Academic Radiology最新文献_第9页

Individualizing Radiation Risk in Lung Cancer Screening: Towards Precision Dosimetry 肺癌筛查中的个体化辐射风险：迈向精确剂量学。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.10.003

Harleen Kaur, Ritu R. Gill MD, MPH

引用次数: 0

Virtual Clinical Shadowing: The Future of Medical Student Education Through Telemedicine 虚拟临床阴影：医学生远程医疗教育的未来。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.05.030

Minahil Cheema , Omer A. Awan MD, MPH, CIIP

引用次数: 0

Deep Learning Radiomic Signature Predicts the Overall Survival of Patients with Lung Adenocarcinoma by Reflecting the Tumor Heterogeneity and Microenvironment 深度学习放射学特征通过反映肿瘤异质性和微环境来预测肺腺癌患者的总生存期。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.09.033

Chunlei Dai , Bo Huang , Zhe Yu , Jingwei Xu , Jian Li , Jian Yang

Rationale and Objectives

The need for prediction of overall survival (OS) in patients with lung adenocarcinoma (LUAD) has been increasingly recognized. We aimed to generate a computed tomography-derived radiomic signature for predicting prognosis in LUAD patients, and then explored the relationship between radiomic features and tumor heterogeneity and microenvironment.

Materials and Methods

Data of 306 eligible LUAD patients from three institutions were obtained between January 2019 and January 2024. The mainstream Residual Network 50 (ResNet50) was used to develop an image-based deep learning radiomic signature (DLRS). We developed a clinical model and calculated the conventional radiomics score using pyradiomics package. An external cohort from a public database called The Cancer Imaging Archive was obtained for further validation. We performed the time-dependent receiver operator characteristic curve to assess the performance of the models. We divided the whole dataset into high and low-score groups with the help of the DLRS. The differences in tumor heterogeneity and microenvironment between different score groups were investigated using the sequencing data from the corresponding LUAD cohort from the Cancer Genome Atlas.

Results

In the test cohort, the DLRS outperformed the conventional radiomics score and clinical model, with the area under the curves (95%CI) for 1, 3, and 5-year OS of 0.912 (0.881–0.952), 0.851 (0.824–0.901), and 0.841 (0.807–0.878), respectively. Significant differences in survival time were observed between different groups stratified by this signature. It showed great discrimination, calibration, and clinical utility (all p<0.05). Distinct gene expression patterns were identified. The tumor heterogeneity and microenvironment significantly varied between different score groups.

Conclusion

The DLRS could effectively predict the prognosis of LUAD patients by reflecting the tumor heterogeneity and microenvironment.

理由和目的：预测肺腺癌（LUAD）患者总生存期（OS）的必要性已经越来越被认识到。我们的目的是生成一个计算机断层扫描衍生的放射组学特征来预测LUAD患者的预后，然后探索放射组学特征与肿瘤异质性和微环境之间的关系。材料与方法：2019年1月至2024年1月，来自三家机构的306例符合条件的LUAD患者的数据。使用主流的残余网络50 （ResNet50）来开发基于图像的深度学习放射特征（DLRS）。我们开发了一个临床模型，并使用放射组学包计算常规放射组学评分。为了进一步验证，我们从一个名为“癌症影像档案”的公共数据库中获得了一个外部队列。我们进行了随时间变化的接收算子特征曲线来评估模型的性能。在DLRS的帮助下，我们将整个数据集分为高分组和低分组。使用来自Cancer Genome Atlas的相应LUAD队列的测序数据，研究不同评分组之间肿瘤异质性和微环境的差异。结果：在测试队列中，DLRS优于常规放射组学评分和临床模型，1、3和5年OS的曲线下面积（95%CI）分别为0.912（0.881-0.952）、0.851（0.824-0.901）和0.841（0.807-0.878）。通过该特征分层的不同组之间观察到生存时间的显著差异。结论：DLRS可通过反映肿瘤异质性和微环境，有效预测LUAD患者的预后。

