Accurate segmentation of breast tumors in Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) is critical for early cancer diagnosis and treatment planning. However, prevailing segmentation methods frequently fail to fully leverage the temporal enhancement dynamics inherent in DCE-MRI, which are essential for characterizing tumor pathology. To address this limitation, we propose STMNet, a novel deep learning framework designed for explicit 4D spatio-temporal modeling in breast DCE-MRI. The network incorporates a hierarchical encoder featuring dedicated Spatial Mamba and Temporal Mamba modules, which independently capture intra-phase spatial contexts and inter-phase temporal dependencies across multiple scales. Furthermore, to augment the representation of temporal features, we introduce a Global Temporal Attention Feature Fusion (GTAFF) module that constructs a global query to guide multi-phase integration in a clinically inspired manner. Extensive evaluations on two distinct breast DCE-MRI datasets demonstrate that STMNet consistently surpasses state-of-the-art methods, achieving superior segmentation accuracy and robustness across tumors with diverse morphological characteristics and enhancement kinetics. The code is available at STMNet-DCE-MRI

Objective.Accurate lung motion estimation from 4D computed tomography (CT) is essential for image-guided radiotherapy and thoracic motion analysis, but remains challenging in vessel-rich regions where weak contrast, fine structures, and heterogeneous mechanics limit conventional registration. This work aims to develop an anatomy-informed finite-element digital volume correlation (FE-DVC) framework for accurate and mechanically plausible lung motion estimation.Approach.We propose a structure-tensor-guided heterogeneous anisotropic FE-DVC method. Structure tensors extracted from the reference CT image are used as heuristic anatomical priors to modulate element-wise regularization strength and align an effective anisotropic regularization frame with bronchovascular directions. A lung-specific multi-mesh strategy uses finer elements in vessel-containing regions and coarser elements in parenchyma. A practical workflow based on L-curve analysis and Jacobian-based deformation regularity is introduced to reduce empirical parameter tuning. The method was evaluated on three lung 4D-CT datasets and compared with Demons and pTV registration.Main results.The proposed method consistently reduced landmark target registration error across all datasets, with most errors concentrated below 2 mm and fewer large outliers. In a representative DIR-Lab case, normalized correlation residuals were reduced to 0.059 and 0.100 in axial and coronal views, compared with 0.172/0.256 for Demons and 0.088/0.138 for pTV. The recovered strain fields showed coherent vessel-oriented deformation, and boundary-driven FE simulations confirmed vessel-aligned strain concentrations only under heterogeneous anisotropic regularization. Singular value decomposition further showed that the first motion mode explained more than 90% of deformation energy and the first three modes exceeded 99%.Significance.This framework integrates CT-derived anatomical organization with mechanically interpretable regularization, improving vessel-scale lung motion recovery and supporting future strain-based biomarkers and pulmonary structure-function modeling.

Accurate and biomechanically consistent quantification of cardiac motion remains a major challenge in cine MRI analysis. While classical feature-tracking and recent deep learning methods have improved frame-wise strain estimation, they often lack biomechanical interpretability and temporal coherence. In this study, we propose a spacetime-regularized finite-element digital image/volume correlation (FE-DIC/DVC) framework that enables 2D/3D+T myocardial motion tracking and strain analysis using only routine cine MRI. The method unifies Multiview alignment and 2D/3D+T motion estimation into a coherent pipeline, combining region-specific biomechanical regularization with data-driven based temporal decomposition to promote spatial fidelity and temporal consistency. A correlation-based Multiview alignment module further enhances anatomical consistency across short- and long-axis views. We evaluate the approach on one synthetic dataset (with ground-truth motion and strain fields), three public datasets (with ground-truth landmarks or myocardial masks), and a clinical dataset (with ground-truth myocardial masks). 2D+T motion and strain are evaluated across all datasets, whereas Multiview alignment and 3D+T motion estimation is assessed only on the clinical dataset. Compared with two classical feature-tracking methods and four state-of-the-art deep-learning baselines, the proposed method improves 2D+T motion and strain estimation accuracy as well as temporal consistency on the synthetic data, achieving a displacement RMSE of 0.35 pixels (vs. 0.73 pixels), an equivalent-strain RMSE of 0.05 (vs. 0.097), and a temporal consistency of 0.97 (vs. 0.91). On public and clinical data, it achieves superior performance in terms of a landmark error of 1.96 mm (vs. 3.15 mm), a boundary-tracking Dice of 0.80-0.87 (a 2-4% improvement over the best-performing baseline), and overall registration quality that consistently ranks among the top two methods. By leveraging only standard cine MRI, this work enables 2D/3D+T myocardial mechanics and provides a practical route toward 4D cardiac function assessment.

