathology was used as the gold standard. We performed univariate and binary logistic regression analysis to explore potential risk factors associated with TRG. Additionally, we assessed the correlation between preoperative ultrasound measurements and postoperative gross specimen measurements.

Results

The study included 126 patients and identified 126 lesions, achieving a detection rate of 100 %. Changes in ultrasound thickness between the two time points served as a protective factor for postoperative TRG staging (Odds Ratio = 0.012; P < 0.001). A higher rate of thickness change was associated with a lower TRG stage, indicating a more effective NAT. A significant positive correlation was observed between preoperative ultrasound length and postoperative lesion length (P < 0.001).

Conclusion

Gastric ultrasound is a viable method to evaluate the efficacy of preoperative NAT in patients with gastric carcinoma.

目的探讨胃超声在胃癌新辅助治疗中的应用价值。方法回顾性研究2022年4月至2024年6月在我院行术前新辅助治疗（NAT）及手术治疗的局部进展期胃癌患者。超声检查分别在术前和术前一周内两个时间点进行。我们计算了超声测量的变化，包括长度和厚度，以及这些变化的速率。手术病理所得肿瘤消退等级（TRG）作为金标准。我们采用单因素和二元逻辑回归分析来探讨与TRG相关的潜在危险因素。此外，我们评估了术前超声测量和术后大体标本测量之间的相关性。结果本研究纳入126例患者，发现126个病变，检出率100%。两个时间点间超声厚度的变化是术后TRG分期的保护因素（优势比= 0.012;P < 0.001）。厚度变化率越高，TRG分期越低，表明NAT越有效。术前超声长度与术后病变长度呈显著正相关（P < 0.001）。结论胃超声检查是评价胃癌患者术前NAT治疗效果的一种可行方法。

{"title":"Application of gastric ultrasound in evaluating the efficacy of preoperative neoadjuvant therapy for gastric carcinoma","authors":"Xiaohong Deng, Zhongshi Du, Zhougui Wu, Weiqin Huang, Yaoqin Wang, Lina Tang","doi":"10.1016/j.ejrad.2025.112546","DOIUrl":"10.1016/j.ejrad.2025.112546","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the efficacy of gastric ultrasound in the neoadjuvant treatment of gastric carcinoma.</div></div><div><h3>Methods</h3><div>This retrospective study involved patients with locally advanced gastric carcinoma treated with preoperative neoadjuvant therapy (NAT) and surgery atour institute fromApril 2022 toJune 2024.Ultrasound examinationswere conductedat two time points: pre-NAT and within a week pre-surgery. We calculated changes in ultrasound measurements, including length and thickness, as wellas the rate of these changes. The tumor regression grade (TRG) obtained from surgical pathology was used as the gold standard. We performed univariateand binary logisticregression analysis to explore potentialrisk factorsassociated with TRG. Additionally, we assessed thecorrelation between preoperative ultrasound measurementsand postoperative gross specimen measurements.</div></div><div><h3>Results</h3><div>The study included 126 patients and identified 126 lesions, achieving a detectionrate of 100 %. Changes in ultrasound thickness betweenthe two time points served asa protective factor for postoperative TRG staging (Odds Ratio = 0.012; <em>P</em> < 0.001). Ahigher rate of thicknesschange was associated with a lower TRG stage, indicating a more effective NAT. Asignificant positive correlation was observed between preoperative ultrasound length and postoperative lesion length (P < 0.001).</div></div><div><h3>Conclusion</h3><div>Gastric ultrasound is a viable methodto evaluate the efficacy of preoperative NAT in patients withgastric carcinoma.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112546"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146074620","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

Performance analysis of liver segmentation using nn-UNet TotalSegmentator: Focus on atypical livers, pathologies, and variants 使用nn-UNet TotalSegmentator进行肝脏分割的性能分析：关注非典型肝脏、病理和变异

IF 3.3 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

European Journal of Radiology

Pub Date : 2026-03-01 Epub Date: 2026-01-14 DOI: 10.1016/j.ejrad.2026.112674

Joy-Marie Kleiß , Sebastian Arndt , Lisa Sommerfeld , Maximilian Schmidt MD , Florian Putz , Teresa Graetz , Leonard Stepansky , Kaan Türkan , Simon Mayr , Michael Uder , Matthias S May

Rationale and Objectives

This study evaluates the accuracy of the nn-UNet TotalSegmentator (TS) by Wasserthal et al. (2023) in segmenting atypical livers with pathologies and variants in CT scans.

Materials and Methods

CT scans were retrospectively collected from our RIS and divided into two cohorts: a reference group (67 healthy livers) and a study group (55 scans across eleven pathology and variant subgroups). TS performed automatic segmentation for all groups. For reference, the images were then manually segmented, with corrections reviewed by two radiologists. Accuracy was assessed using Dice similarity score, Hausdorff distance (HD), mean surface distance (MSD), volume difference, and clinical ratings.