{"title":"Deep Learning Radiomic Signature Predicts the Overall Survival of Patients with Lung Adenocarcinoma by Reflecting the Tumor Heterogeneity and Microenvironment","authors":"Chunlei Dai , Bo Huang , Zhe Yu , Jingwei Xu , Jian Li , Jian Yang","doi":"10.1016/j.acra.2025.09.033","DOIUrl":"10.1016/j.acra.2025.09.033","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>The need for prediction of overall survival (OS) in patients with lung adenocarcinoma (LUAD) has been increasingly recognized. We aimed to generate a computed tomography-derived radiomic signature for predicting prognosis in LUAD patients, and then explored the relationship between radiomic features and tumor heterogeneity and microenvironment.</div></div><div><h3>Materials and Methods</h3><div>Data of 306 eligible LUAD patients from three institutions were obtained between January 2019 and January 2024. The mainstream Residual Network 50 (ResNet50) was used to develop an image-based deep learning radiomic signature (DLRS). We developed a clinical model and calculated the conventional radiomics score using pyradiomics package. An external cohort from a public database called The Cancer Imaging Archive was obtained for further validation. We performed the time-dependent receiver operator characteristic curve to assess the performance of the models. We divided the whole dataset into high and low-score groups with the help of the DLRS. The differences in tumor heterogeneity and microenvironment between different score groups were investigated using the sequencing data from the corresponding LUAD cohort from the Cancer Genome Atlas.</div></div><div><h3>Results</h3><div>In the test cohort, the DLRS outperformed the conventional radiomics score and clinical model, with the area under the curves (95%CI) for 1, 3, and 5-year OS of 0.912 (0.881–0.952), 0.851 (0.824–0.901), and 0.841 (0.807–0.878), respectively. Significant differences in survival time were observed between different groups stratified by this signature. It showed great discrimination, calibration, and clinical utility (all p<0.05). Distinct gene expression patterns were identified. The tumor heterogeneity and microenvironment significantly varied between different score groups.</div></div><div><h3>Conclusion</h3><div>The DLRS could effectively predict the prognosis of LUAD patients by reflecting the tumor heterogeneity and microenvironment.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"33 1","pages":"Pages 224-235"},"PeriodicalIF":3.9,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145276538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Efficacy and Safety of Spontaneous Portosystemic Shunts Embolization for Hepatic Encephalopathy: A Meta-analysis 自发性门静脉分流栓塞治疗肝性脑病的疗效和安全性：一项荟萃分析。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.09.049

Xing Wang , Zhengyu Wang , Bohan Luo , Yong Lv , Guohong Han

Background & Aims

Spontaneous portosystemic shunt (SPSS) embolization represents a promising intervention for refractory hepatic encephalopathy (HE). This systematic review and meta-analysis evaluate the efficacy and safety of SPSS embolization in cirrhotic patients without transjugular intrahepatic portosystemic shunts (TIPS).

Methods

We systematically searched PubMed, Web of Science, Embase, and the Cochrane Library through June 12, 2024 to identify studies investigating SPSS embolization for HE. Meta-analysis was performed using fixed-effect or random-effects models to calculate clinical success (defined as HE remission), procedural success rates, and complication frequencies.

Results

Analysis of 10 retrospective studies encompassing 289 cirrhotic patients yielded the following pooled outcomes: hepatic encephalopathy remission rate of 83.1% (95% CI: 70.4%–93.1%), procedural success rate of 99.8% (95% CI: 98.3%–100%), and long-term adverse event rate of 42.9% (95% CI: 34.7%–51.4%). The predominant long-term complications included ascites (51.6% of complications), variceal progression (23.4%), and thrombosis (8.0%), while primary procedure-related adverse reactions were infection (37%) and fever (29%). Subgroup analyses demonstrated no statistically significant effect of etiology (p = 0.788) or shunt type (p = 0.271) on disease remission rates, but revealed significant differences between surgical approaches (p<0.001), with balloon-occluded retrograde transvenous obliteration (BRTO) showing the highest efficacy (97.4%–100%).

Conclusion

SPSS embolization demonstrates both high efficacy for refractory hepatic encephalopathy (83.1% remission rate) and exceptional procedural success (99.8%). Despite substantial long-term complications (42.9%, predominantly portal hypertension sequelae), current evidence from predominantly retrospective studies supports its consideration as a therapeutic option. Technique selection should be individualized pending further validation of BRTO's superiority.

