Magnetic resonance imaging最新文献_第4页

Time-dependent diffusion MRI combined with enhanced MRI and clinical indicators for preoperative prediction of CK19 expression status in hepatocellular carcinoma: a prospective study 时间依赖性弥散MRI联合增强MRI及临床指标预测肝细胞癌CK19表达状态的前瞻性研究

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-26 DOI: 10.1016/j.mri.2025.110602

Yu-chen Wei , Xing-Qing Qin , Jian-sun Li , Yuan-fang Tao , Chongze Yang , Qing ling Huang , Yan-yan Yu , Huiting Zhang , Haodong Qin , Thorsten Feiweier , Jin-yuan Liao

Objective

To explore the value of time-dependent diffusion MRI(Td-dMRI) in predicting the expression status of cytokeratin 19(CK19) in hepatocellular carcinoma(HCC) before surgery.

Materials and methods

Prospective inclusion of 72 HCC patients confirmed by surgical pathology (43 CK19-negative and 29 CK19-positive). All patients underwent time-dependent diffusion MRI (Td-dMRI) using a 3.0 T MR scanner before surgery, and quantitative parameters were calculated. Clinical data and MRI features of the patients were collected. Using univariate and multivariate logistic regression analysis to identify the risk factors for CK19 positive expression and establish a predictive model. The diagnostic performance of the model was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve, decision curve analysis (DCA), and calibration analysis.

Results

CK19-positive HCC exhibited significantly lower d and higher cellularity compared to CK19-negative HCC. CK19-positive HCC demonstrated significantly higher proportions of AFP (>200 ng/ml), CEA (>5 ng/ml), arterial phase rim enhancement, peritumoral hypointensity on hepatobiliary phase, peritumoral arterial hyperenhancement, and intratumoral necrosis/hemorrhage than CK19-negative HCC. Univariate and multivariate logistic regression analyses identified cellularity, AFP (>200 ng/ml), arterial phase rim enhancement, and peritumoral hypointensity on hepatobiliary phase as independent predictors of CK19 positivity. The combined model incorporating these four factors achieved an AUC of 0.889 (95 % CI: 0.809–0.968), with a sensitivity of 82.8 % and specificity of 86.0 %.

Conclusions

The cellularity value based on Td-dMRI was a potential quantitative biomarker for predicting CK19-positive HCC.

目的：探讨时间依赖性弥散MRI（Td-dMRI）在预测肝细胞癌（HCC）术前细胞角蛋白19（CK19）表达状况中的价值。材料和方法：前瞻性纳入手术病理证实的72例HCC患者（43例ck19阴性，29例ck19阳性）。所有患者术前均采用3.0 T MR扫描仪行时间依赖性弥散MRI （Td-dMRI）检查，并计算定量参数。收集患者的临床资料和MRI特征。采用单因素和多因素logistic回归分析，确定CK19阳性表达的危险因素，建立预测模型。采用受试者工作特征（ROC）曲线下面积（AUC）、决策曲线分析（DCA）和校准分析来评估模型的诊断性能。结果：与ck19阴性HCC相比，ck19阳性HCC表现出明显的低d和高细胞性。与ck19阴性HCC相比，ck19阳性HCC表现出AFP（>200 ng/mL）、CEA（>5 ng/mL）、动脉期边缘强化、肝胆期低密度、瘤周动脉高强化和瘤内坏死/出血的比例显著高于ck19阴性HCC。单因素和多因素logistic回归分析发现，细胞结构、AFP（>200 ng/mL）、动脉期边缘增强和肝胆期肿瘤周围低密度是CK19阳性的独立预测因素。纳入这四个因素的联合模型的AUC为0.889(95 % CI: 0.809 ~ 0.968)，敏感性为82.8 %，特异性为86.0 %。结论：基于Td-dMRI的细胞度值是预测ck19阳性HCC的潜在定量生物标志物。

{"title":"Time-dependent diffusion MRI combined with enhanced MRI and clinical indicators for preoperative prediction of CK19 expression status in hepatocellular carcinoma: a prospective study","authors":"Yu-chen Wei , Xing-Qing Qin , Jian-sun Li , Yuan-fang Tao , Chongze Yang , Qing ling Huang , Yan-yan Yu , Huiting Zhang , Haodong Qin , Thorsten Feiweier , Jin-yuan Liao","doi":"10.1016/j.mri.2025.110602","DOIUrl":"10.1016/j.mri.2025.110602","url":null,"abstract":"<div><h3>Objective</h3><div>To explore the value of time-dependent diffusion MRI(Td-dMRI) in predicting the expression status of cytokeratin 19(CK19) in hepatocellular carcinoma(HCC) before surgery.</div></div><div><h3>Materials and methods</h3><div>Prospective inclusion of 72 HCC patients confirmed by surgical pathology (43 CK19-negative and 29 CK19-positive). All patients underwent time-dependent diffusion MRI (Td-dMRI) using a 3.0 T MR scanner before surgery, and quantitative parameters were calculated. Clinical data and MRI features of the patients were collected. Using univariate and multivariate logistic regression analysis to identify the risk factors for CK19 positive expression and establish a predictive model. The diagnostic performance of the model was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve, decision curve analysis (DCA), and calibration analysis.</div></div><div><h3>Results</h3><div>CK19-positive HCC exhibited significantly lower d and higher cellularity compared to CK19-negative HCC. CK19-positive HCC demonstrated significantly higher proportions of AFP (>200 ng/ml), CEA (>5 ng/ml), arterial phase rim enhancement, peritumoral hypointensity on hepatobiliary phase, peritumoral arterial hyperenhancement, and intratumoral necrosis/hemorrhage than CK19-negative HCC. Univariate and multivariate logistic regression analyses identified cellularity, AFP (>200 ng/ml), arterial phase rim enhancement, and peritumoral hypointensity on hepatobiliary phase as independent predictors of CK19 positivity. The combined model incorporating these four factors achieved an AUC of 0.889 (95 % CI: 0.809–0.968), with a sensitivity of 82.8 % and specificity of 86.0 %.</div></div><div><h3>Conclusions</h3><div>The cellularity value based on Td-dMRI was a potential quantitative biomarker for predicting CK19-positive HCC.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110602"},"PeriodicalIF":2.0,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145850324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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

MR cytometry: More effective than conventional diffusion MRI in differentiating benign and malignant musculoskeletal tumors MR细胞术：在鉴别肌肉骨骼良恶性肿瘤方面比常规弥散MRI更有效。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-24 DOI: 10.1016/j.mri.2025.110603

Zhanxing Yan , Jian Zhao , Jianling Cui , Lisha Duan , Xiaohan Feng , Xinying Zhang , Shengnan Zhang , Wenhua Liang , Xiaohui Cao , Hong Yu

Objectives

To investigate the clinical significance of microstructure imaging based on MR cytometry in the qualitative diagnosis of musculoskeletal tumors (MSTs).

