http://med-sci.cn/cgzcx/en/contents_list.asp?issue=202607 zh-cn An RSS feed for Chinese Journal of Magnetic Resonance Imaging http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.001 ObjectiveThe glutamate-weighted chemical exchange saturation transfer (GluCEST) imaging technique was used to evaluate the changes of glutamate (Glu) in the thalamus of children with negative MRI - confirmed epilepsy. Furthermore, the volume changes of the subregions of the thalamus in these children were explored, and the correlation between the Glu changes measured by GluCEST and the volume of the thalamic subregions was analyzed. Materials and MethodsA total of 73 pediatric patients diagnosed with epilepsy were prospectively enrolled between January 2024 and May 2026 at the Department of Pediatric Neurology, Binzhou Medical University Hospital. They were divided into the focal epilepsy (FE) group with 37 cases and the generalized epilepsy (GE) group with 36 cases. Thirty-six healthy control (HCs) subjects were recruited. The brain magnetic resonance images of the three groups were collected to exclude data with brain diseases. Those meeting the requirements were subjected to three-dimensional T1-weighted magnetization prepared rapid gradient echo (MPRAGE) imaging and GluCEST imaging. The GluCEST images were processed using Matlab software to obtain the asymmetric magnetization rate (MTR_asym) value of the thalamus region, representing the relative Glu concentration value of this area; the MPRAGE images were automatically segmented using FreeSurfer to obtain the volume of the thalamic subregions. SPSS was used to compare the MTR_asym values of the left and right thalamus in the three groups, and to compare the thalamic MTR_asym values and subregion volumes between the GE group and the HCs group, as well as between the FE group and the HCs group. ResultsThe MTR_asym value of the ipsilateral thalamus in the focal epilepsy (FE) group was significantly higher than that of the contralateral thalamus, with a statistically significant difference (t = 3.252, P = 0.002). In the generalized epilepsy (GE) group, both the left (Z = -4.944, P < 0.001) and right (t = 4.816, P < 0.001) thalamic MTR_asym values were significantly elevated compared to those in the healthy control (HC) group. Similarly, both the ipsilateral (t = 4.547, P < 0.001) and contralateral (t = 3.293, P = 0.002) thalamic MTR_asym values in the FE group were significantly higher than those in the HC group. Regarding thalamic subregion volumes, the volume of the right medial nucleus in the GE group was significantly smaller than that in the HC group (t = -2.667, P = 0.009); however, no significant differences were observed between the FE group and the HC group in any thalamic subregion (P > 0.05). Furthermore, no significant correlation was found between thalamic MTR_asym values and the volumes of thalamic subregions in either the GE or FE groups (P > 0.05). ConclusionsThis study employed GluCEST imaging in conjunction with structural MRI to investigate glutamate concentration and volumetric alterations in the thalamus among children with epilepsy and negative conventional MRI findings. Furthermore, the correlation between MTR_asym values and the volumes of specific thalamic subregions was examined. It is helpful to understand the neuro-metabolic differences of different types of epilepsy, and it opens up a new perspective for the diagnosis, treatment and prognosis of children with epilepsy. GluCEST technology is expected to provide a new, non-invasive imaging tool for exploring the pathophysiological mechanism of childhood epilepsy. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.002 ObjectiveTo investigate structural and functional brain alterations in type 2 diabetes mellitus-associated cognitive dysfunction (TDACD) and to explore potential neural mechanisms. Materials and MethodsThis study has been registered on the Prospective Register of Systematic Reviews (PROSPERO) with the registration number CRD420251115489. Whole-brain voxel-based morphometry (VBM) and resting-state functional magnetic resonance imaging (rs-fMRI) studies were systematically identified and selected according to PRISMA guidelines. A meta-analysis of gray matter volume (GMV) and resting-state functional connectivity (rs-FC) was conducted using seed-based d mapping with permutation of subject images (SDM-PSI), together with sensitivity, heterogeneity, and publication bias testing. Meta-regression analyses were additionally conducted using demographic and clinical variables [age, sex, body mass index (BMI), education, disease duration, glycated hemoglobin A1c (HbA1c), Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA)]. ResultsTwelve studies (7 functional and 5 structural) were included. Compared with healthy controls (HC), patients with TDACD demonstrated reduced functional activity in the left supramarginal gyrus [SMG, MNI coordinates: (-54, -54, 28), SDM-Z = -4.678, P < 0.000 5] and right inferior frontal gyrus, orbital part [IFGorb, MNI coordinates: (48, 40, -4), SDM-Z = -3.649, P < 0.001)], accompanied by significant gray matter volume reduction in the right putamen [PUT, MNI coordinates: (34, -4, 4), SDM-Z = -2.933, P < 0.01]. Sensitivity analyses confirmed the robustness of these findings. Higher levels of education were associated with milder functional impairment in the right IFGorb (r = 0.558, P < 0.005), whereas no other variables showed significant moderating effects. ConclusionsPatients with TDACD exhibit structural and functional abnormalities in the cortical-subcortical circuits; in particular, damage to core brain regions centered on the PUT-SMG-IFGorb may constitute a potential neuroimaging biomarker. These findings provide important clues for the pathological mechanisms, early identification, and neuromodulation interventions of TDACD. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.003 ObjectiveTo investigate the characteristics of glymphatic system dysfunction and choroid plexus (ChP) morphological remodeling in patients with neuropsychiatric systemic lupus erythematosus (NPSLE) using diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) and deep learning-based ChP segmentation, and to further explore their correlations with disease activity. Materials and MethodsThis prospective cross-sectional study included 47 patients with NPSLE and 59 concurrently recruited healthy controls between December 2023 and October 2025. All participants underwent conventional magnetic resonance imaging, high-resolution 3D T1-weighted imaging, and diffusion tensor imaging (DTI). The DTI-ALPS index was calculated to quantitatively evaluate glymphatic clearance. Furthermore, a 3D U-Net model was utilized to automatically segment the ChP of the lateral ventricles and quantify its volume. Partial correlation analyses, adjusting for age, years of education, and total intracranial volume, were performed to assess the associations between these neuroimaging metrics and the SLE Disease Activity Index (SLEDAI) score, Montreal Cognitive Assessment (MoCA) score, and disease duration. ResultsCompared to healthy controls, patients with NPSLE exhibited significantly reduced left DTI-ALPS indices (F = 6.235, P = 0.014), right DTI-ALPS indices (F = 10.168, P = 0.002) and global mean DTI-ALPS indices (F = 9.764, P = 0.002), and concomitantly increased left normalized ChP volume ratio (F = 9.532, P = 0.003), right normalized ChP volume ratio (F = 9.153, P = 0.003), and total normalized ChP volume ratio (F = 11.208, P = 0.001). Correlation analysis showed that, in the NPSLE group, the mean DTI-ALPS index was negatively correlated with the SLEDAI score (r = -0.436, P = 0.003), whereas there were no significant correlations between the mean DTI-ALPS index and the normalized ChP volume ratio, nor between these two parameters and disease duration or MoCA score (all P > 0.05). ConclusionsNPSLE patients exhibit glymphatic dysfunction and ChP morphological remodeling, and DTI-ALPS, which is closely correlated with SLEDAI, may serve as an imaging biomarker for monitoring central nervous system involvement and assessing disease fluctuation. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.004 ObjectiveTo compare the topological properties of brain networks during right-hand motor execution and motor imagery in healthy individuals, and to explore the neural mechanisms underlying the two motor conditions. Materials and MethodsForty right-handed healthy volunteers underwent task-based functional magnetic resonance imaging. Whole-brain functional connectivity matrices were constructed using the Schaefer 2018 atlas (17-network parcellation). The left somatomotor network and its subnetworks (A and B) were extracted. Graph-theoretical methods were used to calculate network density, global efficiency, nodal efficiency, clustering coefficient, and betweenness centrality at the module-internal and module-to-whole-brain levels. Paired t-tests were performed, with false discovery rate (FDR) correction applied for multiple comparisons. ResultsCompared with motor execution, motor imagery showed increased network density, global efficiency, and betweenness centrality among the module-to-whole-brain metrics in the left somatomotor network A (t = 4.272, 4.280, and 8.981; P = 0.004, 0.004, and < 0.001, respectively; FDR-corrected), and decreased nodal efficiency and clustering coefficient (t = -3.867 and -4.107; P = 0.009 and 0.005, respectively; FDR-corrected). In the left somatomotor network B, network density, global efficiency, and betweenness centrality among the module-to-whole-brain metrics were also increased during motor imagery (t = 3.403, 3.217, and 4.543; P = 0.022, 0.029, and 0.003, respectively; FDR-corrected). ConclusionsMotor imagery and motor execution showed distinct brain network topological patterns within the left somatomotor network, and the left somatomotor networks A and B responded differently to the two task conditions. This finding may provide imaging evidence for understanding the differences in neural mechanisms between motor imagery and motor execution. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.005 ObjectiveTo identify consistent alterations in brain functional activity in patients with acute leukemia (AL) compared with healthy controls (HCs) using functional magnetic resonance imaging (fMRI) and the activation likelihood estimation (ALE) meta-analytic approach, and to explore the potential neurobiological mechanisms underlying these changes. Materials and MethodsThis study was prospectively registered in PROSPERO (CRD42025117006). A systematic search was conducted for studies published before October 16, 2025, that used task-based or resting-state fMRI to investigate brain functional alterations in AL patients. Eligible studies were selected according to predefined inclusion and exclusion criteria. The ALE meta-analysis was performed using the GingerALE software with cluster-level family-wise error (cFWE) correction, and the results were visualized using Mango software. The reproducibility of the findings was further assessed using Jackknife sensitivity analysis and cluster-level reproducibility analysis. ResultsSeven studies were included, comprising 102 AL patients and 96 HCs. Among them, six studies involved acute lymphoblastic leukemia (ALL) and one study involved acute myeloid leukemia (AML). Compared with HCs, AL patients exhibited significantly decreased activation in the left postcentral gyrus, left inferior parietal lobule, left middle frontal gyrus, left lingual gyrus, and right cerebellar culmen (P < 0.05, cluster volume = 1088 - 1672 mm³). These clusters were reproduced in 4, 4, 4, 3, and 3 of 5 Jackknife analyses, respectively. Conversely, increased activation was observed in the right cerebellar tonsil (P < 0.05, cluster volume = 1760 mm³), which remained significant across all 5 Jackknife iterations. ConclusionsThe ALE meta-analysis revealed abnormal functional activity in key brain regions, including the somatosensory cortex, frontal lobe, occipital lobe, and cerebellum, in patients with AL. These findings provide neuroimaging evidence for AL-related brain functional alterations and may facilitate a better understanding of AL-related brain dysfunction. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.006 ObjectiveTo explore the value of histogram parameters derived from diffusion tensor imaging (DTI) and arterial spin labeling (ASL) in the non-invasive preoperative prediction of the promoter methylation status of O⁶-methylguanine-DNA methyltransferase (MGMT), and to further analyze their correlation with patients^, survival and prognosis. Materials and MethodsWe retrospectively analyzed DTI and ASL data from 349 patients with IDH-wildtype glioblastoma. Histogram parameters—including fractional anisotropy (FA), mean diffusivity (MD), and cerebral blood flow (CBF)—were extracted. Binary logistic regression models were built using DTI parameters, ASL parameters, and their combination. Patients were divided into two groups based on MGMT promoter methylation status (methylated vs. unmethylated), and comparisons were made between groups. Model performance was assessed using receiver operating characteristic (ROC) curves, calibration curves, and the Hosmer-Lemeshow goodness-of-fit test. For survival analysis, we performed Kaplan-Meier analysis and used both univariate and multivariate Cox regression to examine associations between imaging features, molecular markers, and overall survival (OS). Bootstrap mediation analysis was conducted to test whether specific imaging parameters mediate the relationship between MGMT methylation status and survival. ResultsThe DTI model (AUC = 0.842) significantly outperformed the ASL model (AUC = 0.619) in predicting MGMT methylation status. The combined DTI-ASL model (AUC = 0.859) was not significantly different from the DTI model alone (Z = 0.53, P = 0.593). Kaplan-Meier analysis confirmed that MGMT promoter methylation was a significant protective factor for prognosis (Log-rank P = 0.041, HR = 0.737, 95% CI: 0.550 to 0.989). However, Cox regression showed that although imaging parameters were significantly associated with MGMT methylation status, they were not independent predictors of OS. Mediation analysis revealed that the 10th percentile of FA mediated the relationship between MGMT methylation status and OS. Skewness of MD showed a statistical association with both factors but its mediation effect was not significant. ConclusionsDTI histogram parameters serve as effective noninvasive imaging markers for predicting MGMT promoter methylation before surgery. The 10th percentile of FA is a key imaging factor linking unmethylated MGMT status to poor prognosis. MD skewness is statistically associated with this link, suggesting that disruption of tumor microstructure and increased tissue heterogeneity may mediate the effect of MGMT methylation status on patient outcomes. These findings offer a new imaging-based perspective for understanding the aggressive biological behavior of glioblastoma. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.007 ObjectiveThis study aimed to assess the left ventricular hemodynamic characteristics in patients with myocardial infarction (MI) and left ventricular thrombus (LVT) using four-dimensional flow magnetic resonance imaging (4D Flow MRI), and to explore their potential relationship with LVT formation. Materials and MethodsA total of 45 subjects were enrolled, including 15 subjects in each group: MI with LVT [MI-LVT(+)], MI without LVT [MI-LVT(-)], and healthy controls. All participants underwent steady-state free precession and 4D Flow MRI scans. Left ventricular functional parameters, four flow components (direct flow, retained inflow, delayed ejection flow, and residual volume), and kinetic energy (KE) parameters, including global KE, peak systolic KE, systolic KE, diastolic KE, E-wave and A-wave peak KE, were quantitatively analyzed using MASS software. All KE parameters were normalized to left ventricular end-diastolic volume (LVEDV) (denoted as KEi_EDV). Myocardial scar extent was assessed using CVI.42 software. Binary logistic regression analysis was used to identify variables independently associated with the presence of LVT, and receiver operating characteristic (ROC) curves were employed to evaluate the diagnostic performance of 4D Flow MRI parameters. ResultsCompared with healthy controls, MI patients exhibited significant left ventricular remodeling, with increased left ventricular end-diastolic volume index (LVEDVi) and left ventricular end-systolic volume index (LVESVi) and decreased left ventricular ejection fraction (LVEF) (all P < 0.001). LVEF was lower in the MI-LVT(+) group compared with the MI-LVT(-) group (31% vs. 40%, P = 0.034). MI patients showed significantly reduced direct flow (P < 0.001) and increased residual volume (P < 0.001), with more pronounced changes observed in the MI-LVT(+) group (P < 0.05). All KEi_EDV parameters were significantly reduced in the MI group (all P < 0.001); however, no statistically significant differences were found between the LVT subgroups. Logistic regression analysis showed that direct flow was independently associated with the presence of LVT. ROC curve analysis demonstrated an area under the curve of 0.747, with an optimal cutoff value of 16.24%, a sensitivity of 80.00%, and a specificity of 66.67%. ConclusionsMI patients demonstrated notable LV remodeling and hemodynamic abnormalities compared to healthy controls. Reduced direct flow and increased residual flow were key characteristics, especially in patients with LVT. Overall KE parameters were decreased, especially in the MI-LVT(+) group. Direct flow is independently associated with the presence of LVT and can serve as a supplementary tool to conventional structural assessment, assisting clinicians in multi-dimensional risk stratification. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.008 ObjectiveTo systematically evaluate whether acute Takotsubo cardiomyopathy is accompanied by both myocardial tissue-level inflammatory changes and systemic inflammatory activation, and to investigate their correspondence and temporal evolution using cardiac magnetic resonance (CMR) tissue characterization combined with serum cytokine profiling. Materials and MethodsForty-two patients with acute Takotsubo cardiomyopathy and 29 healthy controls were enrolled. All participants underwent CMR with acquisition of native T1 mapping, native T2 mapping, and extracellular volume (ECV) for myocardial tissue characterization, along with left ventricular functional assessment. Patients received follow-up CMR at 3 and 6 months. Concurrent measurements included cardiac injury biomarkers, NT-proBNP, and multiple cytokines (IL-1β, IL-6, IL-8, IL-10, MCP-1, TNF-α, IFN-γ). Acute-phase and follow-up changes in tissue parameters and inflammatory markers were compared, and correlations between acute values and longitudinal changes were analyzed. ResultsPatients in the acute phase exhibited significantly elevated native T1 and T2 (both P < 0.05), indicating diffuse myocardial edema, with the largest increases observed in ballooning segments. Both T1 and T2 declined significantly during recovery (both P < 0.05), although ECV remained elevated at 6 months in a subset of patients. Left ventricular ejection fraction improved over time, accompanied by reductions in end-systolic volume index and myocardial mass index (both P < 0.01). IL-6, IL-8, and MCP-1 were elevated in the acute phase, and their declines showed moderate positive correlations with improvements in T1/T2 (all P < 0.05), indicating parallel changes between myocardial abnormalities and systemic inflammation with temporal correspondence during recovery. TNF-α, IFN-γ, and IL-10 did not rise acutely and remained stable throughout follow-up, suggesting a non-necrotic and non-severe inflammatory phenotype in Takotsubo cardiomyopathy. ConclusionsAcute Takotsubo cardiomyopathy is characterized by marked myocardial edema and elevations in proinflammatory cytokines. Although most imaging and inflammatory abnormalities improve during the recovery phase, residual structural alterations persist in some patients, indicating that inflammation-related myocardial injury may not be fully reversible. Accordingly, CMR combined with cytokine profiling provides valuable tools for assessing myocardial edema, changes in quantitative tissue parameters, and inflammatory responses, as well as for dynamically tracking the recovery process in Takotsubo cardiomyopathy. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.009 ObjectiveTo evaluate the clinical value of magnetic resonance imaging (MRI)-based radiomics and deep learning (DL) features in predicting axillary lymph node (ALN) status in breast cancer patients with 1 to 2 positive sentinel lymph nodes (SLNs). Materials and MethodsWe retrospectively analyzed clinicopathological and MRI data from breast cancer patients with pathologically confirmed 1 to 2 positive SLNs at our institution between January 2017 and December 2024. The tumor regions of interest (ROIs) were manually delineated on diffusion-weighted imaging (DWI) and third-phase dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) images using ITK-SNAP 3.8.0 software. Radiomics features were extracted using PyRadiomics, and DL features were extracted using a pretrained Inception-v3 model. Features strongly associated with ALN metastasis were selected using statistical tests, minimum redundancy maximum relevance (mRMR), Spearman correlation analysis, and the least absolute shrinkage and selection operator (LASSO) algorithm. Clinicopathological risk factors associated with ALN metastasis were identified using multivariate logistic regression analysis. Radiomics model, DL model, deep learning radiomics (DLR) model, clinical model and combined model incorporating clinical features were constructed to predict ALN status. The area under the curve (AUC), sensitivity, specificity, and accuracy were calculated to evaluate model performance. Decision curve analysis was performed to assess clinical utility. ResultsA total of 256 patients were enrolled, with 105 patients in the ALN-positive group and 151 in the ALN-negative group. Logistic regression analysis demonstrated that maximum tumor diameter, lymphovascular invasion (LVI), and the number of positive SLNs were independent risk factors for ALN metastasis (all P < 0.05). In the validation cohort, the DLR model achieved an AUC of 0.800 (95% CI: 0.661 to 0.909), outperforming the radiomics model (AUC = 0.734, 95% CI: 0.580 to 0.861) and the DL model (AUC = 0.774, 95% CI: 0.632 to 0.903). The combined model incorporating clinical features showed optimal performance, achieving an AUC, sensitivity, specificity, and accuracy of 0.851 (95% CI: 0.722 to 0.953), 76.2%, 90.3%, and 84.6%, respectively. It also achieved a higher clinical net benefit within a reasonable threshold probability range. ConclusionsMultiparametric MRI-based radiomics, DL, and DLR models can assist in assessing the risk of ALN metastasis in breast cancer patients with 1 to 2 positive SLNs preoperatively. The combined model incorporating clinical features demonstrated favorable clinical utility for comprehensive risk stratification of ALN metastasis after sentinel lymph node biopsy (SLNB). Following further multicenter prospective validation, it may provide auxiliary support for individualized clinical decision-making regarding axillary management. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.010 ObjectiveTo quantitatively explore changes in the fat fraction (FF) of the supraclavicular fat depot in patients with colorectal cancer using the MRI multi-echo DIXON (mDIXON) technique. Materials and MethodsNewly diagnosed colorectal cancer patients (n = 60) were prospectively recruited and underwent neck MRI mDIXON. Neck mDIXON images from healthy volunteers (n = 60) during the same period were collected as controls. Representative region of interest for brown adipose tissue (BAT) in the supraclavicular fat depot and region of interest for white adipose tissue (WAT) in the dorsal subcutaneous fat at the same level were manually delineated on the mDIXON-FF images of both groups, avoiding blood vessels, nerves, and edge areas. Histogram features were extracted. Two independent samples t-tests (for normal parameters) or Mann-Whitney U tests (for non-normal parameters) were used to compare the FF histogram features between the two groups. Colorectal cancer patients were further subdivided based on pathological or clinical characteristics (such as tumor staging, whether there is lymph node metastasis, etc.). Propensity score matching was used to control for confounding factors like age, gender, and body mass index, and the FF histogram features were compared between the matched subgroups. In addition, 10 healthy volunteers were recruited. MRI mDIXON images of their necks were collected before and after a 30-day interval to evaluate the measurement stability of BAT in the supraclavicular fossa and subcutaneous WAT. The manual delineation of the ROI was also tested for its repeatability. Finally, the consistency of delineating BAT ROIs on the left and right sides of the supraclavicular fat depot was verified. ResultsHistogram analysis of the supraclavicular fat depot revealed that three features representing FF magnitude [10th Percentile (85.69 vs. 82.59, P < 0.001), Mean (90.24 vs. 86.92, P < 0.001), and Minimum (83.22 vs. 79.71, P < 0.001)] were all significantly higher in the colorectal cancer group compared to the control group. None of the histogram features for subcutaneous white adipose tissue showed statistically significant differences between the two groups. After internal grouping of colorectal cancer patients based on characteristics like lymph node metastasis, significant differences were observed between the lymph node metastasis-negative and positive groups for three features: InterquartileRange (5.27 vs. 3.55, P = 0.048), MeanAbsoluteDeviation (2.83 vs. 