背景与目的：自发性门系统分流（SPSS）栓塞是治疗难治性肝性脑病（HE）的一种有希望的干预手段。本系统综述和荟萃分析评估了SPSS栓塞治疗肝硬化患者无经颈静脉肝内门静脉系统分流术（TIPS）的有效性和安全性。方法：我们系统地检索PubMed、Web of Science、Embase和Cochrane Library，检索时间截止到2024年6月12日，以确定调查SPSS栓塞治疗HE的研究。采用固定效应或随机效应模型进行meta分析，计算临床成功率（定义为HE缓解）、手术成功率和并发症频率。结果：对包含289例肝硬化患者的10项回顾性研究的分析得出了以下汇总结果：肝性脑病缓解率为83.1% (95% CI: 70.4%-93.1%)，手术成功率为99.8% (95% CI: 98.3%-100%)，长期不良事件发生率为42.9% （95% CI: 34.7%-51.4%）。主要的长期并发症包括腹水（51.6%的并发症）、静脉曲张进展（23.4%）和血栓形成（8.0%），而主要的手术相关不良反应是感染（37%）和发烧（29%）。亚组分析显示，病因学（p=0.788）或分流管类型（p=0.271）对疾病缓解率没有统计学意义，但不同手术入路之间存在显著差异(p)。结论：SPSS栓塞治疗难治性肝性脑病疗效高（缓解率83.1%），手术成功率高（99.8%）。尽管有大量的长期并发症（42.9%，主要是门脉高压后遗症），目前主要来自回顾性研究的证据支持将其作为一种治疗选择。在进一步验证BRTO的优势之前，技术选择应该个性化。

{"title":"Efficacy and Safety of Spontaneous Portosystemic Shunts Embolization for Hepatic Encephalopathy: A Meta-analysis","authors":"Xing Wang , Zhengyu Wang , Bohan Luo , Yong Lv , Guohong Han","doi":"10.1016/j.acra.2025.09.049","DOIUrl":"10.1016/j.acra.2025.09.049","url":null,"abstract":"<div><h3>Background & Aims</h3><div>Spontaneous portosystemic shunt (SPSS) embolization represents a promising intervention for refractory hepatic encephalopathy (HE). This systematic review and meta-analysis evaluate the efficacy and safety of SPSS embolization in cirrhotic patients without transjugular intrahepatic portosystemic shunts (TIPS).</div></div><div><h3>Methods</h3><div>We systematically searched PubMed, Web of Science, Embase, and the Cochrane Library through June 12, 2024 to identify studies investigating SPSS embolization for HE. Meta-analysis was performed using fixed-effect or random-effects models to calculate clinical success (defined as HE remission), procedural success rates, and complication frequencies.</div></div><div><h3>Results</h3><div>Analysis of 10 retrospective studies encompassing 289 cirrhotic patients yielded the following pooled outcomes: hepatic encephalopathy remission rate of 83.1% (95% CI: 70.4%–93.1%), procedural success rate of 99.8% (95% CI: 98.3%–100%), and long-term adverse event rate of 42.9% (95% CI: 34.7%–51.4%). The predominant long-term complications included ascites (51.6% of complications), variceal progression (23.4%), and thrombosis (8.0%), while primary procedure-related adverse reactions were infection (37%) and fever (29%). Subgroup analyses demonstrated no statistically significant effect of etiology (p=0.788) or shunt type (p=0.271) on disease remission rates, but revealed significant differences between surgical approaches (p<0.001), with balloon-occluded retrograde transvenous obliteration (BRTO) showing the highest efficacy (97.4%–100%).</div></div><div><h3>Conclusion</h3><div>SPSS embolization demonstrates both high efficacy for refractory hepatic encephalopathy (83.1% remission rate) and exceptional procedural success (99.8%). Despite substantial long-term complications (42.9%, predominantly portal hypertension sequelae), current evidence from predominantly retrospective studies supports its consideration as a therapeutic option. Technique selection should be individualized pending further validation of BRTO's superiority.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"33 1","pages":"Pages 147-156"},"PeriodicalIF":3.9,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145402275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Prognostic Value of Pseudotime from Texture Parameters of [18F]fluorodeoxyglucose PET/CT in Resectable Pancreatic Ductal Adenocarcinoma [18F]氟脱氧葡萄糖PET/CT纹理参数伪时间对可切除胰导管腺癌的预后价值。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.10.016

Wonseok Whi MD, PhD , Seung Hyup Hyun MD, PhD , Hyunjong Lee MD, PhD , Jeong Il Yu MD, PhD , Kwang Hyuck Lee MD, PhD , Jin Seok Heo MD, PhD , Joon Young Choi MD, PhD

Rationale and Objectives

This study evaluates the prognostic value of pseudotime derived from radiomics texture parameters on [¹⁸F]fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) images of resectable pancreatic ductal adenocarcinoma (PDAC) patients.

Materials and Methods

We retrospectively analyzed data from 711 patients who underwent FDG PET/CT before surgery. We extracted 56 radiomics features and calculated the pseudotime, a continuous metric estimating disease progression, using the Phenopath algorithm. Clinicopathologic features and other conventional PET parameters were also obtained. Correlation analyses were performed between the conventional PET parameters and pseudotime, and survival analysis was performed according to the clinicopathologic variables.