Materials and methods

Participants with clinically suspected MSTs between March 2024 and March 2025 were prospectively enrolled. Conventional apparent diffusion coefficient (ADC) value and four microstructural parameters, including cell diameter (d), intracellular volume fraction (v_in), Cellularity index, and extracellular diffusivity (D_ex), were estimated using the IMPULSED method (a form of MR cytometry). The ADC_PGSE, ADC_{25 Hz} and ADC_{40 Hz} were measured at three different effective diffusion times, and the change ADC (cADC) and relative change ADC (rcADC) were measured. The performance was evaluated using the area under the receiver operating characteristic curve (AUC) and compared using the DeLong test.

Results

A total of 62 participants with benign and malignant MSTs (mean age, 47.7 ± 20.9 [SD] years; 21 women, 41men) were enrolled. Among these ADC parameters, rcADC had the highest AUC (AUC = 0.737; 95 %Cl: 0.600, 0.874; p = 0.002). Among the four microstructure parameters derived from MR cytometry, Cellularity index had the highest AUC (AUC = 0.735; 95 %Cl: 0.601, 0.868; p = 0.002). Among all parameters, the AUC of rcADC and Cellularity index is higher than that of conventional ADC (AUC = 0.728; 95 %Cl: 0.594, 0.863; p = 0.003). The combination of the four microstructure parameters of MR cytometry further improved the diagnostic performance (AUC = 0.758; 95 %Cl: 0.626, 0.890; p = 0.001).

Conclusion

MR cytometry was an effective method for helping to predict the benign and malignant MSTs and is superior to conventional ADC values.

目的：探讨基于MR细胞术的微结构成像在肌肉骨骼肿瘤（MSTs）定性诊断中的临床意义。材料和方法：前瞻性纳入2024年3月至2025年3月期间临床疑似MSTs的参与者。常规表观扩散系数（ADC）值和四个微观结构参数，包括细胞直径(d)、细胞内体积分数（vin）、细胞度指数和细胞外扩散率（Dex），使用impulse方法（一种MR细胞术）进行估计。测定三种不同有效扩散时间下ADCPGSE、ADC25 Hz和ADC40 Hz，并测定变化ADC （cADC）和相对变化ADC （rcADC）。采用受试者工作特性曲线下面积（AUC）评价其性能，并采用DeLong试验进行比较。结果：共纳入62例良恶性MSTs患者（平均年龄47.7 ± 20.9 [SD]岁，女性21例，男性41例）。在ADC参数中，rcADC的AUC最高（AUC = 0.737;95 %Cl: 0.600, 0.874; p = 0.002）。在磁共振细胞术得到的4个微观结构参数中，细胞度指数的AUC最高（AUC = 0.735;95 %Cl: 0.601, 0.868; p = 0.002）。在所有参数中，rcADC的AUC和细胞度指数均高于常规ADC （AUC = 0.728;95 %Cl: 0.594, 0.863; p = 0.003）。结合磁共振细胞术的四个显微结构参数进一步提高了诊断效能（AUC = 0.758;95 %Cl: 0.626, 0.890; p = 0.001）。结论：MR细胞术是预测MSTs良恶性的有效方法，优于传统的ADC值。

{"title":"MR cytometry: More effective than conventional diffusion MRI in differentiating benign and malignant musculoskeletal tumors","authors":"Zhanxing Yan , Jian Zhao , Jianling Cui , Lisha Duan , Xiaohan Feng , Xinying Zhang , Shengnan Zhang , Wenhua Liang , Xiaohui Cao , Hong Yu","doi":"10.1016/j.mri.2025.110603","DOIUrl":"10.1016/j.mri.2025.110603","url":null,"abstract":"<div><h3>Objectives</h3><div>To investigate the clinical significance of microstructure imaging based on MR cytometry in the qualitative diagnosis of musculoskeletal tumors (MSTs).</div></div><div><h3>Materials and methods</h3><div>Participants with clinically suspected MSTs between March 2024 and March 2025 were prospectively enrolled. Conventional apparent diffusion coefficient (ADC) value and four microstructural parameters, including cell diameter (<em>d</em>), intracellular volume fraction (<em>v</em><sub>in</sub>), Cellularity index, and extracellular diffusivity (<em>D</em><sub>ex</sub>), were estimated using the IMPULSED method (a form of MR cytometry). The ADC<sub>PGSE</sub>, ADC<sub>25 Hz</sub> and ADC<sub>40 Hz</sub> were measured at three different effective diffusion times, and the change ADC (cADC) and relative change ADC (rcADC) were measured. The performance was evaluated using the area under the receiver operating characteristic curve (AUC) and compared using the DeLong test.</div></div><div><h3>Results</h3><div>A total of 62 participants with benign and malignant MSTs (mean age, 47.7 ± 20.9 [SD] years; 21 women, 41men) were enrolled. Among these ADC parameters, rcADC had the highest AUC (AUC = 0.737; 95 %Cl: 0.600, 0.874; <em>p</em> = 0.002). Among the four microstructure parameters derived from MR cytometry, Cellularity index had the highest AUC (AUC = 0.735; 95 %Cl: 0.601, 0.868; <em>p</em> = 0.002). Among all parameters, the AUC of rcADC and Cellularity index is higher than that of conventional ADC (AUC = 0.728; 95 %Cl: 0.594, 0.863; <em>p</em> = 0.003). The combination of the four microstructure parameters of MR cytometry further improved the diagnostic performance (AUC = 0.758; 95 %Cl: 0.626, 0.890; <em>p</em> = 0.001).</div></div><div><h3>Conclusion</h3><div>MR cytometry was an effective method for helping to predict the benign and malignant MSTs and is superior to conventional ADC values.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110603"},"PeriodicalIF":2.0,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145844082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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

Multicenter validation of amide proton transfer imaging for the classification of adult-type diffuse gliomas 酰胺质子转移成像对成人型弥漫性胶质瘤分级的多中心验证

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-24 DOI: 10.1016/j.mri.2025.110599

Tongling Jiang , Minghao Wu , Junjiao Hu , Hu Guo , Xiaoyue Ma , Yaou Liu , Yi Zhang

