Inhomogeneous magnetization transfer (ihMT) is sensitive to dipolar order associated with motion-restricted macromolecules and can be characterized by the dipolar relaxation time T 1 D $$ {T}_{1D} $$ . In this study, we propose a spin-lock MRI framework for T 1 D $$ {T}_{1D} $$ quantification. Specifically, we introduce a T 1 D $$ {T}_{1D} $$ -sensitive metric, RATI O dosl $$ RATI{O}_{dosl} $$ , derived from the distinct relaxation rate R dosl $$ {R}_{dosl} $$ , defined as the difference between dual-frequency and single-frequency R 1 ρ $$ {R}_{1\rho } $$ measurements. To enable dual-frequency spin-lock acquisition, we developed a dedicated rotary-echo spin-lock sequence. Based on this framework, we further estimated T 1 D $$ {T}_{1D} $$ and the macromolecular proton fraction (MPF) within a unified acquisition. The proposed method was evaluated using numerical simulations, phantom experiments, and in vivo imaging in the healthy human brain. Simulations demonstrated high sensitivity of RATI O dosl $$ RATI{O}_{dosl} $$ to T 1 D $$ {T}_{1D} $$ and supported the robustness of the proposed approach under the investigated conditions. Phantom experiments showed measurable ihMT contrast and supported the feasibility of T 1 D $$ {T}_{1D} $$ estimation using RATI O dosl $$ RATI{O}_{dosl} $$ . In vivo experiments demonstrated simultaneous T 1 D $$ {T}_{1D} $$ and MPF mapping using only three spin-lock-prepared images. Across 10 healthy volunteers, mean white matter T 1 D $$ {T}_{1D} $$ values ranged from approximately 3.70 to 4.80 ms. By requiring only three contrast-prepared images, the proposed technique provides a rapid framework for simultaneous T 1 D $$ {T}_{1D} $$ and MPF mapping and may facilitate further investigation of dipolar-order-sensitive microstructural imaging in vivo.

Purpose To examine whether time-dependent diffusion MRI (Td-dMRI) and macromolecular proton fraction (MPF) mapping-derived quantitative metrics can effectively distinguish between cervical cancer with and without lymph node metastasis (LNM) before treatment. Materials and Methods In this prospective study of adults with clinically suspected cervical cancer who underwent Td-dMRI, MPF mapping, and pulsed gradient spin-echo diffusion-weighted imaging (DWIPGSE) examinations between October 2023 and June 2025, authors calculated Td-dMRI-derived parameters (cellularity, diameter, intracellular volume fraction [Vin], and extracellular diffusivity [Dex]), MPF, and DWIPGSE-derived parameter (pulsed gradient spin-echo apparent diffusion coefficient [ADCPGSE]). Through Ridge regression analysis, the authors identified independent predictors of LNM and developed a composite diagnostic tool using logistic regression analysis. To evaluate tool performance, the area under the receiver operating characteristic curve was determined. Results Among 98 female individuals with cervical cancer (mean age, 56.69 years ± 11.63 [SD]), participants who were LNM positive exhibited higher cellularity, Vin, and MPF but lower diameter, Dex, and ADCPGSE than their counterparts who were LNM negative (P < .001 to P = .007). Cellularity, maximum tumor diameter, and MPF were independent predictors of LNM status, with their combination yielding the best diagnostic performance (area under the receiver operating characteristic curve, 0.95; 95% CI: 0.89, 0.98). The performance of this combination surpassed that of individual imaging modality, including DWIPGSE (ADCPGSE), and MPF, as well as any individual parameter, including cellularity, Vin, diameter, and Dex. Conclusion Td-dMRI and MPF mapping were effective for predicting LNM in cervical cancer, with the combination of cellularity, maximum tumor diameter, and MPF showing the best diagnostic performance. Keywords: Time-Dependent Diffusion MRI, Macromolecular Proton Fraction, Cervical Cancer, Lymph Node Metastases © RSNA, 2026.