Results

Correlation analysis revealed that pseudotime correlates weakly with SUVmax and SUVmean and strongly with the metabolic tumor volume (MTV) and total lesion glycolysis (TLG). A multivariate survival analysis revealed that pseudotime is an independent predictor of disease-free survival (hazard ratio [HR] = 1.67, p < 0.001, c-index = 0.59), showing stronger prognostic performance than MTV (HR = 1.48, p = 0.009, c = 0.57) and TLG (HR = 1.39, p = 0.03, c = 0.56). When pseudotime was combined with TLG for risk stratification, the integrated model demonstrated the strongest prognostic separation among subgroups. Texture parameters related to homogeneity correlated positively with pseudotime, and those representing heterogeneity showed mixed correlations, highlighting the complexity of tumor biology.

Conclusion

Our findings indicate that pseudotime is a meaningful prognostic biomarker in resectable PDAC patients undergoing surgery, with stronger predictive power than established metabolic parameters. Stratification performance improved when it was combined with conventional markers.

基本原理和目的：本研究评估可切除胰导管腺癌（PDAC）患者的放射组学纹理参数所得伪时间的预后价值。材料和方法：我们回顾性分析了711例术前接受FDG PET/CT检查的患者的资料。我们提取了56个放射组学特征，并使用Phenopath算法计算了假时间，这是一种估计疾病进展的连续度量。同时获得临床病理特征及其他常规PET参数。常规PET参数与假时间进行相关性分析，根据临床病理变量进行生存分析。结果：相关分析显示，假时间与SUVmax和SUVmean相关性较弱，与代谢肿瘤体积（MTV）和病变总糖酵解（TLG）相关性较强。多变量生存分析显示，假时间是无病生存的独立预测因子（风险比[HR] = 1.67, p < 0.001, c-index = 0.59），比MTV （HR = 1.48, p = 0.009, c = 0.57）和TLG （HR = 1.39, p = 0.03, c = 0.56）表现出更强的预后效果。当假时间与TLG相结合进行风险分层时，综合模型显示亚组之间的预后分离最强。同质性相关的纹理参数与伪时间呈正相关，异质性相关的纹理参数与伪时间呈混合相关，凸显了肿瘤生物学的复杂性。结论：我们的研究结果表明，假时间是可切除的PDAC患者接受手术的有意义的预后生物标志物，比既定的代谢参数具有更强的预测能力。与常规标记物联合使用可提高分层效果。

{"title":"Prognostic Value of Pseudotime from Texture Parameters of [18F]fluorodeoxyglucose PET/CT in Resectable Pancreatic Ductal Adenocarcinoma","authors":"Wonseok Whi MD, PhD , Seung Hyup Hyun MD, PhD , Hyunjong Lee MD, PhD , Jeong Il Yu MD, PhD , Kwang Hyuck Lee MD, PhD , Jin Seok Heo MD, PhD , Joon Young Choi MD, PhD","doi":"10.1016/j.acra.2025.10.016","DOIUrl":"10.1016/j.acra.2025.10.016","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>This study evaluates the prognostic value of pseudotime derived from radiomics texture parameters on [<sup>18</sup>F]fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) images of resectable pancreatic ductal adenocarcinoma (PDAC) patients.</div></div><div><h3>Materials and Methods</h3><div>We retrospectively analyzed data from 711 patients who underwent FDG PET/CT before surgery. We extracted 56 radiomics features and calculated the pseudotime, a continuous metric estimating disease progression, using the Phenopath algorithm. Clinicopathologic features and other conventional PET parameters were also obtained. Correlation analyses were performed between the conventional PET parameters and pseudotime, and survival analysis was performed according to the clinicopathologic variables.</div></div><div><h3>Results</h3><div>Correlation analysis revealed that pseudotime correlates weakly with SUVmax and SUVmean and strongly with the metabolic tumor volume (MTV) and total lesion glycolysis (TLG). A multivariate survival analysis revealed that pseudotime is an independent predictor of disease-free survival (hazard ratio [HR] = 1.67, p < 0.001, c-index = 0.59), showing stronger prognostic performance than MTV (HR = 1.48, p = 0.009, c = 0.57) and TLG (HR = 1.39, p = 0.03, c = 0.56). When pseudotime was combined with TLG for risk stratification, the integrated model demonstrated the strongest prognostic separation among subgroups. Texture parameters related to homogeneity correlated positively with pseudotime, and those representing heterogeneity showed mixed correlations, highlighting the complexity of tumor biology.</div></div><div><h3>Conclusion</h3><div>Our findings indicate that pseudotime is a meaningful prognostic biomarker in resectable PDAC patients undergoing surgery, with stronger predictive power than established metabolic parameters. Stratification performance improved when it was combined with conventional markers.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"33 1","pages":"Pages 189-200"},"PeriodicalIF":3.9,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145394975","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 and Radiologic Contextualization of Automated BAC Quantification: A Commentary 自动BAC定量的临床和放射背景：评论。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.08.057

Ahmet Gürkan Erdemir MD , Gamze Durhan Assoc. Prof.