While the discovery of the glymphatic system has greatly advanced our understanding of waste clearance and fluid dynamics in central nervous system diseases, the neurofluid dynamics in pediatric acute leukemia remain to be elucidated. This study sought to evaluate MRI markers putatively related to neurofluid dynamics in pediatric acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML), including perivascular space (PVS) burden, free water (FW) fraction, and diffusion tensor imaging along the PVS (DTI-ALPS). Seventy-two children with acute leukemia and 72 age- and sex-matched typically developing (TD) children (50 ALL and TDs1; 22 AML and TDs2) were included in this prospective study. Group differences in brain volumetric measures and glymphatic-related MRI markers were evaluated. In addition, MRI metrics that differed significantly between groups were further examined for associations with clinical variables and total sleep scale scores using partial correlation analyses. Compared to TDs, ALL and AML showed no significant differences in intracranial volume. Compared with TDs, decreased brain parenchymal volume and gray matter volume were found in both children with ALL and AML (all FDR-corrected P ≤ 0.04). Compared with TDs, the ALL group exhibited reduced white matter volume and increased cerebrospinal fluid volume (all FDR-corrected P ≤ 0.001), while the AML group showed no significant differences in these measures. For glymphatic-related MRI markers, decreased PVS volume and count, FW value, and DTI-ALPS index were observed in both ALL and AML (all FDR-corrected P ≤ 0.02). In addition, the DTI-ALPS index was negatively correlated with risk stratification and total scale scores in children with ALL (all P ≤ 0.03). This preliminary, cross-sectional study identified a neuroimaging pattern in pediatric acute leukemia, characterized by reduced brain parenchyma volume and alterations in MRI markers putatively linked to neurofluid dynamics. In patients with ALL, a lower ALPS index was correlated with higher clinical risk stratification and greater sleep disturbance. These observations support future studies to clarify the biological relevance of neurofluid-related imaging features in acute leukemia. ChiCTR2000031353; registered date: 2020-04.