RATIONALE AND OBJECTIVES:To investigate the value of time-dependent diffusion MRI (Td-dMRI) and macromolecular proton fraction (MPF) imaging in assessing the pathological grade of cervical cancer (CC). MATERIALS AND METHODS:A total of 92 CC patients, comprising 35 with high-grade (grade III) cancer and 57 with low-grade (grade I/II) cancer, who underwent Td-dMRI and MPF, were prospectively enrolled. Td-dMRI derived parameters including cellularity (cell density), diameter (tumor cell size), Dex (extracellular diffusivity), Vin (intracellular volume fraction), and three apparent diffusion coefficients (ADCPGSE, ADC17 Hz, ADC33Hz) and MPF derived parameter MPF (tissue macromolecular) were calculated and compared. Diagnostic performance was assessed via area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and decision curve analysis (DCA); internal validation was performed using 1000 bootstrap resamples to mitigate model optimism. Multiple pairwise AUC comparisons were adjusted using the DeLong test. RESULTS:Cellularity, Vin, and MPF were higher and diameter, Dex, ADCPGSE, ADC17 Hz, and ADC33 Hz were lower in high-grade group than in low-grade group (all P < 0.05). Cellularity, Vin, and MPF were independent predictors and their combination achieved optimal diagnostic efficacy (AUC = 0.960; 95% CI: 0.897-0.990; sensitivity = 97.14%; specificity = 82.46%), which was significantly higher than any individual parameter (AUC range: 0.685-0.923; all P < 0.05 after Bonferroni correction). Internal validation confirmed stable performance (AUC = 0.955; 95% CI: 0.939-0.960), and DCA demonstrated higher net benefit for patients. Vin strongly correlated with pathological nuclear fraction (r = 0.752, P < 0.001). CONCLUSION:Td-dMRI and MPF were effective methods of predicting pathological grade in CC, and the combination of cellularity, Vin, and MPF has the potential to serve as a new imaging marker, facilitating preoperative grading and personalized treatment.

Semi-supervised learning (SSL) has shown notable potential in relieving the heavy demand of dense prediction tasks on large-scale well-annotated datasets, especially for the challenging multi-organ segmentation (MoS). However, the prevailing class-imbalance problem in MoS caused by the substantial variations in organ size exacerbates the learning difficulty of the SSL network. To address this issue, in this paper, we propose an innovative semi-supervised network with BAlanced Subclass regularIzation and semantic-Conflict penalty mechanism (BASIC) to effectively learn the unbiased knowledge for semi-supervised MoS. Concretely, we construct a novel auxiliary subclass segmentation (SCS) task based on priorly generated balanced subclasses, thus deeply excavating the unbiased information for the main MoS task with the fashion of multi-task learning. Additionally, based on a mean teacher framework, we elaborately design a balanced subclass regularization to utilize the teacher predictions of SCS task to supervise the student predictions of MoS task, thus effectively transferring unbiased knowledge to the MoS subnetwork and alleviating the influence of the class-imbalance problem. Considering the similar semantic information inside the subclasses and their corresponding original classes (i.e., parent classes), we devise a semantic-conflict penalty mechanism to give heavier punishments to the conflicting SCS predictions with wrong parent classes and provide a more accurate constraint to the MoS predictions. Extensive experiments conducted on two publicly available abdominal datasets, i.e., the WORD dataset and the MICCAI FLARE 2022 dataset, have verified the superior performance of our proposed BASIC compared to other state-of-the-art methods.