1.99, P = 0.029), and RobustMeanAbsoluteDeviation (2.20 vs. 1.45, P = 0.013). No statistically significant differences were observed in other features between the other subgroups. The reproducibility test of manual ROI delineation showed that all the above six features had good consistency [intra-class correlation coefficient (ICC) ≥ 0.75]. The self-control before and after showed that the FF histogram features of supraclavicular BAT and subcutaneous WAT obtained by scanning after 30 days had good stability. There was no significant statistical difference in delineating BAT ROI from the left and right supraclavicular fat depots respectively. ConclusionsIn patients with colorectal cancer, the relative content of supraclavicular brown adipose tissue (BAT) is decreased and the FF is elevated, suggesting that the reduction in BAT proportion may be accompanied by a decrease in BAT content, and its changes are correlated with disease progression indicators such as lymph node metastasis. Furthermore, the histogram analysis method based on mDIXON-FF provides a non-invasive and stable new approach for assessing brown adipose tissue, showing future potential as an imaging biomarker for metabolic evaluation and prognosis prediction in colorectal cancer. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.011 ObjectiveTo investigate the correlation between multidimensional quantitative parameters derived from the magnetic resonance image compilation (MAGiC) sequence and the iterative decomposition of water and fat with echo asymmetry and least‑squares estimation quantitative (IDEAL‑IQ) sequence with lumbar disc degeneration (LDD), and to construct an interpretable machine learning model for predicting LDD. Materials and MethodsA total of 199 volunteers with chronic low back pain were prospectively recruited and divided into a normal group (n = 83) and a degeneration group (n = 116) based on Pfirrmann grading (> grade Ⅱ defined as LDD) on T2‑weighted images. The cohort was randomly split into training and validation sets at a 7∶3 ratio. All participants underwent conventional lumbar MRI, MAGiC, and IDEAL‑IQ sequence scans. Collected data included the vertebral bone quality (VBQ) score calculated from conventional T1‑weighted images, T1 relaxation time (T1), T2 relaxation time (T2), and proton density (PD) values obtained from the MAGiC sequence, the average fat fraction (FFav) measured by the IDEAL‑IQ sequence, as well as general clinical information, laboratory results, and body mass index (BMI). Spearman correlation analysis and logistic regression were used to screen for degeneration‑related predictors. Six machine learning algorithms [logistic regression (LR), decision tree (DT), extreme gradient boosting (XGBoost), support vector machine (SVM), k‑nearest neighbors (KNN), and light gradient boosting machine (LightGBM)] were employed to build prediction models. Model performance was comprehensively evaluated using the receiver operating characteristic (ROC) curve, area under the curve (AUC), accuracy, sensitivity, specificity, F1‑score, positive predictive value (PPV), and negative predictive value (NPV), DeLong test was used to compare the differences in AUC among the models. Clinical net benefit was assessed by decision curve analysis (DCA). The optimal model was interpreted using SHapley Additive exPlanations (SHAP). ResultsSignificant differences were observed between the normal and degeneration groups in age, albumin (ALB), T1, T2, PD, FFav, and VBQ scores (P < 0.05). Disc degeneration showed positive correlations with age, FFav, and VBQ score (r = 0.61, 0.36, 0.39), and negative correlations with ALB, T1, T2, and PD values (r = -0.30, -0.60, -0.70, -0.47). Among the six machine learning models, LightGBM performed best, achieving an AUC of 0.967 (95% CI: 0.942 to 0.993) in the training set, with accuracy of 91.2%, sensitivity of 90.0%, specificity of 93.3%, F1‑score of 0.922, PPV of 94.7%, and NPV of 87.5%. DCA indicated a high clinical net benefit within the medium‑risk threshold range. SHAP analysis further revealed that T2, T1, and age were the key predictors of LDD. ConclusionsMachine learning models built on multidimensional quantitative parameters from MAGiC and IDEAL‑IQ sequences can effectively predict LDD, with the LightGBM model demonstrating the best performance. The SHAP method enhances model interpretability, providing a quantitative tool and decision‑making reference for early identification and clinical intervention of LDD. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.012 ObjectiveTo investigate the value of combined T1‑weighted and T2‑weighted magnetic resonance black‑blood thrombus imaging (BTI) for staging diagnosis of lower extremity deep vein thrombosis (DVT). Materials and MethodsA total of 55 patients with first-diagnosed DVT underwent T1-BTI and T2-BTI scanning of the lower extremity deep veins and were divided into acute (n = 29), subacute (n = 13), and chronic (n = 13) stages based on comprehensive clinical diagnosis. On T1-BTI and T2-BTI images, the signal intensity ratio (SIR) at the proximal and distal ends of the thrombus (proximal T1-SIR, proximal T2-SIR, distal T1-SIR, distal T2-SIR, denoted as pT1-SIR, pT2-SIR, dT1-SIR, dT2-SIR, respectively), lumen area ratio, and soft tissue edema score (0 to 3) were measured. Differences in each indicator between acute and subacute stages, as well as between acute and chronic stages, and the diagnostic performance of individual and combined indicators in the above differential scenarios were compared. ResultsIn differentiating acute from subacute DVT, acute-stage pT1-SIR, pT2-SIR, dT1-SIR, and dT2-SIR were all lower than those in subacute stage (all P < 0.05); soft tissue edema score was higher in acute stage (P = 0.037); there was no significant difference in lumen area ratio between groups (P = 0.591). Among single indicators, pT1-SIR had the highest area under the curve (AUC) of 0.944 [95% confidence interval (CI): 0.871 to 1.000], with a sensitivity of 89.7% and specificity of 92.3% for diagnosing acute stage using a cutoff value of ≤1.23. The AUC of pT1-SIR combined with soft tissue edema score was 0.968 (95% CI: 0.911 to 1.000), with a sensitivity of 93.1% and specificity of 100%; the AUC of pT1-SIR combined with dT2-SIR was 0.963 (95% CI: 0.912 to 1.000), with a sensitivity of 86.2% and specificity of 100%. There were no statistically significant differences in AUC between either combined model and single pT1-SIR (both P > 0.05). In differentiating acute from chronic DVT, there were no statistically significant differences in SIR indicators between groups (P > 0.05). Lumen area ratio and soft tissue edema score in acute DVT were significantly higher than those in chronic DVT (P < 0.001). The AUC of soft tissue edema score was 0.996 (95% CI: 0.984 to 1.000), with a sensitivity of 96.6% and specificity of 100% for diagnosing acute stage using a cutoff value of ≥ 1.5; the AUC of lumen area ratio was 0.862 (95% CI: 0.723 to 1.000). ConclusionsCombined T1-BTI and T2-BTI can reflect the course of DVT from the two dimensions of thrombus signal and soft tissue edema. Soft tissue edema score has high diagnostic efficacy in ruling out chronic DVT; pT1-SIR is helpful in differentiating acute from subacute DVT, and combined assessment shows good diagnostic performance, but the incremental value compared with single pT1-SIR needs further validation in large-sample studies. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.013 ObjectiveTo establish a brain tuberculous granuloma model in Sprague-Dawley (SD) rats through BCG vaccination, and investigate the changes in the extracellular space (ECS) of brain tuberculous granulomas and the tracer distribution and clearance. Materials and MethodsForty SD rats were used. Thirty rats were randomly selected to establish a brain tuberculosis granuloma model by stereotactic injection of BCG suspension into the caudate nucleus of the rats, while the remaining ten received saline injections at the same site to serve as the control group. Four weeks after modeling, MRI scans were performed to observe granuloma formation. Rats in the tuberculosis granuloma model group and the saline group underwent tracer-based MRI scans with gadolinium-diethylene triamine pentaacetic acid (Gd-DTPA) as the contrast agent. Ten rats in the tuberculosis granuloma group meeting the inclusion criteria (intracranial tuberculosis granulomas confirmed by pathology) and eight rats in the saline group were used for parameter measurement on tracer-based imaging (effective diffusion coefficient D^*, volume fraction α, clearance rate constant K, and half-life T_1/2). The ECS parameters of the two groups were compared to analyze changes in the ECS and its mechanism. ResultsA brain tuberculosis granuloma model was established using 1×10⁶ CFU/mL BCG suspension, and four weeks after modeling, the formation rate of tuberculosis granulomas in rats was 56.7% (17/30). Experimental results showed that compared with the saline group, the brain tuberculosis granuloma model group exhibited significant changes in multiple ECS parameters: the clearance rate (K) in the granuloma model group was lower than that in the saline group (K: P = 0.006), while the diffusion coefficient (D^*), half-life (T_1/2), and volume fraction (α) were all higher than those in the saline group (D^*: P = 0.024;T_1/2:P = 0.011;α: P = 0.029). ConclusionsA brain tuberculosis granuloma model in SD rats can be established using a BCG bacterial suspension at a dose of 1×10⁶ CFU/mL. Due to cell necrosis and matrix remodeling, the ECS structure in the granuloma area is altered, allowing tracers to diffuse faster and over a wider range. However, the structural barrier of the granuloma wall obstructs local drainage, leading to prolonged tracer retention. MRI tracer techniques can effectively monitor the morphological and functional changes of the ECS under pathological conditions, providing a theoretical basis for drug delivery via the ECS. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.014 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.015 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.016 ^,s disease (PD) has insidious early manifestations, and conventional magnetic resonance imaging (MRI) is limited in detecting brain microstructural abnormalities. Therefore, sensitive and noninvasive imaging methods are urgently needed for early identification and disease assessment. Diffusion kurtosis imaging (DKI) can quantify the non-Gaussian diffusion characteristics of water molecules and sensitively reflect the complexity and heterogeneity of brain tissue microstructure, showing important value in recent studies of PD-related brain microstructural changes. However, current findings remain inconsistent because of differences in sample size, disease stage, clinical subtype, scanning protocols, and post-processing methods, highlighting the need for a systematic review. This article first outlines the technical principles and major parameters of DKI, and then focuses on its applications in gray matter nuclei, white matter fiber tracts, neural circuits, and the glymphatic system in PD. The potential value of DKI combined with artificial intelligence (AI) in auxiliary diagnosis and clinical phenotyping is also discussed. Finally, this review analyzes the current research limitations and future directions, aiming to provide references for the in-depth research and clinical application of DKI in PD. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.017 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.018 ^,s disease (AD), is a critical period where early identification is paramount for effective prevention and intervention. However, the current lack of objective and specific diagnostic markers poses a significant challenge to early identification. In recent years, neuroimaging studies have suggested that abnormalities in the negative functional connectivity between the default mode network (DMN) and the dorsal attention network (DAN) may serve as a potential early neural marker. However, existing research findings are inconsistent, and no systematic synthesis of these findings has been conducted. Therefore, there is an urgent need for a review to integrate current progress and clarify future directions. This review focuses on the physiological basis of DMN-DAN negative functional connectivity and its alterations in individuals with SCD, synthesizes existing evidence regarding its association with AD pathological biomarkers and cognitive function, and summarizes its research value in differentiating etiologies (e.g., vascular or depression-related cognitive decline) and predicting disease progression. Additionally, it highlights methodological challenges in current research, such as participant heterogeneity and the lack of unified analytical standards, and looks ahead to future research directions including the integration of multimodal imaging and artificial intelligence. This review aims to summarize the current state and controversies in this field, providing a theoretical foundation for the development of early biomarkers based on brain network connectivity features. Ultimately, it seeks to guide clinical practice and enhance early screening, precise subtyping, and intervention efficacy for SCD. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.019 ^,s disease is a common neurodegenerative disorder characterized by the loss of dopaminergic neurons in the substantia nigra and abnormal aggregation of α-synuclein. Deep brain nuclei, such as the subthalamic nucleus, globus pallidus, and substantia nigra, play critical roles in motor control, and their structural and functional abnormalities are key contributors to the symptoms of Parkinson^,s disease. In recent years, the rapid advancement of MRI techniques, especially high-field MRI, diffusion tensor imaging, and resting-state functional MRI, has provided powerful tools for in vivo, non-invasive investigation of the microstructure, white matter connectivity, and functional network activity of deep brain nuclei in Parkinson^,s disease patients. This review summarizes recent progress in structural and functional MRI studies of deep brain nuclei in Parkinson^,s disease, focusing on findings related to volume, morphology, iron deposition, white matter integrity, and abnormalities in local brain activity and network connectivity. It also discusses the correlations between these imaging alterations and disease stages, clinical symptoms (such as bradykinesia, tremor, and gait disturbances), and non-motor symptoms of Parkinson^,s disease. The review also analyzes current technical limitations in MRI studies of deep brain nuclei and, in light of emerging trends in neuroimaging, identifies future research directions, thereby providing imaging-based theoretical foundations and research references for exploring the pathogenesis of Parkinson^,s disease, enabling precise diagnosis and treatment, and guiding targeted interventions. ]]> Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.020 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.021 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.022 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.023 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.024 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.025 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.026 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.027 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.028 Mon,20 Jul 2026 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2026.07.029 Mon,20 Jul 2026 00:00:00 GMT