Accurate classification of adult-type diffuse gliomas is essential for treatment planning, particularly following the 2021 WHO classification update. While amide proton transfer (APT) imaging shows promise, multicenter validation is needed to establish its clinical utility. This study aims to evaluate the multicenter performance of the APT-weighted (APTw) metric for differentiating genotypes, grades, and subtypes of adult-type diffuse gliomas. MRI and clinical data were collected from three centers between July 2020 and January 2024. A standardized whole-brain SPACE CEST imaging protocol was used, and molecular diagnoses were confirmed. Tumor core regions were delineated on T2-weighted images, and multi-modality data were co-registered to T1-weighted images. Mean APTw values were assessed for IDH genotyping and grading among three centers, and for subtyping across two centers, using unpaired t-tests and receiver operating characteristic (ROC) curve analysis. A total of 123 patients (mean age: 48 ± 12 years; 67 males) were included. APTw indices were significantly higher in IDH-wildtype and high-grade gliomas compared to IDH-mutant and low-grade groups (p < 0.01). ROC analysis demonstrated strong classification performance: for IDH genotyping, the pooled area under the ROC curve (AUC) value was 0.84 (individual centers: 0.86, 0.84, and 0.93); for glioma grading, the pooled AUC was 0.83 (individual centers: 0.86, 0.84, and 0.83). Subtyping results showed good performance, particularly in differentiating glioblastomas from oligodendrogliomas (AUC: 0.93) and astrocytomas (AUC: 0.81). APT imaging effectively differentiated glioma grades and IDH mutations across centers, demonstrating robust multicenter performance. However, challenges remain in differentiating specific glioma subtypes.

成人型弥漫性胶质瘤的准确分类对于治疗计划至关重要，特别是在2021年世卫组织分类更新之后。虽然酰胺质子转移（APT）成像显示出希望，但需要多中心验证来建立其临床应用。本研究旨在评估apt加权（APTw）指标在区分成人型弥漫性胶质瘤的基因型、分级和亚型方面的多中心表现。2020年7月至2024年1月期间从三个中心收集MRI和临床数据。采用标准化全脑SPACE CEST成像方案，并进行分子诊断。在t2加权图像上描绘肿瘤核心区域，并将多模态数据共同配准到t1加权图像上。采用非配对t检验和受试者工作特征（ROC）曲线分析，评估三个中心之间IDH基因分型和分级的平均APTw值，以及两个中心之间的亚型分型。共纳入123例患者，平均年龄48±12岁，男性67例。与idh突变组和低级别组相比，idh野生型和高级别胶质瘤的APTw指数显著升高（p < 0.01）。ROC分析显示了较强的分类能力：对于IDH基因分型，ROC曲线下合并面积（AUC）值为0.84（个体中心分别为0.86、0.84和0.93）；对于胶质瘤分级，合并AUC为0.83（个别中心分别为0.86、0.84和0.83）。分型结果表现良好，特别是在胶质母细胞瘤与少突胶质胶质瘤（AUC: 0.93）和星形细胞瘤（AUC: 0.81）的区分上。APT成像可以有效地区分胶质瘤分级和IDH突变，显示出强大的多中心表现。然而，在区分特定的胶质瘤亚型方面仍然存在挑战。

{"title":"Multicenter validation of amide proton transfer imaging for the classification of adult-type diffuse gliomas","authors":"Tongling Jiang , Minghao Wu , Junjiao Hu , Hu Guo , Xiaoyue Ma , Yaou Liu , Yi Zhang","doi":"10.1016/j.mri.2025.110599","DOIUrl":"10.1016/j.mri.2025.110599","url":null,"abstract":"<div><div>Accurate classification of adult-type diffuse gliomas is essential for treatment planning, particularly following the 2021 WHO classification update. While amide proton transfer (APT) imaging shows promise, multicenter validation is needed to establish its clinical utility. This study aims to evaluate the multicenter performance of the APT-weighted (APTw) metric for differentiating genotypes, grades, and subtypes of adult-type diffuse gliomas. MRI and clinical data were collected from three centers between July 2020 and January 2024. A standardized whole-brain SPACE CEST imaging protocol was used, and molecular diagnoses were confirmed. Tumor core regions were delineated on T2-weighted images, and multi-modality data were co-registered to T1-weighted images. Mean APTw values were assessed for IDH genotyping and grading among three centers, and for subtyping across two centers, using unpaired <em>t</em>-tests and receiver operating characteristic (ROC) curve analysis. A total of 123 patients (mean age: 48 ± 12 years; 67 males) were included. APTw indices were significantly higher in IDH-wildtype and high-grade gliomas compared to IDH-mutant and low-grade groups (<em>p</em> < 0.01). ROC analysis demonstrated strong classification performance: for IDH genotyping, the pooled area under the ROC curve (AUC) value was 0.84 (individual centers: 0.86, 0.84, and 0.93); for glioma grading, the pooled AUC was 0.83 (individual centers: 0.86, 0.84, and 0.83). Subtyping results showed good performance, particularly in differentiating glioblastomas from oligodendrogliomas (AUC: 0.93) and astrocytomas (AUC: 0.81). APT imaging effectively differentiated glioma grades and IDH mutations across centers, demonstrating robust multicenter performance. However, challenges remain in differentiating specific glioma subtypes.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110599"},"PeriodicalIF":2.0,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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

Identification of functional neural networks of human brains with fMRI 用功能磁共振成像技术识别人脑功能神经网络。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-20 DOI: 10.1016/j.mri.2025.110600

Jie Huang

The highly evolved human brain comprises numerous functional systems, ranging from essential sensory, motor, attention and memory networks to higher-order cognitive functions like reasoning and language. These neural systems and cognitive functions are functionally integrated together to perform a task. BOLD-fMRI measures the neural activity across the entire brain at large-scale systems level. We define a functional neural network (FNN) as a network in which the temporal variation of the neural activity is the same across the entire network, i.e., all voxels in the network share the same BOLD time signal. We report a novel data-driven method to objectively and automatically identify FNNs across the entire brain for both brain states measured with resting-state and task-fMRI, respectively. The identified FNNs demonstrate dominating bilateral and symmetric characteristics for both brain states at both individual and group levels. These FNNs and the quantification of the interaction between them characterize the whole-brain activity holistically for each brain state and each individual subject.