RATIONALE AND OBJECTIVES:Accurate identification of active multiple sclerosis (MS) lesions is essential for guiding treatment decisions and monitoring disease response. MATERIALS AND METHODS:To evaluate the performance of T1rho in differentiating active from inactive MS lesions. A total of 275 (27 active, 248 inactive) lesions from patients with relapsing-remitting MS were included. T1rho and quantitative magnetic resonance imaging (MRI) parameters, including T2 relaxation, and diffusion metrics (apparent diffusion coefficient [ADC], fractional anisotropy [FA], mean diffusivity [MD], radial diffusivity [RD], axial diffusivity [AD], were measured for each lesion. Differences between lesion types were assessed using linear mixed-effects models. Discriminative performance was evaluated using receiver operating characteristic (ROC) analysis, and univariate, multivariate, and stepwise logistic regression were applied to identify the optimal MRI combination for lesion differentiation. RESULTS:Active lesions demonstrated significantly lower T1rho values than inactive lesions (91.70 ± 10.19 ms vs. 114.04 ± 28.49 ms, P < 0.01). T1rho showed the highest discriminative performance (area under the curve [AUC] 0.83, 95% confidence interval [CI] 0.78-0.87), outperforming T2 and all diffusion metrics. Both T1rho and ADC were independent predictors of lesion activity, with their combined model achieving excellent discriminatory performance (AUC 0.85, 95% CI 0.81-0.89). CONCLUSION:T1rho imaging is a highly promising non-contrast technique for differentiating active MS lesions, demonstrating superior performance compared with conventional T2 mapping and diffusion metrics. Combining T1rho with ADC provides an approach that may further improve lesion discrimination.

Background/Objectives: Corpus callosum atrophy is a well recognized feature of multiple sclerosis (MS), which has been associated with disease duration and severity. Quantitative T1rho imaging is an MRI technique sensitive to microstructural tissue alterations, but the T1rho characteristics of the corpus callosum in MS remain largely unexplored. This study aimed to characterize T1rho in the corpus callosum in MS patients, and examine relationships between T1rho values, callosal atrophy, and clinical and imaging markers of disease burden. Method: Thirty patients with relapsing-remitting multiple sclerosis (RRMS) and 30 healthy controls underwent T1rho imaging. T1rho values were measured in the genu, body, splenium, and whole corpus callosum. Comparisons between patient and control groups were performed using the Mann–Whitney U test and the independent samples t-test, and Pearson’s and Spearman’s correlation coefficients were used to evaluate the correlation between imaging and clinical measures. Results: T1rho values were significantly increased across all corpus callosum subregions in patients compared to controls. Corpus callosum T1rho values showed correlations with disease duration (r = 0.375–0.408, p < 0.041), lesion load (r = 0.726–0.810, p < 0.001), and the corpus callosum index (CCI) (r = −0.731–−0.640, p < 0.001). Conclusions: T1rho values are elevated in the corpus callosum of MS patients and associated with disease duration, lesion load, and callosal atrophy.

Background:Macromolecular proton fraction (MPF) is a promising noninvasive biomarker for staging liver fibrosis. However, current post-processing requires B1-inhomogeneity acquisition and manual region of interest (ROI) selection, introducing subjectivity and variability. In this study, we propose a deep learning pipeline that enables liver MPF quantification without subject-specific B1 acquisition by leveraging an atlas-derived B1 map constructed from measured B1 data. Methods:This retrospective study used data collected at one institution from April 2019 to October 2019. The pipeline contains three models: a segmentation network for obtaining liver masks, a registration network for aligning the atlas B1 map with liver masks, and a quantification network for MPF quantification. An uncertainty-guided strategy is proposed to automatically select ROIs for assessing liver MPF. The accuracy of MPF quantification was evaluated using mean absolute error (MAE), structural similarity index measure (SSIM), and peak signal-to-noise ratio (PSNR). Reproducibility of liver MPF measurement was assessed by comparing manually selected ROIs from two experts with automated ROIs, employing intraclass correlation coefficient (ICC) and Bland-Altman analysis. Results:The study included 44 patients (mean age, 59.4±9.7 years; 20 male patients, 24 female patients). MAE for MPF quantification within the whole liver and ROI are 0.49%±0.31% and 0.45%±0.26%, respectively. ICC of liver MPF assessments are 0.931 [95% confidence interval (CI): 0.887, 0.963] between expert 1 analyst and automated analysis, 0.949 (95% CI: 0.924, 0.971) between the expert 2 analyst and automated analysis, and 0.938 (95% CI: 0.903, 0.959) between expert 1 and expert 2 analyst. Mean bias [95% limits of agreement (LOA)] were 0.042% (-0.514%, 0.598%), -0.029% (-0.445%, 0.388%), and 0.033% (-0.497%, 0.563%) for expert 1 vs. automated analysis, expert 2 vs. automated analysis, and expert 1 vs. expert 2, respectively. Conclusions:The proposed deep learning pipeline enables liver MPF quantification without subject-specific B1 acquisition by employing an atlas-based B1 substitution strategy, while maintaining high reproducibility in a fully automated manner.