Preoperative distinction between follicular thyroid carcinoma (FTC) and follicular thyroid adenoma (FTA) remains a clinical challenge, largely due to the substantial cytomorphological overlap observed in fine-needle aspiration samples. We hypothesised that deep learning could capture subvisual cytomorphological patterns that reflect underlying biological differences between these entities.Preoperative distinction between follicular thyroid carcinoma (FTC) and follicular thyroid adenoma (FTA) remains a clinical challenge, largely due to the substantial cytomorphological overlap observed in fine-needle aspiration samples. entities. We hypothesised that deep learning could capture subvisual cytomorphological patterns that reflect underlying biological differences between these entities. This retrospective multicentre study enrolled 255 patients with surgically confirmed follicular thyroid neoplasms (FTNs) across 10 institutions, all of whom underwent preoperative liquid-based cytology (LBC) testing. Patients were allocated to a training cohort (n = 127; 48,560 patches), an internal validation cohort (n = 44; 17,226 patches), and an external validation cohort (n = 84; 17,192 patches). We developed a multi‑bag clustering‑ constrained attention multiple instance learning (CLAM_MB) model to classify LBC whole‑slide images. The model leveraged pathologist‑annotated follicular cell clusters to guide instance sampling without requiring instance‑level labels, thereby enabling phenotype pattern discovery under weak supervision. Model performance was benchmarked against single‑bag CLAM (CLAM_SB) and max‑pooling multiple instance learning (MIL) architectures. The CLAM_MB model demonstrated robust classification performance across all cohorts, with no significant performance degradation between validation settings (all P > 0.05). In the internal validation cohort, CLAM_MB achieved an area under the receiver operating characteristic curve (AUC) of 0.817, significantly outperforming CLAM_SB (0.766) and max‑pooling MIL (0.786; both P < 0.001). Similarly, in the external validation cohort, CLAM_MB yielded an AUC of 0.827, compared with 0.796 for CLAM_SB and 0.619 for max‑pooling MIL (both P < 0.001). At optimal cutoffs, the model attained sensitivities of 68.8% and 81.8%, and specificities of 78.4% and 92.9% in the internal and external validation cohorts, respectively. Subgroup analyses further confirmed consistent diagnostic performance across age and sex strata (both P > 0.05). The CLAM_MB model effectively identifies subvisual cytomorphological signatures in LBC samples that distinguish FTC from FTA with high accuracy. These findings support its potential utility as a preoperative decision-support tool for the differential diagnosis of FTNs, warranting further prospective validation. None

PURPOSE:Contrast-enhanced magnetic resonance image (MRI) imaging via administration of contrast agents is critical for diagnosis, staging, and treatment of nasopharyngeal carcinoma (NPC). However, gadolinium-based contrast agents can lead to severe adverse effects, especially in patients with compromised kidney function, necessitating a safer alternative for contrast enhancement. In this study, we aim to develop and assess the clinical feasibility of a federated learning model for synthesizing virtual contrast-enhanced MRI (VCE-MRI) images from contrast-free scans for patients with NPC. METHODS AND MATERIALS:In this multicenter, retrospective study, we developed and clinically evaluated a federated learning-based VCE-MRI synthesis model (FL-VCE-MRI) using pretreatment contrast-free T1-weighted and T2-weighted MRI scans. The model was trained using data from 14 centers involving 2061 patients. External validation was performed with an independent dataset from 25 centers comprising 126 patients. Additionally, 10 clinicians from 8 centers assessed the quality of the synthetic images. RESULTS:The FL-VCE-MRI model demonstrated high generalizability in the external validation dataset, achieving a mean absolute error of 45.85 (95% CI, 43.73-47.96). For centers facing challenges in developing well-performing single-center models, the FL-VCE-MRI improved the average mean absolute error from 53.93 (95% CI, 49.21-58.65) to 45.93 (95% CI, 41.64-50.22). Clinical evaluations indicated that the synthesized VCE-MRI images are both reliable and clinically valuable, with no significant differences compared with gadolinium-enhanced MRI in disease diagnosis, tumor staging, and delineation. CONCLUSIONS:The FL-VCE-MRI model shows potential as a noninvasive alternative to gadolinium-enhanced MRI for patients with NPC.

OBJECTIVES:To compare the imaging performance of 5.0 T and 3.0 T susceptibility-weighted imaging (SWI) in visualizing cerebral veins and deep gray matter nuclei, and to assess the potential clinical implications of 5.0 T SWI. METHODS:Eight healthy adult volunteers underwent SWI scans on both 5.0 T and 3.0 T MRI systems. Two experienced radiologists assessed overall image quality and the visualization of cerebral veins and deep gray matter nuclei qualitatively (grades 1-4) and quantitatively (SNR and CNR). The detection rate of the "swallow tail" sign of the substantia nigra was also evaluated. RESULTS:The 5.0 T SWI demonstrated significantly higher image quality, with CNR improvements exceeding 50% for cerebral veins and more than doubling for multiple deep gray matter nuclei structures compared to 3.0 T. The detection rate of the "swallow tail" sign was 81.25% on 5.0 T, compared to 50% on 3.0 T. However, 5.0 T exhibited more susceptibility artifacts in regions near air-tissue interfaces. CONCLUSIONS:SWI at 5.0 T provides superior visualization of cerebral veins and specific deep gray matter nuclei compared to 3.0 T, suggesting its significant potential in enhancing anatomical visualization in neuroimaging research settings. ADVANCES IN KNOWLEDGE:This study presents a systematic comparison of 5.0 T and 3.0 T SWI in healthy adults, demonstrating that 5.0 T significantly enhances the visualization of cerebral veins and specific deep gray matter nuclei, offering superior contrast and resolution. These improvements could facilitate more precise neuroanatomical assessments and hold promise for advancing research in neurodegenerative disease.