引用次数: 0

Dual-Vessel Microcirculation Imaging in Discriminating Non-Hodgkin Lymphoma Subtypes Using Super-Resolution Ultrasound: An Exploring Study 超分辨率超声双血管微循环成像鉴别非霍奇金淋巴瘤亚型的探索性研究。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.10.015

YiJie Dong MD , Qing Hua MD , ShuJun Xia MD , CongCong Yuan MD , Cheng Li MD , YanYan Song PhD , YuHang Zheng PhD , RuoLin Tao MD , ZhenHua Liu MD , YuLu Zhang MS , FangGang Wu MS , Wei Guo PhD , Yuan Tian MS , JianQiao Zhou MD

Background

Identifying the subtype of intranodal non-Hodgkin lymphoma (NHL) is crucial for clinical management.

Rationale and Objectives

To display dual-vessel systems (microvascular and microlymphatic circulation) of intranodal NHL using super-resolution ultrasound (SRUS), and explore the diagnostic performance of SRUS imaging for predicting B-cell and T-cell subtypes NHL.

Materials and Methods

A total of 49 patients with intranodal NHL underwent dual-vessel system SRUS imaging via intravenous and intra-lymph node routes. Least absolute shrinkage and selection operator (LASSO) regression, fitted the LASSO model and leave-one-out cross-validation (LOOCV) were used for model development and internal validation.

Results

Among the 49 patients, 40 were diagnosed with B-cell NHL and 9 with T-cell NHL. Variables including LDmax, LDLmin, and VCmin were selected and the logistic regression model achieved discrimination of B-cell and T-cell subtype of lymphoma with an AUC of 0.831 (0.594–0.969).

Conclusion

Dual-vessel SRUS imaging can display real time microvascular and microlymphatic circulation of intranodal NHL in physiological status. With quantitative analysis of SRUS offers a potential non-invasive diagnostic alternative in differentiating NHL subtype.

背景：确定结内非霍奇金淋巴瘤（NHL）亚型对临床治疗至关重要。原理和目的：利用超分辨率超声（SRUS）显示结内NHL的双血管系统（微血管和微淋巴循环），并探讨SRUS成像在预测b细胞和t细胞亚型NHL中的诊断性能。材料和方法：共有49例结内NHL患者通过静脉和淋巴结内途径行双血管系统SRUS成像。最小绝对收缩和选择算子（LASSO）回归，拟合LASSO模型和留一交叉验证（LOOCV）用于模型开发和内部验证。结果：49例患者中，40例诊断为b细胞NHL， 9例诊断为t细胞NHL。选取LDmax、LDLmin、VCmin等变量，logistic回归模型实现了b细胞和t细胞亚型淋巴瘤的区分，AUC为0.831（0.594-0.969）。结论：双血管SRUS成像可实时显示结内NHL生理状态下的微血管和微淋巴循环。SRUS的定量分析为区分NHL亚型提供了一种潜在的非侵入性诊断选择。