高度进化的人类大脑包括许多功能系统，从基本的感觉、运动、注意力和记忆网络到推理和语言等高级认知功能。这些神经系统和认知功能在功能上整合在一起以执行任务。BOLD-fMRI在大尺度系统水平上测量整个大脑的神经活动。我们将功能神经网络（FNN）定义为神经活动在整个网络中的时间变化相同的网络，即网络中的所有体素共享相同的BOLD时间信号。我们报告了一种新的数据驱动方法，可以客观和自动地识别整个大脑的fnn，分别用静息状态和任务fmri测量大脑状态。所识别的fnn在个体和群体水平上都表现出两种大脑状态的主要双边和对称特征。这些fnn和它们之间相互作用的量化表征了每个大脑状态和每个个体受试者的全脑活动。

引用次数: 0

Time-dependent diffusion MRI–based microstructural mapping for predicting IDH mutation status in glioma: A multicenter study 预测胶质瘤中IDH突变状态的时间依赖扩散核磁共振显微结构定位：一项多中心研究。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-19 DOI: 10.1016/j.mri.2025.110598

Wanjun Hu , Wentao Liu , Darui Li , Liang Niu , Yuping Han , Junwei Chang , Tianyong Xu , Yuhui Xiong , Zhiqiang Ouyang , Qiang Liang , Jing Zhang

Objectives

To evaluate whether time-dependent diffusion MRI(td-dMRI)-based microstructural histogram parameters can accurately distinguish IDH mutation status in gliomas, with multicenter external validation.

Methods

In this prospective dual-center study, patients with pathologically confirmed glioma underwent preoperative conventional MRI and td-dMRI-based. Microstructural parameters including intracellular volume fraction (Vin), intracellular diffusivity (Din), extracellular diffusivity (Dex), cellularity, and cell diameter—were derived using the IMPULSED model. ADC values at different frequencies (25 Hz, 50 Hz, PGSE) were also computed. Tumor regions were manually segmented by two radiologists, excluding necrosis and edema. Voxel-wise values were extracted, and histogram features were computed using PyRadiomics. Feature selection included the Mann–Whitney U test, Spearman correlation, and logistic regression. A predictive model for IDH mutation status was developed using data from Center A and externally validated in Center B. Model performance was assessed using AUC, calibration curves, and confusion matrices.

Results

Among 147 histogram features extracted from seven td-dMRI–based mapping, eight nonredundant features significantly differed between IDH-wildtype and IDH-mutant gliomas. A logistic regression model based on ADC_PGSE_firstorder_Energy and cellularity_firstorder_10Percentile yielded AUCs of 0.801 (training) and 0.771 (validation). ADC_PGSE and cellularity were positively correlated with the Ki-67 index (P < 0.05).

Conclusion

Histogram features from td-dMRI (ADC_PGSE and cellularity) enabled robust IDH mutation prediction in a multicenter glioma cohort.

目的：通过多中心外部验证，评估基于时间依赖扩散MRI（td-dMRI）的显微结构直方图参数能否准确区分胶质瘤中IDH突变状态。方法：在这项前瞻性双中心研究中，病理证实的胶质瘤患者术前接受常规MRI和td- dmri。微结构参数包括细胞内体积分数（Vin）、细胞内扩散率（Din）、细胞外扩散率（Dex）、细胞度和细胞直径——使用impulse模型推导。还计算了不同频率（25 Hz, 50 Hz， PGSE）下的ADC值。肿瘤区域由两名放射科医生手工分割，排除坏死和水肿。提取体素值，并使用PyRadiomics计算直方图特征。特征选择包括Mann-Whitney U检验、Spearman相关和logistic回归。利用A中心的数据建立了IDH突变状态的预测模型，并在b中心进行了外部验证。使用AUC、校准曲线和混淆矩阵评估模型的性能。结果：从7个基于td- dmri的图谱中提取的147个直方图特征中，有8个非冗余特征在idh -野生型和idh -突变型胶质瘤之间存在显著差异。基于adcpgse_first storder_energy和cellularity_first storder_10percentile的逻辑回归模型的auc为0.801（训练）和0.771（验证）。ADCPGSE和细胞性与Ki-67指数呈正相关（P ）结论：来自dd - dmri的直方图特征（ADCPGSE和细胞性）能够在多中心胶质瘤队列中进行稳健的IDH突变预测。

{"title":"Time-dependent diffusion MRI–based microstructural mapping for predicting IDH mutation status in glioma: A multicenter study","authors":"Wanjun Hu , Wentao Liu , Darui Li , Liang Niu , Yuping Han , Junwei Chang , Tianyong Xu , Yuhui Xiong , Zhiqiang Ouyang , Qiang Liang , Jing Zhang","doi":"10.1016/j.mri.2025.110598","DOIUrl":"10.1016/j.mri.2025.110598","url":null,"abstract":"<div><h3>Objectives</h3><div>To evaluate whether time-dependent diffusion MRI(td-dMRI)-based microstructural histogram parameters can accurately distinguish IDH mutation status in gliomas, with multicenter external validation.</div></div><div><h3>Methods</h3><div>In this prospective dual-center study, patients with pathologically confirmed glioma underwent preoperative conventional MRI and td-dMRI-based. Microstructural parameters including intracellular volume fraction (Vin), intracellular diffusivity (Din), extracellular diffusivity (Dex), cellularity, and cell diameter—were derived using the IMPULSED model. ADC values at different frequencies (25 Hz, 50 Hz, PGSE) were also computed. Tumor regions were manually segmented by two radiologists, excluding necrosis and edema. Voxel-wise values were extracted, and histogram features were computed using PyRadiomics. Feature selection included the Mann–Whitney <em>U</em> test, Spearman correlation, and logistic regression. A predictive model for IDH mutation status was developed using data from Center A and externally validated in Center B. Model performance was assessed using AUC, calibration curves, and confusion matrices.</div></div><div><h3>Results</h3><div>Among 147 histogram features extracted from seven td-dMRI–based mapping, eight nonredundant features significantly differed between IDH-wildtype and IDH-mutant gliomas. A logistic regression model based on ADC<sub>PGSE</sub>_firstorder_Energy and cellularity_firstorder_10Percentile yielded AUCs of 0.801 (training) and 0.771 (validation). ADC<sub>PGSE</sub> and cellularity were positively correlated with the Ki-67 index (<em>P</em> < 0.05).</div></div><div><h3>Conclusion</h3><div>Histogram features from td-dMRI (ADC<sub>PGSE</sub> and cellularity) enabled robust IDH mutation prediction in a multicenter glioma cohort.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"128 ","pages":"Article 110598"},"PeriodicalIF":2.0,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145804885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Accuracy and precision of random walk with barrier model fitting: Simulations and applications in head and neck cancers 具有屏障模型拟合的随机行走的准确性和精度：头颈癌的模拟和应用。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-16 DOI: 10.1016/j.mri.2025.110597