ObjectiveTo evaluate the diagnostic performance of radiomics features extracted from diffusion-derived Vessel Density (DDVD) in differentiating hepatocellular carcinoma (HCC) from hepatic hemangioma (HG).MethodsThis retrospective study enrolled 232 patients (104 with pathologically confirmed HCCs and 128 with clinically diagnosed HGs). The cohort was randomly divided into training and testing sets (7:3 ratio). We generated DDVD maps (subtraction maps of b0-b50 voxel-by-voxel). Features were extracted from b0, b50, b800, ADC, and DDVD maps, respectively. Feature selection was sequentially performed for each type of image using Mann-Whitney U test, Pearson correlation (|r| > 0.8), and LASSO regression. Five Logistic regression models (b0, b50, b800, ADC, and DDVD) were independently constructed to differentiate HCC from HG, with model performance evaluated using receiver operating characteristic analysis, with AUC, sensitivity, specificity, NPV, PPV, and accuracy as primary metrics. Delong test was utilized to evaluate the difference in performance of models.ResultsFrom a total of 1,197 features initially extracted, 10 most informative features from each image type were retained. The DDVD-based model demonstrated comparable performance to b0, b50, and ADC models, achieving AUC values of 0.926 (95% CI: 0.914 - 0.929) in the validation cohort and 0.977 (95% CI: 0.948 - 1.000) in the independent test cohort. Comparative analysis revealed that the b0, b50, ADC, and DDVD models significantly outperformed the b800 model in the test cohort (all p < 0.05).ConclusionsDDVD-based radiomics demonstrates an effective approach for differentiating HCC from HG by quantifying spatial heterogeneity in microvascular characteristics.

Purpose:To investigate the value of radiomics features extracted from plain and enhanced spectral CT-derived metrics in differentiating osteoblastic bone metastasis (OBM) and bone island (BI) in newly diagnosed cancer patients. Methods:From January to November 2020, 51 newly diagnosed cancer patients with 204 bone lesions (OBM = 116, BI = 88) receiving spectral CT were retrospectively enrolled. 40-140 keV mono-energy images were generated from plain CT and contrast-enhanced CT, and material-decomposition images, including water (calcium) and calcium (water) substrate density images from plain CT and Iodine (calcium) substrate density images from contrast-enhanced CT. Radiomics features were extracted from the manually segmented lesions, including shape feature set, material-separation feature set, plain spectral CT feature set, and enhanced spectral CT feature set. U-test and LASSO analysis were sequentially used to select the most relevant features. The shape model, material-separation model, plain CT model, contrast-enhanced CT model, and combined model were built using Random Forest with model performance evaluated using ROC analysis and compared using the Delong test. Results:After feature selection, four features were selected for the shape set, seven features for the material-separation set, seven features for the plain spectral CT set, and nine features for the enhanced spectral CT set. The AUC of the shape model was significantly smaller than that of the other four models (all P < 0.05). The combined model (AUC = 0.874, 95%CI: 0.821-0.916) outperformed the material-separation model (AUC = 0.828, 95%CI: 0.769-0.877, P = 0.005), the plain spectral CT model (AUC = 0.820 95%CI: 0.760-0.870, P = 0.005) and the enhanced spectral CT model (AUC = 0.838, 95%CI: 0.780-0.886, P = 0.005). Conclusion:The radiomics features derived from spectral CT metrics will enhance the differentiation of de novo OBM and BI in newly diagnosed cancer patients.

