http://med-sci.cn/cgzcx/en/contents_list.asp?issue=202505 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.1674-8034.2025.05.001 Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.002 ObjectiveTo explore the diagnostic value of non-contrast MRI in evaluating benign and malignant mural nodules of intraductal papillary mucinous neoplasms (IPMN) of the pancreas. Materials and MethodsPatients with pancreatic IPMN who were pathologically diagnosed after radical surgical resection in the First Affiliated Hospital of Naval Medical University from August 2013 to February 2024 were collected. A total of 238 patients with IPMN containing mural nodules were screened out through non-contrast MRI. According to the pathological grading criteria of IPMN, the patients were divided into the benign mural nodule group (low-grade dysplasia) and the malignant mural nodule group (high-grade dysplasia, invasive carcinoma). Independent sample t-test, rank sum test and chi-square test were used for statistical analysis to compare the differences in non-contrast MRI features between the two groups. Multivariate logistic regression was used to analyze the independent predictors of the malignant mural nodule group of pancreatic IPMN, and a diagnostic model was constructed. The receiver operating characteristic (ROC) curve was drawn and the area under the curve (AUC) was calculated. ResultsThe diameter of mural nodules in the malignant mural nodule group [15.00 (10.00, 23.00) mm] was larger than that in the benign mural nodule group [7.00 (6.00, 9.75) mm], and the difference was statistically significant (P < 0.001). In the malignant mural nodule group, the manifestations such as main pancreatic duct or mixed pancreatic duct type, high signal of cystic fluid on T1WI, main pancreatic duct dilation ≥ 10 mm, cysts ≥ 40 mm, thickening of the cyst wall, and abrupt change in caliber of pancreatic duct with distal pancreatic atrophy. There were statistically significant differences when compared with the benign group of mural nodules (all P < 0.05). The results of multivariate logistic regression analysis showed that the threshold diameter of mural nodules of 9.5 mm, thickening of the cyst wall, and abrupt change in caliber of pancreatic duct with distal pancreatic atrophy were independent predictors for the malignant mural nodule group (all P < 0.05). A model was established by combining these three independent predictors to diagnose the malignant mural nodule group, and the AUC was 0.851 [95% confidence interval (CI): 0.802 to 0.900], with a sensitivity of 70.7% and a specificity of 90.5%. ConclusionsThe threshold diameter of mural nodules of 9.5 mm, thickening of the cyst wall, and abrupt change in caliber of pancreatic duct with distal pancreatic atrophy are independent predictors for evaluating malignant mural nodules of pancreatic IPMN based on the features of non-contrast MRI. The diagnostic model combining these three factors shows good diagnostic efficacy. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.003 ObjectiveTo investigate the clinical utility of baseline three-dimensional magnetic resonance elastography (3D-MRE) viscoelastic parameters in evaluating chemotherapy response and prognosis in patients with advanced pancreatic cancer (APC). Materials and MethodsThis prospective study enrolled patients with histologically confirmed APC between January 2022 and June 2024, all of whom underwent 3D-MRE scans prior to chemotherapy. Chemotherapy response was categorized into responders and non-responders based on RECIST 1.1 criteria. The Youden index was used to determine the optimal cut-off values for imaging parameters in predicting chemotherapy response. Univariate and multivariate logistic regression analyses were performed to identify independent predictors of chemotherapy response, while Cox proportional hazards models were used to analyze factors influencing progression-free survival (PFS) and overall survival (OS). ResultsA total of 59 patients were included, with 21 in the responder group and 38 in the non-responder group. The area under the curve (AUC) for baseline shear stiffness (SS) in diagnosing chemotherapy response was 0.771 (95% CI: 0.640 to 0.902), with an optimal cutoff value of 3.85 kPa. Multivariate logistic regression analysis identified SS ≥ 3.85 kPa as an independent predictor of poor chemotherapy response [odds ratio (OR) = 6.760, 95% CI: 1.197 to 38.174, P=0.030]. Cox regression analysis revealed that SS ≥ 3.85 kPa was independently associated with shorter OS [hazard ratio (HR) = 4.217, 95% CI: 1.796 to 9.889, P=0.001] and PFS (HR = 3.860, 95% CI: 1.801 to 8.273, P=0.001). Patients with SS ≥ 3.85 kPa had significantly shorter PFS (P < 0.001) and OS (P < 0.001). ConclusionsBaseline SS measured by 3D-MRE is an independent predictor of chemotherapy response in APC patients, with higher baseline SS associated with poorer prognosis. As a non-invasive and quantitative biomechanical assessment tool, 3D-MRE holds promise as an important adjunct for personalized treatment in pancreatic cancer. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.004 ObjectiveTo investigate the histological relevance and potential clinical value of multiparametric diffusion-weighted imaging (DWI) parameters for pancreatic cancer through animal experiments and a prospective clinical study. Materials and MethodsTwelve xenograft mouse models of pancreatic cancer and twenty-five patients with histologically confirmed pancreatic ductal adenocarcinoma were enrolled. Multi-model DWI analysis was performed using the mono-exponential model (Mono), intravoxel incoherent motion (IVIM) model, diffusion kurtosis imaging (DKI), stretched exponential model (SEM), fractional-order calculus (FROC) model, and continuous-time random walk (CTRW) model. The extracted parameters included Mono_ADC, IVIM_D, DKI_MD, SEM_D, FROC_D, CTRW_D, IVIM_D^*, IVIM_f, DKI_MK, SEM_α, FROC_β, FROC_mμ, CTRW_α, and CTRW_β. Histological correlations with DWI parameters were evaluated using Masson and Ki-67 staining in the animal cohort. In the clinical study, differences in DWI parameters between pancreatic cancer and pancreatic normal tissue were assessed, and ROC analysis was used to evaluate discriminative ability. ResultsIn animal studies, DKI_MD was significantly negatively correlated with the degree of fibrosis (r = -0.85, P < 0.001), while CTRW_β (r = -0.82, P = 0.001) and FROC_β (r = -0.78, P = 0.002) were closely associated with Ki-67 expression. In clinical data, DWI parameters including DKI_MD, IVIM_f, and FROC_β differed significantly between pancreatic cancer and normal tissues (P < 0.05). DKI_MD showed the highest diagnostic performance individually (AUC = 0.757, 95% CI: 0.615 to 0.867), while the combined model (DKI_MD + FROC_β) achieved improved accuracy (AUC = 0.866, 95% CI: 0.739 to 0.945) with significantly better sensitivity (76%) and specificity (88%). ConclusionsMultiparametric analysis using non-Gaussian DWI models provides valuable insights into the microstructural features of pancreatic cancer. Among them, DKI_MD and FROC_β demonstrated significant advantages in quantifying fibrosis and heterogeneity, indicating their potential as imaging biomarkers. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.005 ObjectiveTo investigate the feasibility and clinical value of quantitative parameters derived from the mDixon-Quant sequence in preoperative non-invasive prediction of histological differentiation grade and Ki-67 expression level in patients with pancreatic ductal adenocarcinoma (PDAC). Materials and MethodsA retrospective analysis was conducted on the clinical, radiological, and a cohort of 57 cases exhibiting pathologically confirmed PDAC. According to the histological differentiation degree, 57 patients were divided into well-differentiated group (n = 30) and poorly differentiated group (n = 27). The basic clinical data of the two groups (age, gender, abdominal pain, jaundice, preoperative CA19-9 level, etc.), conventional imaging features (location, morphology, boundary, long and short diameters of the tumor, whether there is dilation of the pancreatic duct, vascular invasion, etc.) and quantitative parameters [water phase value, fat phase value, T2^* value, R2^* value and fat fraction (FF)] were analyzed. The quantitative parameter values of healthy pancreases were collected for normal control according to the ratio of the case group to the normal group (1∶1). At the same time, 31 cases with Ki-67 expression level results were analyzed, and they were divided into high expression (Ki-67 ≥ 50%) and low expression groups (Ki-67 < 50%). The intra-class correlation coefficient (ICC) was used to evaluate the repeatability. The Mann-Whitney U test, t test or χ² test was used to compare the differences of various parameters between the two groups. The receiver operating characteristic (ROC) curve was drawn and the area under the curve (AUC) was calculated to evaluate the predictive efficacy of relevant indicators. The efficacy of different AUCs was compared by employing the DeLong test. ResultsA statistically significant age disparity was observed between the well-differentiated and poorly differentiated subgroups. Patients with poorly differentiated demonstrated a modestly younger compared to the well-differentiated cohort (P = 0.006). However, there were no statistically significant differences in gender, symptoms, CA19-9 level, mass morphology, location, and whether there was dilation of the pancreatic duct, etc. Except