Objective Recent advancements in deep learning (DL) have advanced knee cartilage segmentation in Magnetic Resonance Imaging (MRI), offering scalable, automated solutions that markedly reduce reader time and address the limitations of traditional manual approaches. Automated segmentation can substantially aid osteoarthritis (OA) assessment using MRI, facilitating consistent, reproducible quantification across large longitudinal cohorts, reduces inter-/intra-observer variability, capabilities that are impractical with manual workflows. Method This study presents a concise review of state-of-the-art DL-based approaches for knee cartilage segmentation, focusing on the evaluation of various architectures, techniques, and their adaptability to diverse datasets and imaging protocols. This review highlights key challenges in knee cartilage segmentation, including data scarcity, domain shifts, and imaging variability, while also discussing proposed solutions such as semi-supervised learning, domain adaptation, augmentation strategies, and foundation models. Additionally, the clinical significance of knee cartilage segmentation is underscored through its diverse applications. Results The study highlights substantial improvements against conventional methods in segmentation accuracy and efficiency using DL-based methods, given challenging scenarios of knee MRI. Solutions to key challenges are presented, and clinical applications showcase the potential of automated segmentation for cartilage thickness mapping and OA assessment. Conclusion DL-based segmentation is advancing musculoskeletal imaging by offering reliable and automated solutions. Despite persistent challenges such as data scarcity, domain shifts, and imaging variability, advancements in areas like semi-supervised learning, domain adaptation, augmentation strategies, and foundation models present significant opportunities to enhance model robustness and expand clinical applicability.

PURPOSE:Recent studies have shown that spin-lock MRI can simplify quantitative magnetization transfer (MT) by eliminating its dependency on water pool parameters, removing the need for a T1 map in macromolecular proton fraction (MPF) quantification. However, its application is often limited by the requirement for long radiofrequency (RF) pulse durations, which are constrained by RF hardware capabilities despite remaining within specific absorption rate (SAR) safety limits. METHODS:To address this challenge, we propose a novel method, MPF mapping using pulsed spin-lock (MPF-PSL). MPF-PSL employs a pulsed spin-lock train with intermittent free precession periods, enabling extended total spin-lock durations without exceeding hardware and specific absorption rate limits. A comprehensive analytical framework was developed to model the magnetization dynamics of the two-pool MT system under pulsed spin-lock, demonstrating that MPF-PSL achieves MT-specific quantification while minimizing confounding effects from the water pool. The proposed method is validated with Bloch-McConnell simulations, phantoms, and in vivo studies at 3T. RESULTS:Both Bloch-McConnell simulations and phantom validation demonstrated that MPF-PSL exhibits insensitivity to water pool parameters while enabling robust MPF quantification. In vivo validation studies confirmed the method's clinical utility in detecting collagen deposition in patients with liver fibrosis. CONCLUSION:MPF-PSL presents a practical solution for quantitative MT imaging, with strong potential for clinical applications.

Segmentation of cartilages to examine the Knee osteoarthritis is a challenging problem in medical image analysis, due to distribution gaps in source and target domains. Existing models trained on one MRI domain struggle to generalize to MRI scans produced from different scanners, highlighting the need to develop novel approaches to adapt to cross-modalities. In this paper, we propose a novel Eigen Low-rank subspace-assisted Mean Teacher Knowledge Distillation framework (MTKD-LRS) using a semi-supervised learning approach. This framework leverages the low-rank approximations within the deep feature subspaces to capture meaningfullatent patterns and construct the domain-invariant feature representations. By preserving robust feature maps associated with larger singular values and leveraging the lower singular value feature maps as successive truncated noise, the student model is optimized with more robust supervision to bridge the gap between the cross-modality MRI data. Extensive experiments on public and private datasets demonstrate the effectiveness of MTKD-LRS over existing state-of-the-art approaches.