Preoperative ternary stratification of lung adenocarcinoma-spectrum nodules remains challenging, particularly for minimally invasive adenocarcinoma (MIA). We developed and externally validated a multimodal deep-learning framework integrating three-dimensional CT nodule patches with structured radiological semantic features to stratify atypical adenomatous hyperplasia/adenocarcinoma in situ (AAH/AIS), MIA, and invasive adenocarcinoma (IAC). Consecutive patients with surgically resected, pathologically confirmed nodules were retrospectively enrolled from three centers, with postoperative pathology as the reference standard. The Center 1 development cohort included 2004 patients/2208 nodules and was split at the patient level into training (1603/1764) and internal validation (401/444); external validation used Center 2 (446/483) and Center 3 (276/378). Internally, the multimodal model achieved an AUC of 0.914 (95% CI, 0.894–0.932), exceeding CT image-only (0.874; absolute gain, 0.040; P < 0.001) and semantic-only (0.867; absolute gain, 0.047; P < 0.001) models. External AUCs were 0.879 (0.856–0.902) in Center 2 and 0.895 (0.868–0.917) in Center 3, with significant improvements over CT image-only and semantic-only models. These findings support structured radiological semantic features as a clinically interpretable complementary input to CT representations and could inform preoperative invasiveness stratification in decision-support workflows for preoperatively suspected and surgically considered lung adenocarcinoma-spectrum nodules.

Background:The widespread use of breast magnetic resonance imaging (MRI) for monitoring tumor response to neoadjuvant chemotherapy (NAC) is restricted by its limited accessibility and substantial interpretive workload. This study aimed to evaluate various abbreviated breast MRI protocols and identify the optimal approach for the early assessment of NAC response in breast cancer patients. Methods:A total of 359 patients with invasive breast cancer from three centers, who underwent full-protocol breast MRI at baseline and after two cycles of NAC, were retrospectively included in the study [primary cohort, n=169; external validation cohorts (EVCs), n=89 and 101]. Six abbreviated protocols were reconstructed from the full protocol, each comprising T2-weighted imaging plus a single dynamic contrast-enhanced (DCE) phase acquired at approximately 20, 90, or 270 s after injection; diffusion-weighted imaging (DWI) was also included in Protocols 4-6. Percentage changes in tumor size, the tumor-to-parenchyma signal enhancement ratio (SER), and the apparent diffusion coefficient (ADC) between MRI at the baseline and after two cycles of NAC (Δ%Size, Δ%SER, and Δ%ADC) were calculated. Diagnostic performance was assessed using receiver operating characteristic (ROC) analysis with the area under the curve (AUC), and compared using the DeLong test. Scan acquisition and interpretation times were also recorded to assess efficiency. Results:Δ%Size, Δ%SER, and Δ%ADC were associated with pathologic complete response (all P<0.05). The AUCs ranged from 0.734 to 0.876 in the primary cohort and from 0.759 to 0.878 and 0.771 to 0.880 in the two EVCs. Under the same enhancement duration, adding DWI significantly improved the AUC value in the primary cohort (P1 vs. P4, P=0.005; P2 vs. P5, P=0.003; P3 vs. P6, P<0.001), an improvement which was confirmed in both the EVCs (all P<0.05). Extending the post-contrast phase did not improve the AUC values among the DWI-based protocols (all P>0.05). The abbreviated protocols reduced the acquisition time by 11-65% compared with the full protocol across cohorts, while the addition of DWI increased the interpretation time. Conclusions:The abbreviated MRI protocols showed potential for the early assessment of NAC response in breast cancer and shortened the acquisition time. Prolonged enhancement phases did not yield diagnostic gains, while the inclusion of DWI improved diagnostic performance but lengthened the interpretation time. Integrating DWI with shorter enhancement may offer a balanced approach between accuracy and efficiency.