Results

Automatic segmentation underestimated liver volume by a mean of 48.11 ml (3.1%) in the reference group and overestimated it in 84% of study group cases by 79.09 ml (4%).

The average Dice score was 0.980 ± 0.007 for the reference group and 0.933 ± 0.113 for the study group. Hepatomegaly achieved the highest score (0.979 ± 0.006), Polycystic liver disease (PLD) the lowest (0.656 ± 0.230). Cirrhosis with Ascites, Beavertail, and PLD had significantly lower Dice scores than the reference group. Clinical ratings were often lower than Dice scores suggested, especially in Beavertail, Cirrhosis with Ascites, Ablation defects, Metastases, and Hemihepatectomy.

Conclusion

TS performs excellently on healthy and well on most pathological livers. Despite high Dice scores in many pathological cases, clinical ratings reveal limitations. Clinical evaluation remains essential. Inclusion of PLD and Beavertail cases in training data may reduce bias and improve performance.

理由和目的本研究评估了Wasserthal等人（2023）使用nn-UNet TotalSegmentator （TS）对CT扫描中病理和变异的非典型肝脏进行分割的准确性。材料和方法回顾性收集RIS的sct扫描，并将其分为两组：参照组（67个健康肝脏）和研究组（11个病理和变异亚组的55个扫描）。TS对所有组进行自动分割。作为参考，这些图像随后被手动分割，并由两名放射科医生进行校正。使用Dice相似度评分、Hausdorff距离（HD）、平均表面距离（MSD）、体积差和临床评分来评估准确性。结果自动分割在参照组中平均低估了48.11 ml(3.1%)，在84%的研究组中平均高估了79.09 ml（4%）。参照组的平均Dice评分为0.980±0.007，研究组的平均Dice评分为0.933±0.113。肝肿大评分最高（0.979±0.006），多囊性肝病评分最低（0.656±0.230）。肝硬化合并腹水、海狸尾和PLD的Dice评分明显低于对照组。临床评分通常低于Dice评分，特别是在海狸尾、肝硬化合并腹水、消融缺陷、转移和半肝切除术中。结论ts对健康肝脏治疗效果良好，对多数病理肝脏治疗效果良好。尽管在许多病理病例中Dice得分很高，但临床评分显示出局限性。临床评估仍然是必要的。在训练数据中加入PLD和Beavertail案例可以减少偏差并提高性能。

{"title":"Performance analysis of liver segmentation using nn-UNet TotalSegmentator: Focus on atypical livers, pathologies, and variants","authors":"Joy-Marie Kleiß , Sebastian Arndt , Lisa Sommerfeld , Maximilian Schmidt MD , Florian Putz , Teresa Graetz , Leonard Stepansky , Kaan Türkan , Simon Mayr , Michael Uder , Matthias S May","doi":"10.1016/j.ejrad.2026.112674","DOIUrl":"10.1016/j.ejrad.2026.112674","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>This study evaluates the accuracy of the nn-UNet TotalSegmentator (TS) by Wasserthal et al. (2023) in segmenting atypical livers with pathologies and variants in CT scans.</div></div><div><h3>Materials and Methods</h3><div>CT scans were retrospectively collected from our RIS and divided into two cohorts: a reference group (67 healthy livers) and a study group (55 scans across eleven pathology and variant subgroups). TS performed automatic segmentation for all groups. For reference, the images were then manually segmented, with corrections reviewed by two radiologists. Accuracy was assessed using Dice similarity score, Hausdorff distance (HD), mean surface distance (MSD), volume difference, and clinical ratings.</div></div><div><h3>Results</h3><div>Automatic segmentation underestimated liver volume by a mean of 48.11 ml (3.1%) in the reference group and overestimated it in 84% of study group cases by 79.09 ml (4%).</div><div>The average Dice score was 0.980 ± 0.007 for the reference group and 0.933 ± 0.113 for the study group. Hepatomegaly achieved the highest score (0.979 ± 0.006), Polycystic liver disease (PLD) the lowest (0.656 ± 0.230). Cirrhosis with Ascites, Beavertail, and PLD had significantly lower Dice scores than the reference group. Clinical ratings were often lower than Dice scores suggested, especially in Beavertail, Cirrhosis with Ascites, Ablation defects, Metastases, and Hemihepatectomy.</div></div><div><h3>Conclusion</h3><div>TS performs excellently on healthy and well on most pathological livers. Despite high Dice scores in many pathological cases, clinical ratings reveal limitations. Clinical evaluation remains essential. Inclusion of PLD and Beavertail cases in training data may reduce bias and improve performance.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"196 ","pages":"Article 112674"},"PeriodicalIF":3.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146035336","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