{"title":"Dual-Vessel Microcirculation Imaging in Discriminating Non-Hodgkin Lymphoma Subtypes Using Super-Resolution Ultrasound: An Exploring Study","authors":"YiJie Dong MD , Qing Hua MD , ShuJun Xia MD , CongCong Yuan MD , Cheng Li MD , YanYan Song PhD , YuHang Zheng PhD , RuoLin Tao MD , ZhenHua Liu MD , YuLu Zhang MS , FangGang Wu MS , Wei Guo PhD , Yuan Tian MS , JianQiao Zhou MD","doi":"10.1016/j.acra.2025.10.015","DOIUrl":"10.1016/j.acra.2025.10.015","url":null,"abstract":"<div><h3>Background</h3><div>Identifying the subtype of intranodal non-Hodgkin lymphoma (NHL) is crucial for clinical management.</div></div><div><h3>Rationale and Objectives</h3><div>To display dual-vessel systems (microvascular and microlymphatic circulation) of intranodal NHL using super-resolution ultrasound (SRUS), and explore the diagnostic performance of SRUS imaging for predicting B-cell and T-cell subtypes NHL.</div></div><div><h3>Materials and Methods</h3><div>A total of 49 patients with intranodal NHL underwent dual-vessel system SRUS imaging via intravenous and intra-lymph node routes. Least absolute shrinkage and selection operator (LASSO) regression, fitted the LASSO model and leave-one-out cross-validation (LOOCV) were used for model development and internal validation.</div></div><div><h3>Results</h3><div>Among the 49 patients, 40 were diagnosed with B-cell NHL and 9 with T-cell NHL. Variables including LDmax, LDLmin, and VCmin were selected and the logistic regression model achieved discrimination of B-cell and T-cell subtype of lymphoma with an AUC of 0.831 (0.594–0.969).</div></div><div><h3>Conclusion</h3><div>Dual-vessel SRUS imaging can display real time microvascular and microlymphatic circulation of intranodal NHL in physiological status. With quantitative analysis of SRUS offers a potential non-invasive diagnostic alternative in differentiating NHL subtype.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"33 1","pages":"Pages 35-46"},"PeriodicalIF":3.9,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145402303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 Predictive Model for False-Negative Results in Ultrasound-Guided Percutaneous Transthoracic Needle Lung Biopsy 超声引导下经皮经胸肺穿刺活检假阴性结果的预测模型。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.10.023

Jiawei Yi , Ke Bi , Mengjun Shen , Kaiwen Wu , Xinyu Zhao , Runhe Xia , Yang Cong , Yi Li , Yin Wang

Objectives

This study aimed to develop a post-procedural predictive model for assessing the risk of false-negative results in ultrasound-guided percutaneous transthoracic needle lung biopsy (US-PTLB).

Material and Methods

Two prospective cohorts were designed for model development and validation. Patients scheduled for US-PTLB underwent B-mode ultrasound (B-US), color Doppler flow imaging (CDFI), ultrasound elastography, and contrast-enhanced ultrasound (CEUS) of the lesions, with the final diagnosis confirmed through comprehensive evaluation. Risk factors associated with false-negative results were identified, and multivariate logistic regression was used to construct the predictive model. The model's performance was further evaluated in an independent cohort to assess its impact on reducing the incidence of false-negative results through targeted interventions.

Results

The US-PTLB false-negative risk prediction model was constructed using data from 129 patients, of whom 35 (29.1%) were ultimately diagnosed with false-negative results. Predictors included age, lesion size, elasticity score, lesion necrosis, and enhancement intensity on CEUS. The model demonstrated excellent discrimination, with an area under the curve of 0.922, sensitivity of 88.6%, and specificity of 90.4%. Internal validation in 70 independently collected patients confirmed robust model performance. Application of the model in 423 patients, coupled with second biopsies for high-risk patients, led to a significant reduction in the incidence of false-negative results.

Conclusion

This predictive model, combining clinical parameters with multimodal ultrasound features, serves as a robust post-procedural tool for objectively assessing false-negative risk in ultrasound-guided percutaneous transthoracic needle lung biopsy. Its clinical application enables early risk stratification, minimizes false-negative rates, and enhances diagnostic precision.

目的：本研究旨在建立一种术后预测模型，用于评估超声引导下经皮经胸穿刺肺活检（US-PTLB）假阴性结果的风险。材料和方法：设计了两个前瞻性队列进行模型开发和验证。行US-PTLB的患者对病变行b超（B-US）、彩色多普勒血流显像（CDFI）、超声弹性成像（ultrasound elastography，超声造影）、超声造影（contrast-enhanced ultrasound， CEUS）检查，综合评价后确定最终诊断。确定与假阴性结果相关的危险因素，并采用多因素logistic回归构建预测模型。在一个独立的队列中进一步评估了该模型的性能，以评估其通过有针对性的干预措施减少假阴性结果发生率的影响。结果：利用129例患者的数据构建US-PTLB假阴性风险预测模型，其中35例（29.1%）最终诊断为假阴性。预测因素包括年龄、病变大小、弹性评分、病变坏死和超声造影增强强度。该模型具有良好的鉴别能力，曲线下面积为0.922，灵敏度为88.6%，特异度为90.4%。在70名独立收集的患者中进行的内部验证证实了模型的稳健性能。在423例患者中应用该模型，再加上对高危患者进行第二次活检，导致假阴性结果的发生率显著降低。结论：该预测模型将临床参数与多模态超声特征相结合，可作为超声引导下经皮经胸肺穿刺活检假阴性风险客观评估的可靠术后工具。它的临床应用使早期风险分层，最大限度地减少假阴性率，提高诊断精度。