PurposeThere is a lack of research evaluating the clinical performance of virtual contrast-enhanced MRI (VCE-MRI). This study aims to assess the clinical utility of an established VCE-MRI technique in NPC patients and to establish patient selection criteria.Materials and methodsWe retrospectively collected data from 333 NPC patients across six institutions (2012-2023). VCE-MRI was synthesized from T1-weighted and T2-weighted MRI for each patient using the multimodality-guided synergistic neural network (MMgSN-Net), which was pre-trained on a large cohort of 1682 NPC patients from 14 institutions. Three experienced radiologists independently assessed image quality using a 5-point Likert scale, with scores below 4 deemed clinically unacceptable. The association between assessment results and patient clinical characteristics of corresponding patients (tumor diameter, T-stage, shape, gender, age) was analyzed to stratify the patients suitable for MMgSN-Net-based VCE-MRI.ResultsClinically acceptability of VCE-MRI significantly decreased with larger tumor diameters (p=0.001), advanced T-stage (T1: 100%, T2: 95.1%, T3: 90.0%, T4: 69.6%; p<0.001), and complex tumor shape (regular: 94.8% vs. complex: 74.0%; p<0.001). Age and gender had no significant impact (p=0.327, 0.773). 13 unacceptable patients were found due to severe VCE-MRI artifacts. VCE-MRI achieved high clinical acceptability in T1/T2 patients (109/114, 95.6%) and T3/T4 patients with regular tumor shapes (115/120, 95.8%). In contrast, acceptability significantly decreased to 73.0% (65/89) for T3/T4 tumors with complex shapes. The T-staging result discordance rate between VCE-MRI and CE-MRI is 8.4%, with overstaging in 15 cases.ConclusionMMgSN-Net-based VCE-MRI demonstrates favorable clinical feasibility in selected patient subgroups, T1/T2-stage NPC and T3/T4-stage NPC with regular tumor morphology, potentially reducing GBCAs administration in 67.3% of eligible patients when applying these stratification criteria.

Tumor hemodynamic heterogeneity holds significant potential for improving breast lesion characterization, diagnosis, and prognosis. Traditional methods typically focus on either spatial morphological heterogeneity or temporal dynamics, but rarely integrate both. This research proposes a novel methodology to concurrently visualize and quantify spatiotemporal hemodynamic heterogeneity. We generate voxel- wise time-intensity curve (TIC) maps along with wash-in and wash-out parameter maps to represent the spatial and temporal hemodynamic information. Then radiomics features was extracted to quantify these hemodynamic variations. The proposed method was demonstrated to be clinically effective, with an AUC of 0.8587 (95% CI: 0.8055-0.9056) in differentiating malignant from benign breast lesions. This novel framework offers an intuitive, self- explanatory visualization of hemodynamic heterogeneity and holds promise for broader clinical application in breast cancer management.

Physiological and external motion cause inter-frame misalignment in chemical exchange saturation transfer magnetic resonance imaging (CEST-MRI), thereby compromising quantitative accuracy. In CEST-MRI, saturation effects induce intensity variations, resulting in motion-intensity coupling that makes registration particularly challenging. To address this issue, we extend the finite element digital image correlation (FE-DIC) framework by introducing an alternating correction strategy that iteratively refines both motion and intensity estimation. Unlike conventional FE-DIC approaches that assume intensity constancy, the proposed method incorporates mechanical regularization to suppress non-physical deformations, alongside intensity correction to compensate for reference-target contrast discrepancies. This mutual reinforcement enables progressively improved registration across the CEST sequence. The robustness and effectiveness of the method were evaluated on three datasets. In simulated liver data, it maintains RMSE within 0.4 pixels, reducing error by 0.5 pixels compared to RPCA&PCA (a PCA-based synthetic reference generation method for CEST registration). On clinical brain and pig cardiac data, it achieves average SSIM of 0.83, outperforming RPCA&PCA by 0.03 and surpassing CNN-based registration (e.g., AirLab) by 0.10. The consistent results across datasets highlight its generalizability, making it a promising tool for metabolic quantification in clinical and research settings.

