http://med-sci.cn/cgzcx/en/contents_list.asp?issue=202512 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.1674-8034.2025.12.001 ObjectiveTo investigate the abnormal patterns of brain structure, function, and network in parents who have lost their only child (shidu parents), and to reveal the overall coupling relationship between these abnormalities and clinical symptoms such as depression, anxiety, and post-traumatic stress disorder (PTSD). Materials and MethodsA total of 47 shidu parents (SD) and 36 healthy controls (HC) were included. Multimodal magnetic resonance imaging (MRI) data and clinical symptom ratings were collected from all participants. Voxel-based morphometry (VBM), resting-state functional analysis, and graph-theoretical network analysis were conducted. Canonical correlation analysis (CCA) was further applied to examine the global relationship between multimodal neuroimaging indicators and clinical symptom sets. ResultsCompared with the HC group, the SD group showed reduced gray matter volume in brain regions including the left hippocampus, bilateral insula, and left medial and paracingulate gyri (P < 0.05, FDR corrected), along with functional abnormalities such as decreased degree centrality in the left temporal lobe (P < 0.05, FDR corrected) and increased regional homogeneity in the right middle temporal gyrus (P < 0.05, FDR corrected). The topological properties of the brain network were also altered in the SD group (P < 0.05, FDR corrected). CCA results demonstrated a significant overall coupling between neuroimaging features and clinical symptoms (canonical correlation coefficient of the first pair: r = 0.890, P < 0.001). Among the imaging indicators, network properties contributed the most (loading r = -0.998) and were most strongly associated with PTSD symptoms (cross-loading r = -0.882). ConclusionsShidu parents exhibit widespread brain alterations, with abnormal brain network topology serving as a core neuroimaging feature strongly linked to PTSD and other clinical symptoms. These findings suggest that brain network reorganization may represent a key neural substrate underlying psychological symptoms in this population. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.002 ObjectiveBased on resting-state functional magnetic resonance imaging (rs-fMRI) data, the cross attention mechanism (CAM) combined with static-dynamic graph convolutional network (GCN) technology was utilized to evaluate the classification efficacy of this method for patients with Parkinson^,s disease (PD), and to explore potential imaging biomarkers, providing a new perspective for the clinical diagnosis and pathological mechanism analysis of PD. Materials and MethodsA total of 32 patients with PD were prospectively recruited from the outpatient department of the Affiliated Hospital of Qingdao University, and 30 healthy controls (HC), matched for gender, age and education years were recruited from the community health management center of the Affiliated Hospital of Qingdao University. Resting-state functional magnetic resonance imaging was collected from both groups of subjects. After image preprocessing, static-graph convolutional networks (static-GCN) and dynamic-graph convolutional networks (dynamic-GCN) were constructed for each subject based on the AAL atlas and GCN. Through multi-scale feature extraction and CAM, the complementary information of static-GCN and dynamic-GCN was fused. The performance was evaluated using the accuracy of five-fold cross-validation and the area under the receiver operating characteristic (ROC) curve (AUC). The attention weight coefficients obtained during the process were combined with statistical analysis to identify the abnormal brain regions and static-dynamic functional connections (static-dynamic FC) most related to PD. Two independent sample t-tests were used for inter-group comparisons, and Pearson correlation analysis was used to explore the correlation between the statistically significant static-dynamic FC and clinical scales. ResultsThe method based on CAM combined with static-dynamic graph convolution network has excellent classification performance (with an accuracy of 79.84%, a sensitivity of 80.47%, and a specificity of 78.47%). The ROC curve analysis results show that the AUC for diagnosing PD is 0.814 (95% CI: 0.727 to 0.902, P < 0.001). Five PD high-weight brain regions were identified: the right supplementary motor area, the left posterior cingulate gyrus, the left postcentral gyrus, cerebellar Lobule Ⅵ, and vermis 10. At the same time, two most relevant static-dynamic FC were discovered. Compared with the HC group, the static-dynamic FC in the following two pairs of brain regions was significantly enhanced in the PD group (P < 0.05): (1) the left posterior cingulate gyrus - cerebellar Lobule Ⅵ; (2) the right supplementary motor area - vermis 10 - cerebellar Lobule Ⅵ/the left postcentral gyrus. Moreover, both of these enhanced static-dynamic FC were significantly positively correlated with the UPDRS-Ⅲ score (r = 0.432, P = 0.017; r = 0.420, P = 0.021). ConclusionsThe method combining CAM with static-dynamic graph convolution networks has excellent diagnostic performance, and has discovered abnormal enhanced patterns of specific static-dynamic FC between the cerebellum and the cerebral cortex in patients with PD, providing a new basis for the objective imaging diagnosis of PD. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.003 ObjectiveTo investigate changes in default mode network (DMN) functional connectivity (FC) in adolescents with bipolar disorder (BD) after short-term treatment and their correlation with clinical symptoms. Materials and MethodsThirty patients with BD were enrolled and underwent follow-up assessments after two weeks of naturalistic treatment, along with 33 age- and sex-matched healthy control (HC). Clinical assessments included: the 17-item Hamilton Depression Rating Scale (HAMD-17) for depressive symptoms, the Young Mania Rating Scale (YMRS) for manic symptoms, and the Hamilton Anxiety Rating Scale (HAMA) for anxiety symptoms. Resting-state functional magnetic resonance imaging (rs-fMRI) scans were acquired for all participants at both baseline and the 2-week follow-up timepoint. Using the DMN as seed regions, we performed whole-brain FC analysis. Three primary analyses were conducted: (1) Within-patient longitudinal comparison to identify FC changes between pre- and post-treatment; (2) Between-group cross-sectional comparisons (baseline BD vs. HC; post-treatment BD vs. HC) using two-sample t-tests; (3) Correlation analysis between significant treatment-related FC changes and clinical scale scores. ResultsThe study found that after two weeks of treatment, the functional connectivity of the posterior cingulate cortex (PCC) in patients changed, showing weakened connectivity with the right calcarine cortex and enhanced connectivity between the posterior cingulate cortex and the left middle occipital gyrus (t = 5.79, P = 0.001; t = 4.72, P = 0.004). However, no significant correlations were observed between these changes and clinical scale scores (r = 0.183, P = 0.466; r = 0.238, P = 0.342; r = 0.086, P = 0.684; r = -0.121, P = 0.631; r = -0.031, P = 0.902; r = -0.213, P = 0.308). Compared to the healthy control group, baseline BD patients exhibited abnormalities in the visual network (VN) and the salience network (SN). After treatment, these abnormalities in the VN and SN were not fully resolved, and slight abnormalities emerged in the frontoparietal network (FPN). ConclusionsShort-term treatment can partially modulate the abnormal functional connectivity between the DMN, VN, and SN in BD patients. These neuroimaging findings provide new directions for understanding the neurobiological mechanisms of BD and optimizing early intervention strategies. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.004 ObjectiveTo construct a three-classification model based on T1-weighted imaging (T1WI) radiomics for the diagnosis and early, middle and late stage classification of Parkinson^,s disease (PD), and to explore the diagnostic value of different nuclear features. Materials and MethodsA prospective analysis was conducted on the T1WI data of 146 patients, including 86 cases from the Second Affiliated Hospital of Xinjiang Medical University (development cohort), with 26 cases in the normal group, 35 cases in the early PD group, and 25 cases in the middle and late PD group; and 60 cases from Affiliated First Hospital of Xinjiang Medical University (external validation cohort), with 18 cases in the normal group, 22 cases in the early PD group, and 20 cases in the middle and late PD group. Six nuclei, namely the caudate nucleus (CN), putamen (PUT), globus pallidus (GP), red nucleus (RN), substantia nigra (SN), and nucleus accumbens (NAC), were segmented from all data, and 1688 radiomics features (including first-order statistics, shape, texture, and filter features) were extracted. Logistic regression (LR) algorithm was used to construct 6 single-nucleus models and 1 combined model. Key features were selected through variance threshold method, univariate selection method, and least absolute shrinkage and selection operator (LASSO) algorithm. Model performance and interpretability were analyzed using receiver operator characteristic (ROC) curve, confusion matrix, and SHapley Additive exPlanations (SHAP) value. ResultsThe macro area under the curve (AUC) of the combined model in the training set, internal validation set, and external validation set were 0.93 (95% CI: 0.87 to 1.00), 0.88 (95% CI: 0.69 to 1.00), and 0.84 (95% CI: 0.75 to 0.92), respectively, which were significantly better than most single-nucleus models. Among the 21 key features selected, 9 were related to GP (e.g., wavelet-HLL_glcm_Correlation_GP), and the absolute value of the coefficient of the lbp-3D-k_firstorder_Minimum_RN feature of RN was the largest (0.12). SHAP analysis showed that the classification of the normal group relied on the texture symmetry of GP and SN, early PD focused on the local structural changes of PUT and GP, and middle and late PD was characterized by signal abnormalities of RN and NAc. ConclusionsThe multi-nucleus combined model based on T1WI has high performance in the diagnosis and staging of PD. Nucleus-specific features can reflect the spatial heterogeneity of the pathological process of PD, with GP and SN playing a core role in diagnosis, providing an imaging basis for precise clinical staging. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.005 ObjectiveTo investigate the neuroimaging mechanisms underlying the association between structural alterations of the glymphatic system and cognitive and emotional impairments in patients with sudden sensorineural hearing loss (SSNHL), using structural magnetic resonance imaging. Materials and MethodsForty-eight patients with bilateral SSNHL and 47 healthy controls were enrolled. All participants underwent hearing assessments, multidimensional neuropsychological evaluations, and 3D T1-weighted magnetic resonance imaging scans. Choroid plexus volume, perivascular space volume, and ventricular volumes (bilateral, third, and fourth ventricles) were automatically segmented and quantified using a U-net deep learning model and FreeSurfer tool. Spearman correlation analyses were performed to examine the relationships between imaging-derived metrics and cognitive and emotional scores. ResultsPatients and healthy controls were well matched in age, sex, and education (P > 0.05). Patients showed higher mean binaural hearing thresholds compared to controls, as well as differences in cognitive and depression scale scores (P < 0.05). Compared with controls, bilateral SSNHL patients exhibited larger raw and normalized choroid plexus volumes, perivascular space volumes, and bilateral lateral ventricular volumes (P < 0.05), while no significant group differences were observed in the raw or normalized volumes of the third and fourth ventricles (P > 0.05). Normalized choroid plexus volume was negatively correlated with Symbol Digit Modalities Test performance (r = -0.311, P = 0.032) and positively correlated with mean binaural hearing threshold (r = 0.382, P = 0.007). ConclusionsBilateral SSNHL is associated with structural alterations in glymphatic system-related brain regions. Enlargement of the choroid plexus is significantly related to reduced information processing speed, providing insight into the neuroimaging mechanisms underlying cognitive and emotional impairments in SSNHL. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.006 ObjectiveUsing synthetic MRI (SyMRI) technology to analyze brain microstructural changes in children with global developmental delay and evaluate the diagnostic efficacy of T1 and T2 relaxation values and proton density (PD) values in SyMRI. Materials and MethodsThe study was conducted from May 2024 to October 2024, involving 55 children with global developmental delay (GDD group) and 30 typically developing children (TD group) from the Third Affiliated Hospital of Zhengzhou University. SyMRI sequences and clinical data were collected. Post-processing was performed to derive SyMRI parameters, and T1 and T2 relaxation values and PD values were measured in eight gray matter regions of interest (ROIs) for all children. Correlation analyses were performed between the T1, T2, and PD values of brain regions with significant differences and clinical scales. Receiver operating characteristic (ROC) curve analyses were used to evaluate the diagnostic efficacy SyMRI parameters. ResultsCompared to the TD group, the GDD group showed increased T1 relaxation values in the left parietal gray matter and right occipital gray matter, T2 relaxation values in the left parietal gray matter and bilateral temporal gray matter, as well as PD values in the left temporal gray matter and right occipital gray matter. These differences were statistically significant (P < 0.003). Correlation analyses revealed that the T1 relaxation values in the left parietal gray matter were positively correlated with adaptive behavior developmental quotient (r = 0.327, P = 0.015) and personal-social developmental quotient (r = 0.535, P < 0.001). The T2 relaxation values in the left parietal gray matter were positively correlated with adaptive behavior developmental quotient (r = 0.449, P = 0.001). The T2 relaxation values in the left temporal gray matter were positively correlated with adaptive behavior developmental quotient (r = 0.348, P = 0.009) and personal-social developmental quotient (r = 0.321, P = 0.017). The PD values in the left temporal gray matter were positively correlated with fine motor developmental quotient (r = 0.322, P = 0.017). ROC curve analysis indicated that the T2 relaxation value of left parietal gray matter had the largest area under the curve, at 0.752. ConclusionsSyMRI parameters can indicate abnormal brain microstructure in GDD children, aiding in understanding GDD pathophysiology and offering potential value for early recognition and diagnosis. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.007 ObjectiveTo investigate the alterations in gray matter structure and their correlation with cognitive function in patients with Alzheimer^,s disease (AD) using voxel-based morphometry (VBM) and surface-based morphometry (SBM). Materials and MethodsSixty-one participants from the Alzheimer^,s Disease Neuroimaging Initiative (ADNI) database were enrolled, comprising 32 AD patients and 29 healthy controls (HC). Morphological metrics, including gray matter volume, cortical thickness, and cortical complexity (local gyrification index, fractal dimension, and sulcal depth), were derived using VBM and SBM pipelines. Regional values from brain areas exhibiting significant intergroup differences were extracted and correlated with cognitive scale scores using Pearson^,s analysis. ResultsWhole-brain structural analysis indicated that the AD group had significantly lower total gray matter volume (P = 0.021) and higher cerebrospinal fluid volume (P = 0.011) than the HC group. VBM identified gray matter reduction in the AD group within the left hippocampus, right parahippocampal gyrus, bilateral middle and inferior temporal gyri, and the left mid-cingulate gyrus (voxel-level FWE-corrected P < 0.001). SBM revealed statistically significant group differences in cortical thickness and local gyrification index (cluster-level FWE-corrected P < 0.05). Cortical thinning was observed in the bilateral inferior parietal lobules and bilateral middle, superior, and inferior temporal gyri, whereas the local gyrification index was elevated in the left middle and inferior temporal gyri. No significant between-group differences were detected in fractal dimension or sulcal depth (FWE-uncorrected P > 0.05). Cognitive assessment confirmed marked differences in Mini-Mental State Examination (MMSE) and Clinical Dementia Rating (CDR) scores (P < 0.001). In the AD group, MMSE scores correlated positively with left hippocampal volume, right parahippocampal gyrus volume, and cortical thickness of the right precuneus and left middle temporal gyrus. Conversely, CDR scores were negatively correlated with volumes of the left hippocampus and right parahippocampal gyrus. ConclusionsVBM and SBM analyses effectively identified characteristic gray matter atrophy in the limbic system and association cortices of Alzheimer^,s disease patients, which was significantly correlated with the severity of cognitive impairment. Furthermore, the study revealed an increased local gyrification index in the left temporal lobe. This finding deviates from the conventional model of linear degenerative changes, suggesting that the local gyrification index holds promise as a novel neuroimaging biomarker for detecting microstructural alterations in AD. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.008 ObjectiveTo quantitatively assess metabolic ratios and hippocampal volumetric parameters in patients with drug-refractory temporal lobe epilepsy (TLE) using proton magnetic resonance spectroscopy (¹H-MRS) and artificial intelligence (AI)-based automated brain segmentation, and to evaluate their diagnostic efficacy in lateralizing the epileptogenic focus. Materials and MethodsA total of 27 TLE patients and 27 healthy controls underwent three-dimensional T1-weighted imaging (3D-T1WI) and multi-voxel ¹H-MRS on a 3.0 T MRI scanner. Hippocampal volume ratio (defined as the percentage of hippocampal volume relative to total brain volume) was measured bilaterally in both groups using United Imaging Intelligence brain segmentation technology. Glutamate-to-creatine ratio (Glu/Cr) and N-acetylaspartate-to-creatine ratio (NAA/Cr) were measured in the bilateral hippocampal head, bodyand tail. Results¹H-MRS: The Glu/Cr ratio in the affected hippocampal head of TLE patients was significantly higher than that in the contralateral side (P = 0.012) and the control group (P < 0.001). No significant differences in NAA/Cr ratios were found between the affected and contralateral hippocampal subregions (P > 0.05). Volumetry: The hippocampal volume ratio on the affected side was significantly lower than that on the contralateral side (P = 0.033) and in the control group (P < 0.001). Diagnostic performance [receiver operating characteristic (ROC) analysis, hippocampal head Glu/Cr ratio for lateralization: Sensitivity 88.9%, Specificity 70.4%, area under the curve (AUC) = 0.859]. Hippocampal volume ratio for lateralization: Specificity 88.9%, Sensitivity 70.4% (AUC = 0.808). Combined diagnosis (Glu/Cr + volume ratio): Sensitivity 81.5%, Specificity 96.3% (AUC = 0.941). ConclusionsThis study confirms the presence of specific glutamate excitotoxicity (↑Glu/Cr) and structural atrophy in the affected hippocampal head of TLE patients. The combined application of hippocampal head Glu/Cr ratio (high sensitivity) and hippocampal volume ratio (high specificity) significantly enhances the efficacy for lateralizing the epileptogenic focus (AUC = 0.941), outperforming either single metric. This multimodal MRI biomarker combination provides a high-value tool for non-invasive pre-surgical localization in drug-refractory TLE. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.009 ObjectiveThis study combines resting-state functional magnetic resonance imaging (rs-fMRI) with traditional graph theory analysis methods to see how brain functional network indicators change in patients with different levels of cerebral small vessel disease (CSVD) and to study how this relates to cognitive function. Materials and MethodsThere were altogether 23 cases of patients with mild CSVD burden (CSVD-m), 22 cases of patients with moderate to severe CSVD burden (CSVD-s), and 21 cases of healthy controls (HC). All rs-fMRI and related cranial MRI imaging data, clinical and laboratory data, and related cognitive scale scores were taken. The differences in functional network between the three groups were analyzed, and the correlation between the difference data and brain regions and cognitive scale scores were analyzed. ResultsFor clinical results, the Montreal Cognitive Assessment (MoCA) result in the CSVD-s group were lower than those in the HC group (P < 0.05). For global results, the CSVD-s group had lower global efficiency (Eg) and higher shortest path length (Lp) and nodal clustering coefficient (Cp) compared with the HC group; the CSVD-s group had higher Lp compared with the CSVD-m group (P < 0.05). For local results, compared with the HC group, the brain regions with reduced nodal efficiency (NE) in the CSVD-s group were in the right superior frontal gyrus orbital region and the left inferior parietal lobule, and the brain region with increased nodal clustering coefficient (NCp) in the CSVD-s group was in the left superior frontal gyrus orbital region. In the CSVD-s group, the brain region with decreased NE was in the left inferior parietal lobule compared with the CSVD-m group (both P < 0.05). For cognitive correlation analysis, in the left subparietal lobule, NE was positively correlated with MoCA scores (r = 0.339, P < 0.05). ConclusionsBoth global and local network results were changed in patients with different loads of CSVD, which showed that the brain^,s infomation-processing ability was reduced to different levels, and there was a compensatory system. Moreover, some of the changes in results were correlated with cognitive fuction, which offered a deeper understanding of the pathophysiological mechanisms behind the differences in their clinical symptoms on an imaging basis. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.010 ObjectiveTo investigate the value of multiplexed sensitivity encoding diffusion weighted imaging (MUSE-DWI) combined with amide proton transfer (APT) imaging in preoperatively assessing meningioma consistency. Materials and MethodsA retrospective analysis was performed on 71 patients with meningioma who underwent tumor resection at the General Hospital of Ningxia Medical University between January 2024 and August 2025. All patients had complete pathological results and comprehensive intraoperative surgical records. Preoperatively, each patient underwent conventional MRI, MUSE-DWI, amide proton transfer (APT) imaging, and contrast enhanced T1-weighted imaging (CE-T1WI). During surgery, tumor consistency was assessed according to the Zada classification scale and categorized into a soft group or a non-soft group. The apparent diffusion coefficient (ADC) and APT values were measured within the enhancing region. Independent samples t-test or Mann-Whitney U test was used to compare ADC and APT values between groups. Parameters showing significant differences were included in a multivariate logistic regression analysis. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of individual parameters and their combination for predicting meningioma consistency. The areas under the ROC curves (AUCs) were compared using DeLong^,s test. ResultsThe soft meningioma group showed significantly higher ADC and APT values than the non-soft group (P = 0.011 and P < 0.001, respectively). Among all single parameters, APT value demonstrated the highest diagnostic efficacy for differentiating soft from non-soft meningiomas (AUC = 0.915), which was higher than that of the ADC value (AUC = 0.675). The multiparameter combined prediction model (ADC + APT) achieved an AUC of 0.947, which is higher than that of any single parameter. The DeLong test demonstrated that the multiparameter combined model achieved significantly superior diagnostic performance compared to the ADC value (P < 0.05), while no statistically significant difference was observed in the AUC between the combined model and APT (P = 0.061). ConclusionsBoth MUSE-DWI and APT techniques are useful for the noninvasive preoperative prediction of meningioma consistency. Their combination provides the highest diagnostic performance. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.011 ObjectiveTo construct a radiomics model by extracting high-throughput radiomics features from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) sequences, and further explore its value in preoperatively differentiating high-grade gliomas (HGG) from solitary brain metastases (SBM). Materials and MethodsThis retrospective study included 135 patients (HGG, n = 89; SBM, n = 46) treated at Changzhou First People^,s Hospital from May 2016 to December 2024, all with complete imaging data (DCE-MRI and conventional MRI) and histopathology. Two neuroradiologists (3 and 10 years of experience) manually delineated whole-tumor regions of interest, from which 963 radiomics features were extracted from each region of interest (ROI). Feature redundancy was reduced using the Mann-Whitney U test, Pearson correlation analysis, and hierarchical clustering, followed by least absolute shrinkage and selection operator (LASSO) for optimization. Subsequently, logistic regression (LR) was employed to establish three models: the DCE model, conventional MRI model, and fusion model. Model performance was evaluated by receiver operating characteristic analysis, and differences were compared with the DeLong test (P < 0.05 significant). ResultsIn the test set, both the fusion model and the DCE model demonstrated high diagnostic performance [area under the curves (AUCs) are 0.934 (95% CI: 0.860 to 1.000) and 0.908 (95% CI: 0.812 to 1.000), respectively], each significantly outperforming the conventional MRI model (P = 0.001 and P = 0.011, respectively). Although the fusion model showed a numerically higher AUC than the DCE model, the difference was not statistically significant (P = 0.387). ConclusionsThe DCE model demonstrated high performance in preoperatively discriminating between high-grade gliomas and solitary brain metastases, achieving comparable diagnostic efficacy to the fused model. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.012 ObjectiveTo explore the value of orbital parameters combined with serological parameters in the diagnosis of Graves^,s orbitopathy (GO) activity by quantitative analysis. Materials and MethodsA retrospective collection of 47 patients diagnosed with GO who visited the Endocrinology Department of North China University of Science and Technology Affiliated Hospital from January 2024 to February 2025 was conducted. Patients were divided into active and inactive groups based on Clinical Activity Score (CAS), and a control group consisting of 15 healthy individuals was also collected. All subjects were examined by MRI and serology. The maximal diameter, maximal cross-sectional area, T2 relaxation time (T2RT) and exophthalmos of each extraocular muscle of the three groups were measured respectively, and the clinical data of the three groups were collected. The parameters were statistically analyzed by SPSS 27.0 software, and the parameters with statistical differences were quantified by binary logistic regression combined with MRI. The receiver operating characteristics (ROC) curve was used to evaluate the predictive efficacy of quantitative parameters of extraocular muscles on GO activity, and the area under the curve (AUC) was compared by DeLong method. Spearman^,s correlation analysis was used to analyze the relationship between imaging parameters and CAS score. ResultsThe maximal diameter, maximal cross-sectional area, T2RTmax, T2RTmean and exophthalmos in the activity group were statistically significant (P < 0.05), and the quantitative parameters in the activity group were significantly increased. ROC curve analysis shows that TRAb, T2RTmax and exophthalmos have the best diagnostic efficacy, which is higher than the single diagnostic value of the three groups. The AUC value is 0.905, the sensitivity is 91.7%, the specificity is 81.0%, and the cutoff value is 0.44. The maximumcross-sectional area (r = 0.73, P < 0.05), diameter (r = 0.70, P < 0.05), T2RT (r = 0.83, P < 0.05) and exophthalmos (r = 0.75, P < 0.05) were positively correlated with CAS score. ConclusionsMulti-parameters of magnetic resonance imaging and serological indexes can evaluate the activity of GO, and the combination of them can improve the efficacy of clinical diagnosis of GO. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.013 ObjectiveTo explore the diagnostic value of whole-body MRI (WB-MRI) in postoperative incidental lesions of breast cancer. Materials and MethodsThe WB-MRI images of 222 patients with breast cancer after operation in Wuhan Red Cross Hospital from January 2022 to February 2025 were retrospectively analyzed. Based on international guidelines, all incidental lesions were divided into significant and non-significant groups. The diagnostic efficacy of WB-MRI was compared with conventional imaging examinations (CT plain scan and ultrasound) with pathological or follow-up results as the gold standard, and the relationship between age, postoperative time and the number of lesions was analyzed. ResultsA total of 736 incidental lesions were detected in 222 patients by WB-MRI, and the detection rate was 98.20% (218/222), among which 132 were significant lesions (17.93%). These were mainly distributed in chest (16.67%), liver (12.12%), pancreas (9.85%), uterine adnexa (10.61%) and skeleton (10.61%). For the detection of pulmonary lesions, the sensitivity of WB-MRI was lower than that of CT plain scan (63.64% vs. 95.45%), with a statistically significant difference (P = 0.009), and the specificity of both was low. For the detection of abdominal solid organ lesions. The sensitivity of WB-MRI was higher than that of CT and ultrasound (91.18% vs. 55.88%/41.18%), with statistically significant differences (both P < 0.001), its specificity (66.7%) was also better than that of ultrasound (33.3%). For the detection of uterine adnexal lesions. There was no significant difference in sensitivity between WB-MRI and ultrasound (84.3% vs. 92.2%, P = 0.354), but the specificity was equal (both were 50%). The sensitivity of WB-MRI in detecting bone lesions was significantly higher than that of CT (97.1% vs. 67.4%), and the difference was statistically significant (P = 0.004), and the specificity was also better than that of CT. There was no significant correlation between patients^, age and postoperative time and the proportion of significant lesions (Age: P = 0.121; Postoperative time: P > 0.05). ConclusionsWB-MRI has strong ability to detect postoperative incidental lesions of breast cancer patients, especially abdominal, pelvic and bone lesions, and can be used as an effective tool for systemic screening. It is suggested that it should be combined with CT and other examinations to provide more accurate diagnostic information for clinic. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.014 ObjectiveTo evaluate the application value of 5.0 T ultra-high-field magnetic resonance imaging (MRI) susceptibility weighted imaging (SWI) combined with multi-echo T2^* mapping in microvascular invasion (MVI) of hepatocellular carcinoma (HCC). Materials and MethodsA retrospective analysis was conducted on 44 HCC patients who met the inclusion and exclusion criteria. The SWI features of the lesions were analyzed, and the intratumoral susceptibility signal intensity (ITSS) was graded. Additionally, the R2^* values of the lesions, peritumoral areas, and background liver tissue were calculated, along with the △R2^* values. Statistical analysis was performed to examine the differences in these parameters under different MVI statuses. ResultsAccording to pathological standards, the MVI-positive group demonstrated higher ITSS grading and a greater propensity for associated hemorrhage. The R2^* values of lesion, peritumoral and ΔR2^* values in the MVI-positive and MVI-negative groups were (60.07 ± 25.22) Hz and (52.05 ± 21.51) Hz, (94.32 ± 29.32) Hz and (91.06 ± 30.69) Hz, (0.39 ± 0.40) and (0.13 ± 0.12), respectively. The corresponding P-values were 0.312, 0.740, and < 0.001. The difference in ΔR2^* value was statistically significant, and its diagnostic performance was superior to that of ITSS grading (AUC = 0.838, 0.686). ConclusionsBased on the 5.0 T ultra-high-field MRI SWI sequence, a high ITSS grade and the △R2^* value from the T2^* mapping sequence are potential predictive biomarkers for HCC MVI. The combination of these two parameters can enhance diagnostic performance. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.015 ObjectiveTo explore the predictive value of a machine learning model integrating dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters and clinical risk factors for microsatellite instability (MSI) status in rectal cancer. Materials and MethodsA retrospective analysis was conducted on 150 rectal cancer patients treated at Binzhou Medical University Hospital between May 2022 and July 2024, including 27 with MSI and 123 with microsatellite stability (MSS). MRI axial T2WI, apparent diffusion coefficient (ADC) maps, and dynamic contrast-enhanced T1WI (DCE-T1WI) were used to delineate the tumor^,s largest cross-sectional area as the region of interest (ROI). Radiomic features were extracted and reduced to identify optimal features. Independent clinical predictors and DCE parameters for MSI were selected using multivariate logistic regression. The dataset was split into training and validation sets in a 7∶3 ratio. Nine machine learning algorithms, extreme gradient boosting classifier (XGBoost), logistic regression (LR), light gradient boosting machine classifier (LGBM), random forest (RF), decision tree classifier (DT), Gaussian naive Bayes (GNB), support vector classifier (SVM), multilayer perceptron classifier (MLP), and adaptive boosting classifier (AdaBoost) were employed to construct predictive models. The performance of each model in predicting MSI status was evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA). Additionally, a temporal validation set comprising 30 rectal cancer patients from the same hospital between December 2024 and June 2025 was used to assess model generalizability via ROC analysis. ResultsThe GNB model demonstrated the most stable performance. The combined model, incorporating independent clinical risk factors, radiomics scores, and DCE perfusion parameters, demonstrated superior predictive performance for MSI status in rectal cancer, with area under the curve (AUC) values of 0.920 (95% CI: 0.821 to 1.000), 0.900 (95% CI: 0.786 to 1.000), and 0.817 (95% CI: 0.667 to 0.966) in the training set, validation set, and temporal validation set, respectively. ConclusionsAmong the nine machine learning algorithms evaluated, GNB exhibited the best performance in predicting MSI status in rectal cancer, which was further validated using a temporal validation set. Machine learning models incorporating DCE-MRI parameters and clinical risk factors show promising value in predicting MSI status in rectal cancer. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.016 ObjectiveTo develop and validate a nomogram based on intravoxel incoherent motion magnetic resonance imaging (IVIM-MRI) radiomics and clinical indicators for assessing the severity of renal fibrosis (RF) in patients with chronic kidney disease (CKD). Materials and MethodsThis was a case-control study. The clinical and imaging data from 132 CKD patients confirmed by renal biopsy at the First People^,s Hospital of Changzhou between September 2016 and July 2022 were retrospectively analyzed. The patients were randomly divided into a training set (n = 92) and a test set (n = 40) in a 7∶3 ratio. Based on the T score of the Oxford MEST-C classification, patients were grouped into a mild fibrosis group (T0, fibrosis ≤ 25%) and a moderate-to-severe fibrosis group (T1-T2, fibrosis > 25%). Clinical indicators showing significant differences between groups were selected for subsequent clinical modeling. All patients underwent IVIM-MRI before biopsy, and radiomic features were extracted from true diffusion coefficient (D)、pseudo-diffusion coefficient (D^*) and perfusion fraction (f) maps. Feature selection was performed using the Mann-Whitney U test, Pearson correlation analysis, and least absolute shrinkage and selection operator regression. Four radiomics models (Rad_D, Rad_D^*, Rad_f, and Rad_D_D^*_f) and a clinical model were constructed using logistic regression (LR), random forest, and multilayer perceptron algorithms. The optimal-performing radiomics and clinical models were then integrated to build a nomogram. Model performance was evaluated using receiver operating characteristic (ROC) curve analysis, the DeLong test, decision curve analysis (DCA), and calibration curves. ResultsA total of 9, 8, 11, and 12 features were selected for the construction of the Rad_D, Rad_D^*, Rad_f, and Rad_D_D^*_f models, respectively. Among the four radiomics models, Rad_D_D^*_f demonstrated the best performance in distinguishing between mild and moderate-to-severe RF. Among the three algorithms comparison, both the Rad_D_D^*_f radiomics model and the clinical model achieved the highest diagnostic performance using the LR algorithm. The nomogram, combining the best-performing radiomics and clinical models, further improved diagnostic performance, with area under the curve (AUC) of 0.942 (95% CI: 0.896 to 0.989) and 0.820 (95% CI: 0.687 to 0.954) in the training and test sets, respectively. The DeLong test showed that the nomogram significantly outperformed the clinical model (P < 0.05). DCA and calibration curves confirmed that the nomogram provided higher net clinical benefit and good model calibration. ConclusionsThe nomogram integrating IVIM-MRI radiomics and clinical indicators enables noninvasive identification of RF severity in CKD patients, demonstrating potential clinical applicability. This tool may provide imaging-based support for the precise management and dynamic assessment of CKD. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.017 ObjectiveTo investigate the value of biparametric magnetic resonance imaging (bpMRI) habitat imaging analysis in diagnosing high-risk and low-risk prostate cancner (PCa). Materials and MethodsA retrospective analysis was conducted on 191 PCa patients confirmed by biopsy or surgical pathology at Ma^,anshan People^,s Hospital between December 2023 and August 2024, including 131 high-risk PCa cases and 60 low-risk PCa cases. The 191 patients were randomly divided into training and testing sets at a 7∶3 ratio. All patients underwent bpMRI scans, with preprocessing performed on T2WI, ZOOMit diffusion weighted imaging (ZOOMit-DWI), and apparent diffusion coefficient (ADC) sequences. Nineteen radiomics features were extracted from ADC images. By integrating T2WI and ZOOMit-DWI images, unsupervised K-means clustering was used to generate similar subregions across all tumor voxels. Based on the habitat subregion results, the intratumoral heterogeneity score (ITHscore) was calculated for the 191 patients. Radiomics features were extracted from the subregions, followed by dimensionality reduction and filtering to select features with the highest correlation coefficients. SHAP analysis was employed to visualize feature importance. After feature fusion and selection, 10 habitat radiomics models were established. For each model, the threshold, sensitivity, specificity, accuracy, negative predictive value, and positive predictive value were calculated. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was computed to evaluate diagnostic performance. Decision curve analysis (DCA) was further used to assess the net benefit of the models. ResultsIn both training and testing sets, the difference in total prostate specific antigen (tPSA) between high-risk and low-risk PCa was statistically significant. Based on the Calinski-Harabasz (CH) value, 2 was determined as the optimal number of habitat subregions. SHAP analysis revealed that the original_glszm_ZoneEntropy feature in the h2 subregion had the greatest impact. A total of 10 classifiers were employed for model construction. The habitat radiomics model was compared with the clinical + prostate imaging reporting and data system (PI-RADS) score model. In the habitat radiomics model, the extremely randomized trees (ExtraTrees) model demonstrated the best prediction performance in the test set, with a training set AUC of 0.838 [95% (confidence interval, CI): 0.768 to 0.908] and a test set AUC of 0.796 (95% CI: 0.665 to 0.927). For the model constructed using clinical data and PI-RADS scores, the logistic regression (LR) model achieved the highest predictive performance, with a training set AUC of 0.786 (95% CI: 0.705 to 0.866) and a test set AUC of 0.719 (95% CI: 0.550 to 0.887). The clinical net benefit of both models was evaluated using DCA, and the habitat radiomics model exhibited both a higher AUC value range than the clinical-PI-RADS score model. ConclusionsBased on bpMRI habitat imaging analysis, the pathological classification of PCa can be diagnosed relatively accurately, which is helpful for the clinical diagnosis and risk prediction of PCa. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.018 ObjectiveTo investigate the diagnostic value of synthetic magnetic resonance imaging (SyMRI) combined with amide proton transfer-weighted (APTw) imaging in the ISUP grading of prostate cancer (PCa). Materials and MethodsTotal 78 patients with pathologically confirmed PCa were retrospectively enrolled in The First Affiliated Hospital of Soochow University form April 2024 to April 2025. Based on pathological results, the PCa were classified according to the ISUP grade system. All patients underwent conventional MRI sequences, MAGiC, and APTw imaging scans. Longitudinal relaxation time (T1), transverse relaxation time (T2), proton density (PD), amide proton transfer rate (APT), and apparent diffusion coefficient (ADC) values were measured. Independent Student t test or Mann-Whitney U test were used to assess differences in quantitative values between low-grade PCa (ISUP grade 1) and intermediate/high-grade PCa (ISUP grade ≥ 2). Parametric variables were correlated with the ISUP grades using the Spearman rank correlation coefficient. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic efficiency of individual parameters and combined models in distinguishing intermediate/high-grade PCa. ResultsIntermediate/high-grade PCa group showed significantly lower T2 and ADC values but higher APT value than low-grade group (P < 0.05). There was no significant differences in T1 and PD values (P > 0.05). ISUP grade demonstrated a significant positive correlation with APT value (r = 0.359, P = 0.001) and significant negative correlations with T2 value (r = -0.304, P = 0.007) and ADC value (r = -0.535, P < 0.001). No significant correlations were found with T1 and PD values (r = -0.158, -0.103, both P > 0.05). For diagnosing intermediate/high-grade PCa, the AUCs were 0.71 (95% CI: 0.60 to 0.80) for T2 value and 0.75 (95% CI: 0.64 to 0.84) for APT value. Combined model was established by integrating T2 and APT values. AUC of the combined model was 0.78 (95% CI: 0.67 to 0.87), showing no significant difference compared with ADC value (AUC = 0.86, 95% CI: 0.76 to 0.92, DeLong test: P > 0.05). ConclusionsSyMRI and APTw imaging are beneficial for grading of PCa. The combined model demonstrates diagnostic performance comparable to ADC value in distinguishing intermediate/high-grade PCa, providing valuable guidance for grading and clinical decision of PCa. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.019 ObjectiveTo investigate the value of intratumoral and peritumoral radiomics features based on MRI images for preoperative noninvasively predicting lymphovascular space invasion (LVSI) status in endometrial carcinoma (EC) patients. Materials and MethodsClinical and routine imaging features of 222 patients with histopathologically proved EC were retrospectively analyzed. Radiomics features from both intra- and peritumoral regions in T2-weighted imaging (T2WI), the contrast-enhanced T1-weighted images (CE-T1WI) at delayed phase and apparent diffusion coefficient (ADC) were extracted. The independent risk factors were identified through univariate and multivariate logistic analysis to construct predictive models (clinical, radiomics and combined). Receiver operating characteristic (ROC) curve was used to analyze the prediction efficiency of these models. Decision curve analysis (DCA) and calibration curves were utilized to assess the clinical utility and calibration performance of the models, respectively. ResultsThe radiomics model established based on peritumoral 3 mm in T2WI sequences showed best performance, and the AUC of the training group and validation group were 0.902 and 0.803, respectively. The combined model based on tumor maximum diameter, the value of ADC and radiomics features had the optimal performance and achieved AUC values of 0.903 and 0.801 in the training and validation cohorts respectively. The calibration curve results indicated that the combined model had good calibration, and the net benefit of the model was the highest in the decision curve analysis. ConclusionsThe intratumoral and peritumoral radiomics models of EC based on MRI images have good clinical performance and can be applied to predict LVSI characteristics of EC noninvasively. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.020 ObjectiveTo compare the efficacy of MRI radiomics models combined with clinical features, constructed using different machine learning (ML) algorithms, for predicting treatment outcomes in adenomyosis patients after high-intensity focused ultrasound (HIFU). Materials and MethodsImaging and clinical data from 169 adenomyosis patients who underwent HIFU treatment between September 2021 and May 2024 and met inclusion/exclusion criteria were retrospectively collected. Postoperative non-perfused volume (NPV) was assessed via MRI. Patients were stratified into a significant-efficacy group [NPV ratio (NPVR) ≥ 50%, n = 76] and a non-significant-efficacy group (NPVR < 50%, n = 93) using the threshold NPVR = 50% (NPVR = NPV / total lesion volume). Lesions were segmented using 3D Slicer software for feature extraction. Eight ML algorithms were used to build models: decision tree (DT), Gaussian process (GP), logistic regression (LR), partial least squares discriminant analysis (PLSDA), quadratic discriminant analysis (QDA), random forest (RF), stochastic gradient descent (SGD), and support vector machine (SVM). Model performance was evaluated using receiver operating characteristic (ROC) curves, with calculation of the area under the curve (AUC), sensitivity, specificity, accuracy, precision, and F1-score. DeLong test compared inter-model differences (statistical significance: P < 0.05). ResultsRadiomics-clinical models based on DT, GP, LR, PLSDA, QDA, RF, SGD, and SVM algorithms were constructed. Training set AUCs were: 0.865 (95% CI: 0.806 to 0.924), 0.713 (95% CI: 0.619 to 0.807), 0.666 (95% CI: 0.567 to 0.764), 0.669 (95% CI: 0.571 to 0.767), 0.649 (95% CI: 0.550 to 0.749), 0.796 (95% CI: 0.717 to 0.876), 0.425 (95% CI: 0.341 to 0.508), and 0.666 (95% CI: 0.568 to 0.764), respectively. Test set AUCs were: 0.788 (95% CI: 0.669 to 0.907), 0.738 (95% CI: 0.601 to 0.874), 0.719 (95% CI: 0.578 to 0.860), 0.730 (95% CI: 0.592 to 0.868), 0.738 (95% CI: 0.600 to 0.875), 0.731 (95% CI: 0.587 to 0.876), 0.332 (95% CI: 0.221 to 0.444), and 0.719 (95% CI: 0.575 to 0.863), respectively. The DT model achieved the highest AUC, specificity, precision, and accuracy in the test set, and the highest AUC and F1-score in the training set. SGD and PLSDA models performed poorly in both sets. ConclusionsMRI radiomics-clinical models built using six ML algorithms (DT, GP, LR, QDA, RF, SVM) demonstrated good predictive performance for post-HIFU efficacy in adenomyosis. The DT model exhibited optimal performance and is recommended as the preferred method for outcome prediction, assisting clinicians in developing personalized treatment plans and management strategies. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.021 ObjectiveTo evaluate the efficacy and prognosis of uterine leiomyomas after high-intensity focused ultrasound (HIFU) ablation using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Materials and MethodsA retrospective analysis was performed on 116 patients with uterine fibroids who underwent HIFU treatment from January 2023 to December 2024, completed 6 months of postoperative follow-up, and had complete MRI data. Two physicians divided the patients into the adequate ablation group (NPVR ≥ 80%) and the inadequate ablation group (NPVR < 80%) based on the non-perfusion volume ratio (NPVR). The differences in clinical data and MRI parameters between the two groups were analyzed using t-test or chi-square test. In addition, at 6 months after surgery, the two physicians reclassified the patients based on the proportion of residual fibroid volume: the group with residual fibroid tissue ≥ 10% (incomplete ablation) was defined as the regrowth group, and the group with residual fibroid tissue < 10% as the non-growth group. The differences in MRI parameters between the two groups were compared using t-test. Multivariate logistic regression was used to explore the potential predictors of the efficacy of HIFU in the treatment of uterine fibroids. ResultsIndependent sample t-test showed statistically significant differences between groups in subcutaneous fat thickness, dynamic contrast-enhanced signal intensity values, and the number of myoma recurrence cases (P < 0.05). Multivariate logistic regression analysis indicated that NPVR (OR = 0.219) and pre-treatment tumor volume (OR = 0.993) were protective factors against myoma recurrence, while myoma signal intensity on T2WI (OR = 8.975) was an independent risk factor. The initial area under the curve (iAUC) at 6 months between the regrowth group and the pre-operative measurement (P < 0.05). ConclusionsThe iAUC value can effectively predict the risk of residual myoma regrowth after high-intensity focused ultrasound (HIFU) ablation for uterine leiomyomas. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.022 Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.023 ^, daily lives. The pathological mechanism of PSCI is complex, involving various factors such as impaired brain reserve, blood-brain barrier disruption, gut microbiota dysbiosis, and reduced synaptic plasticity, leading to brain dysfunction; in-depth investigation of the central effects of PSCI is key to clinically developing scientific interventions and improving patients^, quality of life. In recent years, functional magnetic resonance imaging (fMRI) technology has developed rapidly; resting-state fMRI (rs-fMRI), with its advantages of non-invasiveness and high temporal and spatial resolution, has become an important tool for evaluating the neural mechanisms of PSCI; PSCI exhibits widespread functional connectivity (FC) abnormalities, including reduced FC in some resting-state networks (RSN), altered cerebellar activity, and decreased dynamic functional network connectivity (dFNC); directed FC analysis shows enhanced information transmission in the contralateral brain regions and impaired RSN in the early stages of stroke. This paper uses voxel-based and node-based analysis methods for FC, systematically reviewing the literature on abnormal local FC, changes in brain network topological properties, and characteristics of directed FC dynamics in PSCI. It also identifies current research limitations and suggests future directions, aiming to inform early diagnosis and refine personalized treatment strategies. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.024 ^, robustness and universality through multicenter independent cohorts; Furthermore, the molecular mechanisms underlying neuroplastic changes and their causal relationship with clinical manifestations remain unclear. Future research urgently requires large-scale, multicenter prospective cohort studies to validate existing neurobiological markers. Integrating multi-omics technologies with behavioral assessments will elucidate the driving mechanisms of central neural remodeling, ultimately advancing the translation of neuroimaging biomarkers into clinical prediction models and targeted intervention strategies. This research direction holds profound scientific significance and clinical value for overcoming bottlenecks in the clinical management of chronic neck pain. This paper systematically integrates neuroimaging evidence to elucidate the central neural mechanisms underlying the transition from acute pain to chronic status. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.025 Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.026 ^,s disease (AD) is a disease characterized by progressive cognitive decline as the main clinical manifestation, which seriously affects patients^, physical and mental health and social development. Acupuncture is one of the effective methods for treating AD, which can significantly improve AD symptoms and delay disease progression, but its central mechanism has not been fully clarified. Functional magnetic resonance imaging (fMRI) technology, with advantages such as no radiation, high temporal and spatial resolution, and high imaging clarity, provides a visual means for in-depth study of the central mechanism of acupuncture in the treatment of AD. By sorting out and summarizing the studies on the central mechanism of resting-state functional magnetic resonance imaging (rs-fMRI) in acupuncture treatment of AD, elaborates on the limitations of current research and points out the directions for future studies,this paper finds that AD patients have pathological manifestations such as structural damage in brain regions, abnormal functional connections between brain regions, and weakened connection effects of brain networks. Acupuncture can exert its role in treating AD by reducing structural damage in brain regions such as the hippocampus, temporal lobe, cingulate gyrus, and caudate nucleus, improving abnormal functional connections between brain regions such as the posterior cingulate gyrus, temporal lobe, parietal lobe, and frontal lobe, enhancing the connectivity effects of large-scale brain networks, including the default mode network (DMN), fronto-parietal network (FPN), central executive network (CEN), sensorimotor network (SMN), as well as the inter-network connectivity between the dorsal attention network (DAN) and DMN, so as to exert a therapeutic effect on AD, with the aim of providing a more comprehensive reference for the in-depth research on the central mechanism of acupuncture in the treatment of AD and the optimization of clinical protocols in the future. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.027 ^,s disease (AD), and there is no effective cure currently available for Alzheimer^,s disease. Consequently, early diagnosis of MCI is crucial for preventing or slowing the progression of AD. The development of multimodal MRI and artificial intelligence (AI) technologies has introduced novel methodologies and perspectives into research on MCI, demonstrating significant potential in the diagnosis of MCI and the prediction of its progression to AD. Nevertheless, several challenges remain in this field, including insufficient standardization of multi-center data and limited generalizability of computational models. This review systematically summarizes recent advances in the integration of multimodal MRI with machine learning and deep learning for MCI classification and AD conversion prediction. Furthermore, it underscores the necessity of establishing unified protocols for multi-center neuroimaging data and developing standardized frameworks for evaluating model robustness. Finally, we propose a promising future direction that integrates multimodal neuroimaging with genetic profiling, with the aim of constructing a more comprehensive biological characterization of MCI and enhancing early intervention strategies. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.028 Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.029 Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.030 ^, physical and mental health. Tumor-infiltrating lymphocytes (TILs) play a critical role in the diagnosis and treatment of breast cancer, with their levels closely associated with patients^, sensitivity to immunotherapy and prognosis. However, traditional pathological assessment of TILs has several drawbacks, including subjectivity, time-consuming procedures, and the need for invasive biopsies. Emerging radiomics technology extracts quantitative features from medical images and combines them with machine learning models to achieve non-invasive prediction of TILs. This effectively overcomes the limitations of traditional methods, provides an innovative approach for the precise diagnosis and treatment of breast cancer, and holds significant clinical application value. This study reviews literature from PubMed and the China National Knowledge Infrastructure (CNKI) on radiomics-based prediction of TILs in breast cancer. It systematically summarizes key aspects including imaging modality selection and research trends, critically analyzes current limitations in the field, and proposes promising directions for future research. The review aims to provide valuable radiological insights for predicting TILs in breast cancer, facilitate non-invasive assessment of TIL levels, and advance the development of precision medicine. ]]> Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.031 Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.032 Sat,20 Dec 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.12.033 Sat,20 Dec 2025 00:00:00 GMT