{"title":"A Predictive Model for False-Negative Results in Ultrasound-Guided Percutaneous Transthoracic Needle Lung Biopsy","authors":"Jiawei Yi , Ke Bi , Mengjun Shen , Kaiwen Wu , Xinyu Zhao , Runhe Xia , Yang Cong , Yi Li , Yin Wang","doi":"10.1016/j.acra.2025.10.023","DOIUrl":"10.1016/j.acra.2025.10.023","url":null,"abstract":"<div><h3>Objectives</h3><div>This study aimed to develop a post-procedural predictive model for assessing the risk of false-negative results in ultrasound-guided percutaneous transthoracic needle lung biopsy (US-PTLB).</div></div><div><h3>Material and Methods</h3><div>Two prospective cohorts were designed for model development and validation. Patients scheduled for US-PTLB underwent B-mode ultrasound (B-US), color Doppler flow imaging (CDFI), ultrasound elastography, and contrast-enhanced ultrasound (CEUS) of the lesions, with the final diagnosis confirmed through comprehensive evaluation. Risk factors associated with false-negative results were identified, and multivariate logistic regression was used to construct the predictive model. The model's performance was further evaluated in an independent cohort to assess its impact on reducing the incidence of false-negative results through targeted interventions.</div></div><div><h3>Results</h3><div>The US-PTLB false-negative risk prediction model was constructed using data from 129 patients, of whom 35 (29.1%) were ultimately diagnosed with false-negative results. Predictors included age, lesion size, elasticity score, lesion necrosis, and enhancement intensity on CEUS. The model demonstrated excellent discrimination, with an area under the curve of 0.922, sensitivity of 88.6%, and specificity of 90.4%. Internal validation in 70 independently collected patients confirmed robust model performance. Application of the model in 423 patients, coupled with second biopsies for high-risk patients, led to a significant reduction in the incidence of false-negative results.</div></div><div><h3>Conclusion</h3><div>This predictive model, combining clinical parameters with multimodal ultrasound features, serves as a robust post-procedural tool for objectively assessing false-negative risk in ultrasound-guided percutaneous transthoracic needle lung biopsy. Its clinical application enables early risk stratification, minimizes false-negative rates, and enhances diagnostic precision.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"33 1","pages":"Pages 134-146"},"PeriodicalIF":3.9,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145460223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Author Response to “Clinical and Radiologic Contextualization of Automated BAC Quantification: A Commentary ” 作者对“致编辑的信：使用基于unet的深度学习检测心血管疾病来量化乳房x光片中的乳腺动脉钙化”的回复。

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.09.041

Wenbo Li MSc , Qiyu Zhang BSc , Dale J. Black BSc , Huanjun Ding PhD , Carlos Iribarren MD, MPH, PhD , Alireza Shojazadeh MD , Sabee Molloi PhD

引用次数: 0

Predictions of Response in Non-small Cell Lung Cancer Patients Treated with Immune Checkpoint Inhibitors Using Clinical Data, Deep Learning, and Radiomics 使用临床数据、深度学习和放射组学预测免疫检查点抑制剂治疗的非小细胞肺癌患者的反应

IF 3.9 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Academic Radiology

Pub Date : 2026-01-01 DOI: 10.1016/j.acra.2025.09.037

Chunxiao Wang , Yuxin Li , Yang Ji, Kang Yu, Chunhui Qin, Ling Liu, Yunjia Shuai, Jiahui Chen, Ao Li, Tong Zhang

Background

Determining predictive biomarkers for immunotherapy response in non-small cell lung cancer (NSCLC) patients is a complex task.

Objective

This research aimed to develop a multimodal model (CRDL) integrating clinical data, deep learning (DL), and radiomics (Rad) to predict immune responses in NSCLC patients receiving checkpoint blockade therapies. This study also evaluated whether CRDL outperforms unimodal, pre-fusion models (Pre-FMs) and post-fusion models (Post-FMs).

Methods

This is a retrospective study that utilized data from 228 lung cancer patients at the Memorial Sloan Kettering Cancer Center, with varying Programmed Death-Ligand 1(PD-L1) expression levels among the patients. 228 NSCLC patients were randomly divided into two groups in a 7:3 ratio: the training cohort (n = 159) and the validation cohort (n = 69). Image histological features were extracted using the "PyRadiomics" package, and DL features were obtained through the deep convolutional neural network from chest computed tomography images, and clinical data from the patients were also collected. Feature reduction was performed using t-tests and the Least absolute shrinkage and selection operator regression. Unimodal modal and Pre-FMs were constructed using random forests, while the post-fusion model was developed using a support vector machine approach. The performance of the model was measured by the area under the receiver operating characteristic curve (AUC).