Accurate identification of lung adenocarcinoma subtype is crucial for selecting appropriate treatment plans. However, challenges such as integrating diverse data, subtype similarities, and capturing contextual features hinder precise differentiation. To address these challenges, we propose a deep neural network model that integrates CT images, annotated lesion bounding boxes, and electronic health records. Initially, the model leverages combining bounding boxes and CT scans to produce enhanced CT images with detailed lesion location information. These enhanced CT images then undergo feature extraction through a vision transformer module. In addition to imaging data, the model incorporates clinical information, encoded using a fully connected encoder. Features extracted from both CT and clinical data are optimized for cosine similarity using a CLIP module, ensuring their cohesive integration. To further blend features from different modalities, we devise an attention-based feature fusion module, known as DFF, to harmonize them into a unified representation. This integrated feature set is then input to a classifier that effectively differentiates among the three types of adenocarcinomas. To alleviate the effect of unbalanced classes, we incorporate contrastive learning loss and focal loss, which enhance feature representation and improve model performance. Our model achieves a superior accuracy of 81.42% and an area under the curve of 0.9120 on the validation set, significantly outperforming other recent multimodal classification methods. The code is available at https://github.com/fancccc/LungCancerDC.

This study aims to develop a novel segmentation method that utilizes spatio-temporal information for segmenting two-dimensional thyroid nodules on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Leveraging medical morphology knowledge of the thyroid gland, we designed a semi-supervised segmentation model that first segments the thyroid gland, guiding the model to focus exclusively on the thyroid region. This approach reduces the complexity of nodule segmentation by filtering out irrelevant regions and artifacts. Then, we introduced a method to explicitly extract temporal information from DCE-MRI data and integrated this with spatial information. The fusion of spatial and temporal features enhances the model's robustness and accuracy, particularly in complex imaging scenarios. Experimental results demonstrate that the proposed method significantly improves segmentation performance across multiple state-of-the-art models. The Dice similarity coefficient (DSC) increased by 8.41%, 7.05%, 9.39%, 11.53%, 20.94%, 17.94%, and 15.65% for U-Net, U-Net + + , SegNet, TransUnet, Swin-Unet, SSTrans-Net, and VM-Unet, respectively, and significantly improved the segmentation accuracy of nodules of different sizes. These results highlight the effectiveness of our spatial-temporal approach in achieving accurate and reliable thyroid nodule segmentation, offering a promising framework for clinical applications and future research in medical image analysis.

Background:Amide proton transfer (APT), a specific type of chemical exchange saturation transfer (CEST) MRI, has proved valuable in tumor diagnosis and characterization by detecting mobile protein/peptides in cancerous tissues. However, T1 confounds CEST measurements, leading to reduced specificity to amides and potential misinterpretation of APT imaging. Purpose:The study aimed to investigate the feasibility of the quasi-steady-state (QUASS)-based apparent exchange-dependent relaxation (AREX) analysis in correcting T1 for unbiased tumor APT MRI at 3T. Materials and Methods:CEST MRI experiments were conducted on an egg white phantom and on prospectively enrolled brain tumor patients with T1 values modulated by gadolinium (Gd). QUASS algorithm was employed to reconstruct steady-state Z spectra. Conventional T1-uncorrected CEST effect was quantified with a multipool Lorentzian function from QUASS Z spectra. The non-QUASS AREX and QUASS-based AREX with T1 correction were calculated from the inverse of non-QUASS and QUASS Z spectra, respectively. The student's t-test and Bland-Altman plots were performed to assess the statistical difference and consistency between pre- and post-Gd measurements. Results:In the phantom study, vials with different T1 values showed conspicuous discrepancy on the conventional uncorrected APT and non-QUASS AREX maps, but comparable contrast on the QUASS-based AREX map. In the human study, 13 patients were enrolled. The contralateral normal-appearing white matter exhibited no substantial change in T1 and similar CEST effect between uncorrected APT, non-QUASS AREX, and QUASS-based AREX pre- and post-Gd (all P > .05). However, the tumor regions showed significantly reduced T1 post-Gd that altered the CEST measurements obtained from uncorrected APT and non-QUASS AREX (both P < .001). In comparison, QUASS-based AREX measurements were in excellent agreement between pre- and post-Gd (P = .19). Conclusion:QUASS-based AREX analysis can effectively correct T1 contamination in CEST measurements, facilitating unbiased tumor APT MRI at 3T.