Jiaren Zou , Yue Cao

Random Walk with Barrier Model (RWBM) can quantify microstructural parameters (free diffusivity, cell size, and membrane permeability) in head and neck cancers (HNCs) using time-dependent diffusion MRI. However, model fitting remains challenging due to limited number of measurements, low signal-to-noise ratio (SNR) and complex nonlinear biophysical models. In this work, we numerically evaluated RWBM fitting performance under a clinically feasible imaging protocol by comparing fits to the short-time-limit expression (STL-RWBM) and to the general RWBM across a broad range of SNRs. We also examined the model fitting degeneracy and fitting landscape for both fitting methods to elucidate their fitting behaviors. Numerical findings were further compared with model-fitting results from HNC patient data. We observed that under clinically relevant SNRs, fitting the STL-RWBM provided low-variance estimates of free diffusivity and membrane permeability, with biases comparable to those obtained from fitting the general RWBM. However, the general RWBM yielded low-bias estimates for cell size. The flat and tube-like fitting landscapes of the general RWBM led to high variance, frequent convergence to boundary constraints and spurious correlation among fitted parameters in vivo. The more restricted fitting landscape of the STL-RWBM substantially reduced variance. Both fitting methods produced cell size estimates in HNCs that were consistent with prior pathological findings. In conclusion, this work provided a comprehensive analysis of RWBM fitting in clinical settings and may guide optimizations of data acquisition and model fitting methods.

随机行走屏障模型（RWBM）可以量化头颈癌（HNCs）的显微结构参数（自由扩散率、细胞大小和膜通透性）。然而，由于测量数量有限，低信噪比（SNR）和复杂的非线性生物物理模型，模型拟合仍然具有挑战性。在这项工作中，我们通过比较短时限表达（STL-RWBM）和一般RWBM在广泛信噪比范围内的拟合，在临床可行的成像方案下对RWBM的拟合性能进行了数值评估。我们还研究了两种拟合方法的模型拟合退化性和拟合景观，以阐明它们的拟合行为。数值结果进一步与HNC患者数据的模型拟合结果进行了比较。我们观察到，在临床相关信噪比下，拟合STL-RWBM提供了自由扩散率和膜透性的低方差估计，其偏差与拟合一般RWBM所得的偏差相当。然而，一般RWBM对细胞大小的估计偏差较低。一般RWBM的平面和管状拟合景观导致体内拟合参数之间的高方差，频繁收敛于边界约束和虚假相关性。STL-RWBM更严格的拟合景观大大减少了方差。两种拟合方法产生的HNCs细胞大小估计值与先前的病理结果一致。总之，这项工作提供了临床环境下RWBM拟合的全面分析，并可能指导数据采集和模型拟合方法的优化。

{"title":"Accuracy and precision of random walk with barrier model fitting: Simulations and applications in head and neck cancers","authors":"Jiaren Zou , Yue Cao","doi":"10.1016/j.mri.2025.110597","DOIUrl":"10.1016/j.mri.2025.110597","url":null,"abstract":"<div><div>Random Walk with Barrier Model (RWBM) can quantify microstructural parameters (free diffusivity, cell size, and membrane permeability) in head and neck cancers (HNCs) using time-dependent diffusion MRI. However, model fitting remains challenging due to limited number of measurements, low signal-to-noise ratio (SNR) and complex nonlinear biophysical models. In this work, we numerically evaluated RWBM fitting performance under a clinically feasible imaging protocol by comparing fits to the short-time-limit expression (STL-RWBM) and to the general RWBM across a broad range of SNRs. We also examined the model fitting degeneracy and fitting landscape for both fitting methods to elucidate their fitting behaviors. Numerical findings were further compared with model-fitting results from HNC patient data. We observed that under clinically relevant SNRs, fitting the STL-RWBM provided low-variance estimates of free diffusivity and membrane permeability, with biases comparable to those obtained from fitting the general RWBM. However, the general RWBM yielded low-bias estimates for cell size. The flat and tube-like fitting landscapes of the general RWBM led to high variance, frequent convergence to boundary constraints and spurious correlation among fitted parameters in vivo. The more restricted fitting landscape of the STL-RWBM substantially reduced variance. Both fitting methods produced cell size estimates in HNCs that were consistent with prior pathological findings. In conclusion, this work provided a comprehensive analysis of RWBM fitting in clinical settings and may guide optimizations of data acquisition and model fitting methods.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110597"},"PeriodicalIF":2.0,"publicationDate":"2025-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145781614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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

The nature and interpretation of BOLD signals in white matter - A review 脑白质BOLD信号的性质及解释综述。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-16 DOI: 10.1016/j.mri.2025.110596

J.C. Gore , M. Li , K.G. Schilling , L. Xu , Y. Li , Z. Zu , A.W. Anderson , Z. Ding , Y. Gao

This review summarizes selected recent findings demonstrating the dependence of blood oxygenation level dependent (BOLD) signals in white matter (WM) on tissue microstructure, composition, vascular properties and metabolism, as well as their relationships to fMRI signals within gray matter (GM) networks. BOLD signals in WM are robustly detectable after a stimulus, and at rest their temporal variations reveal synchronized networks and correlated neural activities involving both WM and GM. However, to date, most analyses of brain fMRI data have ignored WM signals, and often have removed them as nuisance regressors. However, emerging evidence clearly demonstrates that WM BOLD signals represent potentially important and heretofore overlooked indicators of neural activities that are intimately related to how cortical regions communicate, and so should be incorporated into more complete models of brain functional organization. Here we review recent work that contributes to our understanding of their interpretation and significance.

The factors that affect the magnitude and other characteristics of BOLD responses in WM are becoming more clear, and recent studies have demonstrated and quantified the relationships between BOLD signals and vascular and microstructural properties of WM tracts. These relationships depend on the degree of myelination and neurite and mitochondrial densities, but they also are qualitatively different when comparing different fiber types, notably association versus projection fibers. Some fully myelinated fibers appear to not show detectable BOLD effects. The relationships between WM and GM BOLD signals, the contributions of GM resting state correlations and signals to WM BOLD signals, and the engagement of WM in GM networks, are also becoming more clear. These findings supplement the growing literature demonstrating practical, clinical applications of BOLD in WM. The goal of this review is to highlight recent research that demonstrates how WM and GM activities are related, and to stimulate further investigations that may produce a more complete model of brain organization.