Results

512 DL features and 382 Rad features were extracted. The CRDL model demonstrated superior performance with AUC values of 0.884 in the validation dataset and 0.976 in the training dataset, surpassing the best DL model in both unimodal and pre-fusion settings, which had training and validation AUCs of 0.854 and 0.749.

Conclusion

The CRDL model accurately forecasts immunotherapy responses in NSCLC patients, offering one dependable non-invasive test.

背景：确定非小细胞肺癌（NSCLC）患者免疫治疗反应的预测性生物标志物是一项复杂的任务。目的：本研究旨在建立一种整合临床数据、深度学习（DL）和放射组学（Rad）的多模态模型（CRDL），以预测接受检查点阻断治疗的NSCLC患者的免疫反应。本研究还评估了CRDL是否优于单峰模型、融合前模型（Pre-FMs）和融合后模型（Post-FMs）。方法：这是一项回顾性研究，利用了纪念斯隆凯特琳癌症中心228名肺癌患者的数据，这些患者的程序性死亡配体1（PD-L1）表达水平不同。228例NSCLC患者按7:3的比例随机分为两组：训练组（n=159）和验证组（n=69）。使用“PyRadiomics”软件包提取图像组织学特征，通过深度卷积神经网络提取胸部ct图像的DL特征，并收集患者的临床资料。使用t检验和最小绝对收缩和选择算子回归进行特征缩减。采用随机森林方法构建单峰模型和预融合模型，采用支持向量机方法构建融合后模型。该模型的性能通过接收机工作特性曲线下面积（AUC）来衡量。结果：提取DL特征512个，Rad特征382个。CRDL模型在验证集和训练集的AUC值分别为0.884和0.976，优于单峰和预融合设置下的最佳DL模型，前者的训练AUC和验证AUC分别为0.854和0.749。结论：CRDL模型准确预测非小细胞肺癌患者的免疫治疗反应，提供了一种可靠的无创检测方法。

{"title":"Predictions of Response in Non-small Cell Lung Cancer Patients Treated with Immune Checkpoint Inhibitors Using Clinical Data, Deep Learning, and Radiomics","authors":"Chunxiao Wang , Yuxin Li , Yang Ji, Kang Yu, Chunhui Qin, Ling Liu, Yunjia Shuai, Jiahui Chen, Ao Li, Tong Zhang","doi":"10.1016/j.acra.2025.09.037","DOIUrl":"10.1016/j.acra.2025.09.037","url":null,"abstract":"<div><h3>Background</h3><div>Determining predictive biomarkers for immunotherapy response in non-small cell lung cancer (NSCLC) patients is a complex task.</div></div><div><h3>Objective</h3><div>This research aimed to develop a multimodal model (CRDL) integrating clinical data, deep learning (DL), and radiomics (Rad) to predict immune responses in NSCLC patients receiving checkpoint blockade therapies. This study also evaluated whether CRDL outperforms unimodal, pre-fusion models (Pre-FMs) and post-fusion models (Post-FMs).</div></div><div><h3>Methods</h3><div>This is a retrospective study that utilized data from 228 lung cancer patients at the Memorial Sloan Kettering Cancer Center, with varying Programmed Death-Ligand 1(PD-L1) expression levels among the patients. 228 NSCLC patients were randomly divided into two groups in a 7:3 ratio: the training cohort (<em>n  </em>=69). Image histological features were extracted using the \"PyRadiomics\" package, and DL features were obtained through the deep convolutional neural network from chest computed tomography images, and clinical data from the patients were also collected. Feature reduction was performed using t-tests and the Least absolute shrinkage and selection operator regression. Unimodal modal and Pre-FMs were constructed using random forests, while the post-fusion model was developed using a support vector machine approach. The performance of the model was measured by the area under the receiver operating characteristic curve (AUC).</div></div><div><h3>Results</h3><div>512 DL features and 382 Rad features were extracted. The CRDL model demonstrated superior performance with AUC values of 0.884 in the validation dataset and 0.976 in the training dataset, surpassing the best DL model in both unimodal and pre-fusion settings, which had training and validation AUCs of 0.854 and 0.749.</div></div><div><h3>Conclusion</h3><div>The CRDL model accurately forecasts immunotherapy responses in NSCLC patients, offering one dependable non-invasive test.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"33 1","pages":"Pages 236-254"},"PeriodicalIF":3.9,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145294394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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