BACKGROUND:Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays a crucial role in the diagnosis and monitoring of cancers, as it reveals physiological and vascular characteristics of tumors. Traditional pharmacokinetic modeling necessitates high temporal resolution, resulting in relatively low signal-to-noise ratio (SNR) and spatial resolution with limited allocated time for each phase. PURPOSE:To explore the feasibility of using deep learning with sparse DCE MRI phases to generate dense temporal resolution DCE-MRI-derived parametric map. METHODS:An innovative approach, the vision transformer Pix2Pix generative adversarial network (VP-GAN), was introduced to translate the sparse DCE-MRI series into dense-phase DCE-MRI-based parametric maps, specifically targeting Ktrans and ve. The strengths of both Vision Transformers and GANs were utilized to capture complex temporal dynamics and spatial features. The proposed method was comprehensively compared with several existing deep learning models, both for the entire image and within regions of interest (ROI). Metrics used for comparison included Peak-Signal-to-Noise-Ratio (PSNR), Structural Similarity Index (SSIM), Pearson correlation analysis, and Bland-Altman analysis. Additionally, ROI histogram analysis was performed to assess the distribution of parametric values. RESULTS:The parametric maps generated by the proposed approach were qualitatively and quantitatively consistent with the reference images. The performance of the comparative studies evidenced the superiority of VP-GAN over other approaches. CONCLUSION:The proposed model performs well in converting DCE-MRI with a subset of uniformly spaced time points into physiological parametric maps derived from dense-phase DCE-MRI, allowing for DCE-MRI analysis with much fewer phases.

AIM:The microstructural diffusional heterogeneity of nasopharyngeal carcinoma (NPC) impacts outcome assessment. This study aims to investigate the potential of multidimensional diffusion-weighted magnetic resonance imaging (MDD-MRI) in evaluating microscopic diffusion characteristics of NPC. MATERIALS AND METHODS:Sixty-six NPC patients underwent MDD-MRI to derive microscopic diffusion metrics, including anisotropic mean kurtosis (MK), isotropic MK, total MK, and microscopic fractional anisotropy (FA). Macroscopic metrics, such as the apparent diffusion coefficient (ADC), along with parameters from diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI), were also analysed for comparison. Differences between NPC and normal tissues, as well as among pathological subtypes, clinical stages, and Epstein-Barr virus infection statuses, were assessed. Receiver operating characteristic curve analysis evaluated the ability to distinguish NPC from normal tissue. Spearman's correlation analysis examined associations between diffusion metrics and primary tumour (T) stages. RESULTS:NPC tissues exhibited lower ADC (P<.001) and higher DKI-related MK (MKK; P<.001) compared to normal nasopharyngeal tissues, with area under the curve values of 0.968 (95 % CI: 0.926-1) and 0.749 (95 % CI: 0.633-0.865) for ADC and MKK, respectively. Microscopic metrics and ADC correlated with T stages: microscopic FA (rho = 0.52, P<.001), anisotropic MK (rho = 0.38, P=.002), isotropic MK (rho = 0.27, P=.03), total MK (rho = 0.32, P=.009), and ADC (rho = -0.36, P=.003). CONCLUSION:MDD-MRI is a valuable tool for assessing NPC's microscopic characteristics, possibly improves T-stage evaluation beyond conventional ADC.