这篇综述总结了最近的一些研究结果，这些发现证明了白质（WM）中血氧水平依赖（BOLD）信号对组织微观结构、组成、血管特性和代谢的依赖性，以及它们与灰质（GM）网络中fMRI信号的关系。在刺激后，WM中的BOLD信号可以被稳健地检测到，并且在静止状态下，它们的时间变化揭示了涉及WM和GM的同步网络和相关神经活动。然而，迄今为止，大多数脑功能磁共振数据分析都忽略了WM信号，并且经常将其作为讨厌的回归量去除。然而，新出现的证据清楚地表明，WM BOLD信号代表了与皮层区域如何交流密切相关的神经活动的潜在重要和迄今为止被忽视的指标，因此应该纳入更完整的脑功能组织模型。在这里，我们回顾最近的工作，有助于我们的解释和意义的理解。影响WM中BOLD反应的大小和其他特征的因素越来越清楚，最近的研究已经证明并量化了BOLD信号与WM束血管和微观结构特性之间的关系。这些关系取决于髓鞘、神经突和线粒体密度的程度，但在比较不同纤维类型时，它们也有质的不同，特别是关联纤维和投射纤维。一些完全有髓鞘纤维似乎没有可检测到的BOLD效应。WM和GM BOLD信号之间的关系，GM静息状态相关和信号对WM BOLD信号的贡献，以及WM在GM网络中的参与也变得越来越清楚。这些发现补充了越来越多的文献证明BOLD在WM中的实际临床应用。这篇综述的目的是强调最近的研究，证明WM和GM活动是如何相关的，并刺激进一步的研究，可能会产生一个更完整的大脑组织模型。

{"title":"The nature and interpretation of BOLD signals in white matter - A review","authors":"J.C. Gore , M. Li , K.G. Schilling , L. Xu , Y. Li , Z. Zu , A.W. Anderson , Z. Ding , Y. Gao","doi":"10.1016/j.mri.2025.110596","DOIUrl":"10.1016/j.mri.2025.110596","url":null,"abstract":"<div><div>This review summarizes selected recent findings demonstrating the dependence of blood oxygenation level dependent (BOLD) signals in white matter (WM) on tissue microstructure, composition, vascular properties and metabolism, as well as their relationships to fMRI signals within gray matter (GM) networks. BOLD signals in WM are robustly detectable after a stimulus, and at rest their temporal variations reveal synchronized networks and correlated neural activities involving both WM and GM. However, to date, most analyses of brain fMRI data have ignored WM signals, and often have removed them as nuisance regressors. However, emerging evidence clearly demonstrates that WM BOLD signals represent potentially important and heretofore overlooked indicators of neural activities that are intimately related to how cortical regions communicate, and so should be incorporated into more complete models of brain functional organization. Here we review recent work that contributes to our understanding of their interpretation and significance.</div><div>The factors that affect the magnitude and other characteristics of BOLD responses in WM are becoming more clear, and recent studies have demonstrated and quantified the relationships between BOLD signals and vascular and microstructural properties of WM tracts. These relationships depend on the degree of myelination and neurite and mitochondrial densities, but they also are qualitatively different when comparing different fiber types, notably association versus projection fibers. Some fully myelinated fibers appear to not show detectable BOLD effects. The relationships between WM and GM BOLD signals, the contributions of GM resting state correlations and signals to WM BOLD signals, and the engagement of WM in GM networks, are also becoming more clear. These findings supplement the growing literature demonstrating practical, clinical applications of BOLD in WM. The goal of this review is to highlight recent research that demonstrates how WM and GM activities are related, and to stimulate further investigations that may produce a more complete model of brain organization.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110596"},"PeriodicalIF":2.0,"publicationDate":"2025-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145781590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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

Functional MRI quantitative parameters as biomarkers of proliferation in synovial sarcoma xenografts: A study based on precise MR imaging-pathology correlation 功能性MRI定量参数作为异种滑膜肉瘤增生的生物标志物：一项基于精确MRI成像-病理相关性的研究。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-14 DOI: 10.1016/j.mri.2025.110594

Chengjiang Xu , Yasmin Mushtaq , Xiaoge Liu , Guijiao Qin , Juan Tao , Yajie Liu , Jinge Li , Xinyu Yang , Shaowu Wang

Purpose

To determine whether quantitative parameters from diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and dynamic contrast-enhanced MRI (DCE-MRI) are associated with proliferative activity in synovial sarcoma (SS) xenografts.

Materials and methods

Thirty-four synovial sarcoma (SS) xenograft models were established. All mice underwent MRI, including DWI (ADC), IVIM (D, D*, f), and DCE-MRI (K^trans, K_ep, V_e). An MR imaging–pathology correlation method was used to align imaging ROIs with pathological sampling sites.

Pearson or Spearman correlation analyses were performed to assess associations between MRI parameters and proliferation-related markers (mitotic count, Ki-67, and PTEN). Differences between high and low groups for each marker were evaluated using independent-sample t-tests or Wilcoxon rank-sum tests. Diagnostic performance was assessed using receiver operating characteristic (ROC) curve analysis, and AUCs were compared with DeLong's test.

Results

The D value was significantly associated with mitotic count, Ki-67 expression, and PTEN expression (P = 0.022, 0.004, 0.001). V_e showed a positive association with mitotic count (P = 0.007), while K^trans demonstrated a moderate negative association with PTEN expression (P < 0.001). D and V_e showed moderate ability to distinguish high from low mitotic count (AUC = 0.734 and 0.786). The D value showed moderate differentiation between Ki-67 expression groups (AUC = 0.802), and K^trans provided moderate discrimination between PTEN expression groups (AUC = 0.813).

Conclusion

The proliferative activity of SS xenografts could be assessed using quantitative parameters derived from IVIM and DCE-MRI.