Background Generalized knee tissue segmentation, such as cartilage and meniscus in magnetic resonance imaging (MRI), plays a vital role in the clinical assessment of knee osteoarthritis (OA). However, domain variability between MRI datasets poses a significant challenge for the application of robust segmentation methods in real-world clinical settings. Existing unsupervised domain adaptation (UDA) approaches, which rely on one-to-one assumptions between the source and target domains, often fail to preserve knee tissues such as cartilage and meniscus, which are critical for OA diagnosis in diverse clinical settings. Methods We propose a source-independent segmentation approach tailored for multi-domain knee MRI datasets. Our method emphasizes knee tissue regions to reduce domain gaps and label inconsistencies. By introducing a stepwise adaptation strategy, segmentation performance was refined progressively from intermediate domains to the final target domain. Pseudo-label attention mechanisms were integrated into the adaptation pipeline, enabling iterative fine-tuning of domain-specific segmentations while leveraging unidirectional generative adversarial networks to enhance tissue-specific adaptation. This iterative training process ensures the generation of reliable pseudo-labels, thereby improving segmentation accuracy in diverse clinical MRI datasets. Results We demonstrated the effectiveness of our approach on the OA initiative dataset as the source domain and self-collected, T1-weighted fast field echo (T1FFE) as the intermediate domain and three-dimensional fast spin echo (3D FSE) as the final target domain. Our method achieved an average dice scores of 0.8701 and 0.7990 for source and target domains, respectively, surpassing the typical UDA methods explored in our experiments. Conclusion The experiments conducted on clinical MRI data, spanning OA severity from healthy knees to KL Grades 1-4, validated the effectiveness of the proposed domain adaptation method in precise segmentation of the cartilage and meniscus.

Knee cartilage segmentation for Knee Osteoarthritis (OA) diagnosis is challenging due to domain shifts from varying MRI scanning technologies. Existing cross-modality approaches often use paired order matching or style translation techniques to align features. Still, these methods can sacrifice discrimination in less prominent cartilages and overlook critical higher-order correlations and semantic information. To address this issue, we propose a novel framework called Successive Eigen Noise-assisted Mean Teacher Knowledge Distillation (SEN-MTKD) for adapting 2D knee MRI images across different modalities using partially labeled data. Our approach includes the Eigen Low-rank Subspace (ELRS) module, which employs low-rank approximations to generate meaningful pseudo-labels from domain-invariant feature representations progressively. Complementing this, the Successive Eigen Noise (SEN) module introduces advanced data perturbation to enhance discrimination and diversity in small cartilage classes. Additionally, we propose a subspace-based feature distillation loss mechanism (LRBD) to manage variance and leverage rich intermediate representations within the teacher model, ensuring robust feature representation and labeling. Our framework identifies a mutual cross-domain subspace using higher-order structures and lower energy latent features, providing reliable supervision for the student model. Extensive experiments on public and private datasets demonstrate the effectiveness of our method over state-of-the-art benchmarks. The code is available at github.com/AmmarKhawer/SEN-MTKD.