目的：确定来自扩散加权成像（DWI）、体素内非相干运动（IVIM）和动态对比增强MRI （DCE-MRI）的定量参数是否与滑膜肉瘤（SS）异种移植物的增殖活性相关。材料与方法：建立34例滑膜肉瘤（SS）异种移植模型。所有小鼠均行MRI检查，包括DWI （ADC）、IVIM （D, D*, f）和DCE-MRI （Ktrans, Kep, Ve）。使用磁共振成像-病理相关方法将成像roi与病理采样点对齐。Pearson或Spearman相关分析评估MRI参数与增殖相关标志物（有丝分裂计数、Ki-67和PTEN）之间的相关性。使用独立样本t检验或Wilcoxon秩和检验评估每个标记的高组和低组之间的差异。采用受试者工作特征（ROC）曲线分析评估诊断效果，并将auc与DeLong检验进行比较。结果：D值与有丝分裂计数、Ki-67表达、PTEN表达显著相关（P = 0.022,0.004,0.001）。Ve与有丝分裂计数呈正相关（P = 0.007），而Ktrans与PTEN表达呈中度负相关(P e具有中度区分有丝分裂计数高低的能力（AUC = 0.734和0.786）。D值在Ki-67表达组间存在中度差异（AUC = 0.802），Ktrans在PTEN表达组间存在中度差异（AUC = 0.813）。结论：利用IVIM和DCE-MRI的定量参数可以评估SS异种移植物的增殖活性。

{"title":"Functional MRI quantitative parameters as biomarkers of proliferation in synovial sarcoma xenografts: A study based on precise MR imaging-pathology correlation","authors":"Chengjiang Xu , Yasmin Mushtaq , Xiaoge Liu , Guijiao Qin , Juan Tao , Yajie Liu , Jinge Li , Xinyu Yang , Shaowu Wang","doi":"10.1016/j.mri.2025.110594","DOIUrl":"10.1016/j.mri.2025.110594","url":null,"abstract":"<div><h3>Purpose</h3><div>To determine whether quantitative parameters from diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and dynamic contrast-enhanced MRI (DCE-MRI) are associated with proliferative activity in synovial sarcoma (SS) xenografts.</div></div><div><h3>Materials and methods</h3><div>Thirty-four synovial sarcoma (SS) xenograft models were established. All mice underwent MRI, including DWI (ADC), IVIM (D, D*, f), and DCE-MRI (K<sup>trans</sup>, K<sub>ep</sub>, V<sub>e</sub>). An MR imaging–pathology correlation method was used to align imaging ROIs with pathological sampling sites.</div><div>Pearson or Spearman correlation analyses were performed to assess associations between MRI parameters and proliferation-related markers (mitotic count, Ki-67, and PTEN). Differences between high and low groups for each marker were evaluated using independent-sample <em>t</em>-tests or Wilcoxon rank-sum tests. Diagnostic performance was assessed using receiver operating characteristic (ROC) curve analysis, and AUCs were compared with DeLong's test.</div></div><div><h3>Results</h3><div>The D value was significantly associated with mitotic count, Ki-67 expression, and PTEN expression (<em>P</em> = 0.022, 0.004, 0.001). V<sub>e</sub> showed a positive association with mitotic count (<em>P</em> = 0.007), while K<sup>trans</sup> demonstrated a moderate negative association with PTEN expression (<em>P</em> < 0.001). D and V<sub>e</sub> showed moderate ability to distinguish high from low mitotic count (AUC = 0.734 and 0.786). The D value showed moderate differentiation between Ki-67 expression groups (AUC = 0.802), and K<sup>trans</sup> provided moderate discrimination between PTEN expression groups (AUC = 0.813).</div></div><div><h3>Conclusion</h3><div>The proliferative activity of SS xenografts could be assessed using quantitative parameters derived from IVIM and DCE-MRI.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110594"},"PeriodicalIF":2.0,"publicationDate":"2025-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145768561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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 dense recurrent unrolling network leveraging spatio-temporal priors for highly-accelerated dynamic MRI 一个密集的循环展开网络利用时空先验的高加速动态MRI。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-13 DOI: 10.1016/j.mri.2025.110595

Bin Wang , Yusheng Lian , Wan Zhang , Zilong Liu , Xiaojie Hu , Beiqing Huang , Yuanyuan Wang

Dynamic magnetic resonance imaging (MRI) requires accurate reconstruction from undersampled k-space data to achieve high temporal resolution within clinically acceptable scan times. Deep unrolling architectures have recently emerged as effective solutions by integrating physics-based data consistency with learned priors. However, their ability to exploit temporal relationships remains limited, as many approaches rely on independent stage-wise processing with only final-stage outputs propagated across iterations, which restricts feature interaction and often leads to performance degradation when acceleration factors increase. To enhance temporal prior learning, we introduce a bidirectional recurrent convolutional unit within the sparse prior update module. Our approach strengthens temporal dependency modeling by recurrently aggregating contextual information from both past and future frames, thereby improving stability and representation capacity under highly undersampled conditions. Furthermore, we incorporate inter-stage feature transmission that forwards intermediate representations instead of only single-stage outputs. This design substantially improves multi-stage collaboration, enabling more effective refinement across iterations. Experimental results on accelerated dynamic MRI datasets (6×, 12×, and 24×) demonstrate that the proposed method consistently outperforms state-of-the-art unrolling and deep learning strategies in reconstruction accuracy and temporal fidelity. Ablation studies further validate the contributions of recurrent temporal learning and inter-stage feature transmission.

动态磁共振成像（MRI）需要从欠采样k空间数据精确重建，以在临床可接受的扫描时间内实现高时间分辨率。通过将基于物理的数据一致性与学习先验相结合，深度展开架构最近成为一种有效的解决方案。然而，它们利用时间关系的能力仍然有限，因为许多方法依赖于独立的阶段处理，只有跨迭代传播的最后阶段输出，这限制了特征交互，并且当加速因素增加时经常导致性能下降。为了增强时间先验学习，我们在稀疏先验更新模块中引入了双向循环卷积单元。我们的方法通过循环地聚合过去和未来框架的上下文信息来加强时间依赖性建模，从而提高在高度欠采样条件下的稳定性和表示能力。此外，我们结合了转发中间表示的阶段间特征传输，而不仅仅是单阶段输出。这种设计极大地改进了多阶段协作，在迭代之间实现了更有效的细化。在加速动态MRI数据集（6x、12x和24x）上的实验结果表明，该方法在重建精度和时间保真度方面始终优于最先进的展开和深度学习策略。消融研究进一步证实了反复颞叶学习和阶段间特征传递的作用。

{"title":"A dense recurrent unrolling network leveraging spatio-temporal priors for highly-accelerated dynamic MRI","authors":"Bin Wang , Yusheng Lian , Wan Zhang , Zilong Liu , Xiaojie Hu , Beiqing Huang , Yuanyuan Wang","doi":"10.1016/j.mri.2025.110595","DOIUrl":"10.1016/j.mri.2025.110595","url":null,"abstract":"<div><div>Dynamic magnetic resonance imaging (MRI) requires accurate reconstruction from undersampled k-space data to achieve high temporal resolution within clinically acceptable scan times. Deep unrolling architectures have recently emerged as effective solutions by integrating physics-based data consistency with learned priors. However, their ability to exploit temporal relationships remains limited, as many approaches rely on independent stage-wise processing with only final-stage outputs propagated across iterations, which restricts feature interaction and often leads to performance degradation when acceleration factors increase. To enhance temporal prior learning, we introduce a bidirectional recurrent convolutional unit within the sparse prior update module. Our approach strengthens temporal dependency modeling by recurrently aggregating contextual information from both past and future frames, thereby improving stability and representation capacity under highly undersampled conditions. Furthermore, we incorporate inter-stage feature transmission that forwards intermediate representations instead of only single-stage outputs. This design substantially improves multi-stage collaboration, enabling more effective refinement across iterations. Experimental results on accelerated dynamic MRI datasets (6×, 12×, and 24×) demonstrate that the proposed method consistently outperforms state-of-the-art unrolling and deep learning strategies in reconstruction accuracy and temporal fidelity. Ablation studies further validate the contributions of recurrent temporal learning and inter-stage feature transmission.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110595"},"PeriodicalIF":2.0,"publicationDate":"2025-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145763081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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

APTWI-differential analysis for breast cancer: Association with histopathologic characteristics and early prediction of neoadjuvant chemotherapy response 乳腺癌的aptwi差异分析：与组织病理学特征和新辅助化疗反应的早期预测的关系。

IF 2 4区医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Magnetic resonance imaging

Pub Date : 2025-12-12 DOI: 10.1016/j.mri.2025.110590

Jie Fang , Xiaoxia Wang , Lu Wang , Ying Cao , Yao Huang , Shuling Liu , Huifang Chen , Zhechuan Dai , Tao Yu , Sun Tang , Meng Lin , Yi Zhang , Jiuquan Zhang

The purpose of the study is to investigate the value of Amide proton transfer imaging(APTWI)-differential analysis in association with histopathologic characteristics, and the performance to early predict pathologic complete response (pCR) in participants with breast cancer (BC). Participants with BC who underwent pretreatment APTWI between November 2022 and April 2024 were prospectively enrolled. APT-specific signal quantification was achieved through differential analysis between model-fitted and experimentally acquired Z-spectrum at +3.5 ppm. Univariate analysis was used to identify APT^# values associated with histopathologic characteristics and pCR. The area under the receiver operating characteristic curve (AUC) analysis was performed to evaluate the diagnostic value of APTWI-DIGITAL on histopathologic characteristics and assess the predictive performance for pCR. The analysis ultimately included 123 participants with BC (mean age, 52 years±9 [SD]), 43 participants of whom received neoadjuvant chemotherapy (NAC) and 15 participants who achieved pCR. In the pre-treatment group, the APT^# values showed reasonable performance in identifying the positive status of KI67 proliferation index (P = 0.01, AUC = 0.69), and PR (P = 0.045, AUC = 0.60). In the NAC group, the APT^# values in the pCR participants showed a significant downward trend at the T1 (P = 0.01, AUC = 0.80), and was not significant between pCR and non-pCR at other timepoints. The findings suggest APTWI-differential analysis may be useful imaging biomarkers to characterize the immunohistochemical biomarkers and predict pCR to NAC in BC patients.

本研究的目的是探讨酰胺质子转移成像（APTWI）鉴别分析在乳腺癌（BC）患者中与组织病理特征相关的价值，以及早期预测病理完全缓解（pCR）的性能。在2022年11月至2024年4月期间接受APTWI预处理的BC患者被前瞻性纳入研究。通过模型拟合和实验获得的+3.5 ppm的z谱之间的差异分析，实现了apt特异性信号量化。采用单因素分析确定与组织病理学特征和pCR相关的APT#值。通过受试者工作特征曲线下面积（AUC）分析，评估APTWI-DIGITAL对组织病理特征的诊断价值，并评估pCR的预测性能。分析最终纳入123例BC患者（平均年龄52 岁±9 [SD]）， 43例接受新辅助化疗（NAC）， 15例获得pCR的患者。预处理组APT#值对KI67增殖指数（P = 0.01,AUC = 0.69）和PR （P = 0.045,AUC = 0.60）阳性状态的判断表现合理。在NAC组中，pCR参与者的APT#值在T1处呈显著下降趋势（P = 0.01,AUC = 0.80），在其他时间点pCR与非pCR之间无显著差异。研究结果表明，aptwi差异分析可能是有用的成像生物标志物，用于表征BC患者的免疫组织化学生物标志物和预测pCR到NAC。

{"title":"APTWI-differential analysis for breast cancer: Association with histopathologic characteristics and early prediction of neoadjuvant chemotherapy response","authors":"Jie Fang , Xiaoxia Wang , Lu Wang , Ying Cao , Yao Huang , Shuling Liu , Huifang Chen , Zhechuan Dai , Tao Yu , Sun Tang , Meng Lin , Yi Zhang , Jiuquan Zhang","doi":"10.1016/j.mri.2025.110590","DOIUrl":"10.1016/j.mri.2025.110590","url":null,"abstract":"<div><div>The purpose of the study is to investigate the value of Amide proton transfer imaging(APTWI)-differential analysis in association with histopathologic characteristics, and the performance to early predict pathologic complete response (pCR) in participants with breast cancer (BC). Participants with BC who underwent pretreatment APTWI between November 2022 and April 2024 were prospectively enrolled. APT-specific signal quantification was achieved through differential analysis between model-fitted and experimentally acquired <em>Z</em>-spectrum at +3.5 ppm. Univariate analysis was used to identify APT<sup>#</sup> values associated with histopathologic characteristics and pCR. The area under the receiver operating characteristic curve (AUC) analysis was performed to evaluate the diagnostic value of APTWI-DIGITAL on histopathologic characteristics and assess the predictive performance for pCR. The analysis ultimately included 123 participants with BC (mean age, 52 years±9 [SD]), 43 participants of whom received neoadjuvant chemotherapy (NAC) and 15 participants who achieved pCR. In the pre-treatment group, the APT<sup>#</sup> values showed reasonable performance in identifying the positive status of KI67 proliferation index (<em>P</em> = 0.01, AUC = 0.69), and PR (<em>P</em> = 0.045, AUC = 0.60). In the NAC group, the APT<sup>#</sup> values in the pCR participants showed a significant downward trend at the T1 (<em>P</em> = 0.01, AUC = 0.80), and was not significant between pCR and non-pCR at other timepoints. The findings suggest APTWI-differential analysis may be useful imaging biomarkers to characterize the immunohistochemical biomarkers and predict pCR to NAC in BC patients.</div></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"127 ","pages":"Article 110590"},"PeriodicalIF":2.0,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145756795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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