Cardiac MRI, crucial for evaluating heart structure and function, faces limitations like slow imaging and motion artifacts. Undersampling reconstruction, especially data-driven algorithms, has emerged as a promising solution to accelerate scans and enhance imaging performance using highly under-sampled data. Nevertheless, the scarcity of publicly available cardiac k-space datasets and evaluation platform hinder the development of data-driven reconstruction algorithms. To address this issue, we organized the Cardiac MRI Reconstruction Challenge (CMRxRecon) in 2023, in collaboration with the 26th International Conference on MICCAI. CMRxRecon presented an extensive k-space dataset comprising cine and mapping raw data, accompanied by detailed annotations of cardiac anatomical structures. With overwhelming participation, the challenge attracted more than 285 teams and over 600 participants. Among them, 22 teams successfully submitted Docker containers for the testing phase, with 7 teams submitted for both cine and mapping tasks. All teams use deep learning based approaches, indicating that deep learning has predominately become a promising solution for the problem. The first-place winner of both tasks utilizes the E2E-VarNet architecture as backbones. In contrast, U-Net is still the most popular backbone for both multi-coil and single-coil reconstructions. This paper provides a comprehensive overview of the challenge design, presents a summary of the submitted results, reviews the employed methods, and offers an in-depth discussion that aims to inspire future advancements in cardiac MRI reconstruction models. The summary emphasizes the effective strategies observed in Cardiac MRI reconstruction, including backbone architecture, loss function, pre-processing techniques, physical modeling, and model complexity, thereby providing valuable insights for further developments in this field.

T1rho imaging showed potential applications in cancer imaging but little research explored the underlying biological processes that contribute to the T1rho values in cancer. This study aimed to investigate the potential associations between quantitative imaging biomarkers from T1rho imaging and the well-established diffusion weighted imaging (DWI), with tumour-stromal, immunohistochemical (IHC), and tumour-infiltration-lymphocytes (TIL) biomarkers in nasopharyngeal carcinoma (NPC). Pre-treatment T1rho and DWI imaging of primary NPCs were performed in 50 prospectively recruited patients. The mean T1rho and apparent diffusion coefficient (ADC) of NPC were obtained and correlated with tumour-stromal, IHC, TIL biomarkers using the Pearson Correlation test and the coefficients (R) were calculated. The mean T1rho values negatively correlated with collagenous stroma-lymphoid stroma (R=-0.314, p = 0.03) and positively correlated with percentage of tumour cells positive for Ki-67 (R = 0.402, p < 0.01), but there were no associations between T1rho values and the other tumour-stromal, IHC or TIL biomarkers (p = 0.16–0.98) or between ADC values and any of these biomarkers (p = 0.07–0.82). Our results showed the possible underlying biological mechanisms of T1rho imaging in head and neck cancer. T1rho imaging negatively correlated with the ratio of collagenous to lymphoid stroma, and positively correlated with tumour cell proliferation, which are both known to be predictors of outcome, suggesting that T1rho imaging may have a valuable role in head and neck cancer imaging. As this is a preliminary study with small sample size, further studies are encouraged to validate our findings.

Purpose: Inhomogeneous magnetization transfer (ihMT) effect reflects dipolar order with a dipolar relaxation time ($T_{1D}$), specific to motion-restricted macromolecules. We aim to quantify $T_{1D}$ using spin-lock MRI implemented with a novel rotary-echo sequence. Methods: In proposed method, we defined a relaxation rate $R_{dosl}$ that is specific to dipolar order and obtained as the difference of dual-frequency $R_{1rho}^{dual}$ relaxation and single-frequency $R_{1rho}^{single}$ relaxation. A novel rotary-echo spin-lock sequence was developed to enable dual-frequency acquisition. We derive the framework to estimate $T_{1D}$ from $R_{dosl}$ under macromolecular pool fraction (MPF) map constraints. The proposed approach was validated via Bloch-McConnell-Provotorov simulation, phantom studies, and in-vivo white matter studies on a 3T scanner. Results: Simulations demonstrated that $R_{dosl}$ exhibits an approximately linear relationship with $T_{1D}$. Phantom experiments showed robust ihMT contrast in $R_{dosl}$ and confirmed the feasibility and reliability of $T_{1D}$ quantification via $R_{dosl}$. In vivo white-matter studies further supported the clinical potential of this $T_{1D}$ mapping approach. Conclusion: We propose a novel, clinical feasible method for $T_{1D}$ quantification based on spin-lock MRI. This method requires substantially fewer contrast-prepared images compared to the conventional $T_{1D}$ quantification approach. This technique provides a promising pathway for robust MPF and $T_{1D}$ quantification in a single rapid scan with reduced confounds.

Imaging features of knee articular cartilage have been shown to be potential imaging biomarkers for knee osteoarthritis. Despite recent methodological advancements in image analysis techniques like image segmentation, registration, and domain-specific image computing algorithms, only a few works focus on building fully automated pipelines for imaging feature extraction. In this study, we developed a deep-learning-based medical image analysis application for knee cartilage morphometrics, CartiMorph Toolbox (CMT). We proposed a 2-stage joint template learning and registration network, CMT-reg. We trained the model using the OAI-ZIB dataset and assessed its performance in template-to-image registration. The CMT-reg demonstrated competitive results compared to other state-of-the-art models. We integrated the proposed model into an automated pipeline for the quantification of cartilage shape and lesion (full-thickness cartilage loss, specifically). The toolbox provides a comprehensive, user-friendly solution for medical image analysis and data visualization. The software and models are available at https://github.com/YongchengYAO/CMT-AMAI24paper.

Background:While conventional magnetic resonance imaging (MRI) in multiple sclerosis (MS) primarily evaluates focal lesions, the normal-appearing white matter (NAWM) encompasses brain tissue that appears radiologically normal but harbors subtle pathological changes that contribute to the overall disease burden. The aim of this study was to investigate the role of T1rho MRI in characterising distance-dependent microstructural changes in the perilesional NAWM in patients with relapsing-remitting multiple sclerosis (RRMS). Methods:T1rho and diffusion tensor imaging (DTI) images were acquired from 30 patients with RRMS and 30 age-matched healthy controls. A total of 217 non-contrast-enhancing MS lesions were identified, and five perilesional layers were delineated from the lesion margins. T1rho values in the intralesional and perilesional regions in MS patients and the corresponding normal white matter of controls were quantified and compared. Results:T1rho values progressively decreased from the lesional regions (111.82±27.58 ms) to the perilesional layer 5 (78.06±4.76 ms) (P value <0.05). Significant correlations were found between T1rho values and mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) in the perilesional areas, with a pattern of progressively lower correlation coefficients (r, 0.23 to 0.73, P value <0.01) between T1rho values and the DTI metrics as the distance from the lesion increased. Conclusions:Our findings suggest that T1rho is a sensitive method for the detection of microstructural changes in the NAWM, and may provide valuable information about the spatial extent and severity of these changes.

The meniscus, a fibrocartilaginous structure within the knee joint, plays an essential role in joint stability and the prevention of knee osteoarthritis (OA). Accurate segmentation of the meniscus from magnetic resonance imaging (MRI) is crucial for early diagnosis and monitoring of OA progression. However, manual segmentation is labor-intensive, while automatic approaches face challenges due to variability in meniscal morphology, partial volume effects, and low tissue contrast.To address these challenges, we propose ERANet, a semi-supervised framework that effectively leverages both labeled and unlabeled data through anatomically guided augmentation, consistency regularization, and iterative pseudo label refinement. Central to ERANet is edge replacement augmentation (ERA), a meniscus-specific augmentation strategy that introduces plausible morphological perturbations by modifying peripheral meniscal regions with context-aware background information. ERA is tailored to address the unique anatomical variability of meniscal structures. Alongside ERA, ERANet incorporates two generalizable learning modules: prototype consistency alignment (PCA), which enforces feature compactness via prototype-guided regularization, and conditional self-training (CST), which selectively incorporates reliable pseudo labels based on their temporal stability. The synergistic interaction among these modules enables ERANet to handle small, low-contrast anatomical structures with limited supervision.We validated ERANet on 3D DESS and 3D FSE MRI sequences, demonstrating superior segmentation performance compared to state-of-the-art semi-supervised methods. ERANet maintains high accuracy even with minimal labeled data, and extensive ablation studies confirm the individual and combined benefits of ERA, PCA, and CST. Our results suggest that ERANet offers a robust and scalable solution for meniscus segmentation. Code is available at https://github.com/SiYueLi-MRIandAI/ERANet.