for the water phase value, there were statistically significant differences between the healthy pancreas group and the PDAC group, as well as between the patient^,s lesion and the normal pancreatic area (P < 0.05). Furthermore, the well-differentiated and poorly differentiated cohorts demonstrated significantly divergent T2^* and R2^* parameters (P < 0.05). Compared with the well-differentiated group, the T2^* value of the poorly differentiated group was higher [(58.92 ± 7.84) ms vs. (47.87 ± 6.76) ms]; and the R2^* value was lower [17.73 (15.62, 19.77) s^-1 vs. 21.57 (19.65, 24.69) s^-1]. The AUCs of the T2^* and R2^* values for predicting the histological differentiation degree were 0.866 and 0.827, respectively. The sensitivity and specificity of T2^* were 77.8% and 74.1%, respectively, while those of R2^* were 80.0% and 83.3%, respectively. The combined diagnostic AUC of T2^* and R2^* values predicted pathological differentiation grade was 0.863.There were statistically significant differences in the T2^* and R2^* values between the high and the low Ki-67 expression group (P < 0.05). The T2^* value of the high expression group was higher than that of the low expression group [(55.57 ± 8.77) ms vs. (49.23 ± 6.09) ms], and the R2^* value was lower [18.48 (16.45, 22.05) s^-1 vs. 20.87 (19.56, 22.03) s^-1]. The AUCs of the T2^* and R2^* values for predicting the Ki-67 expression level were 0.727 and 0.662, respectively. The sensitivity and specificity of T2^* were 71.4% and 64.7%, respectively, while those of R2^* were 64.3% and 70.6%, respectively. The combined diagnostic AUC of T2^* and R2^* values predicted Ki-67 expression level was 0.752. Compared with individual parameters, the combined use of T2^* and R2^* values showed no statistically significant difference in predictive efficacy for both pathological differentiation and Ki-67 expression level in PDAC. ConclusionsThe T2^* and R2^* values have good predictive value for the pathological differentiation degree of PDAC and the expression level of Ki-67 among the quantitative parameters of the mDixon-Quant sequence; except for the water phase value, each quantitative parameter can effectively distinguish between the PDAC lesion and the normal pancreatic area. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.006 ObjectiveTo explore the value of diffusion kurtosis imaging (DKI) quantitative parameters in predicting the grading of hypoxia inducible factor-1α (HIF-1α) in pancreatic ductal adenocarcinoma (PDAC). Materials and MethodsA retrospective analysis was conducted on the data of 61 PDAC patients who underwent preoperative 1.5 T MRI examination and were confirmed by surgical pathology. According to the postoperative pathological immunohistochemical score, the patients were divided into a HIF-1α low expression group (32 cases) and a HIF-1α high expression group (29 cases). Two radiologists measured the diffusion weighted imaging (DWI) quantitative parameters, including apparent diffusion coefficient (ADC), mean diffusion coefficient (MD), and mean kurtosis (MK), for two groups of lesions. Independent sample t-test Mann Whitney U test or chi square test were used to analyze the differences in clinical pathological data between two groups, and intra class correlation coefficient (ICC) was used to test the consistency of the measurements of each parameter value by two radiologists. Evaluate the discriminative power of statistically significant parameters through receiver operating characteristic (ROC) curves, DeLong test, Net reclassification improvement and integrated discrimination improvement indicators. ResultsThe low tumor differentiation in the high HIF-1α expression group was significantly higher than that in the low HIF-1α expression group (P = 0.031). The quantitative MRI data measured by two radiologists showed good consistency (ICC values > 0.75). The MD value of the HIF-1α high expression group [(1.17 ± 0.26) × 10^-3 mm²/s] was lower than that of the HIF-1α low expression group [(1.52 ± 0.39) × 10^-3 mm²/s], the MK value of the HIF-1α high expression group (0.72 ± 0.11) was higher than that of the HIF-1α low expression group (0.61 ± 0.11), the difference between the two groups was statistically significant (all P < 0.001). No statistically significant difference in ADC values between the two groups. The AUC, sensitivity, and specificity of MD value, MK value, MD value + MK value, and MD value + MK value + differentiation degree in predicting high and low HIF-1α expression are 0.751, 74.7%, 64.8%; 0.814, 84.4%, 72.4%; 0.862, 82.8%, 78.7%; 0.872, 78.1%, 86.2%. The DeLong test showed that the predictive power of MD value + MK value + differentiation degree was only statistically different from MD value (P = 0.037), no statistically significant difference from other parameters (P > 0.05). The NRI and IDI results showed that MD value + MK value + differentiation degree significantly improved the predictive ability of HIF-1α, and was superior to MD value, MK value and MD value + MK value (P < 0.05). ConclusionsThe combination of MD value, MK value and tumor differentiation degree can help predict different HIF-1α expression in PDAC tumors, providing a basis for preoperative risk stratification and personalized treatment. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.007 ObjectiveTo evaluate the ability of magnetization transfer imaging (MTI) in assessing the fibrosis of pancreatic ductal adenocarcinoma (PDAC). Materials and MethodsThe clinical, imaging and pathological data of 53 patients with PDAC were analyzed retrospectively. All patients underwent MTI examination before surgery and the MTR values of the tumors were measured. Masson staining was used to evaluate tumor fibrosis. The percentage of tumor fibrous area were calculated by ImageJ software. Based on the degree of fibrosis, all patients were divided into two groups: high and low fibrosis. Independent sample t test and one-way ANOVA were used to compare the differences of the magnetic transfer rate (MTR) values and general characteristics between high- and low-grade fibrosis groups. Spearman correlation analysis was used to evaluate the correlation between MTR value and fibrosis degree of PDAC. Receiver operating characteristic (ROC) curve was used to evaluate the diagnostic efficiency of MTR value for PDAC fibrosis classification. ResultsThe MTR value of the low-grade fibrosis group was (0.158 ± 0.053), and that of the high-grade fibrosis group was (0.230 ± 0.063). There was a significant difference of MTR values between the two groups (t = -4.528, P < 0.001), while other general characteristics were not statistically significant (P > 0.05). ROC curve analysis showed that the AUC value of the MTR in evaluating the fibrosis of PDAC patients was 0.822. When a threshold was 0.496, the sensitivity and the specificity were 87.5% and 62.1%, respectively. ConclusionsAs a non-invasive imaging index, MTR value has a potential application value in the evaluation of PDAC fibrosis. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.008 ObjectiveTo observe the abnormal characteristics of generalized anxiety disorder (GAD) through resting-state functional connectivity (rs-FC) and dynamic functional connectivity (dFC) methods. Materials and MethodsA total of 35 patients with GAD and 35 healthy controls (HC) were prospectively recruited. Functional magnetic resonance imaging (fMRI) data were collected for both groups, using the amygdala as a seed point to observe the abnormalities in rs-FC and dFC of the bilateral amygdala. Additionally, a correlation analysis was conducted between the differential brain regions and clinical symptoms. ResultsIn terms of rs-FC, the GAD group exhibited reduced rs-FC between the left amygdala and the left orbitofrontal cortex, left postcentral gyrus, left superior occipital gyrus, and left middle temporal (t = -2.236, -2.220, -2.222, -2.230, P < 0.005) gyrus compared to the HC group. The right amygdala also demonstrated decreased rs-FC with the left middle frontal gyrus, left supplementary motor area, and left postcentral gyrus (t = -2.236, -2.220, -2.222, -2.230, P < 0.005) in the GAD group. Regarding dFC, the GAD group showed reduced dFC between the left amygdala and the right middle temporal gyrus (t = -2.236, -2.220, P < 0.005), as well as the left precuneus compared to HCs. The right amygdala^,s dFC with the right orbitofrontal cortex and right insula (t = -2.297, -2.296, P < 0.005) was also lower in the GAD group. Furthermore, there was a negative correlation between the rs-FC value of the right amygdala and the left middle frontal gyrus and the Hamilton Depression Scale 17 (HAMD-17) score in the GAD group (r = -0.425, P = 0.013). ConclusionsPatients with GAD exhibit abnormal static and dynamic functional connectivity of the amygdala with local brain regions, primarily involving the reward network, default mode network, frontoparietal network, and visual network. This study provides valuable insights into the neurobiological mechanisms underlying the abnormal brain functional activity in GAD. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.009 ObjectiveTo predict attention deficit symptoms in attention deficit hyperactivity disorder (ADHD) based on T1-weighted MRI (T1W-MRI) and to explore brain regions and brain network connections that are significantly associated with the symptoms. Materials and MethodsThe subjects of this experiment included 21 groups of repeated-measurement healthy individuals from Vanderbilt University and 38 patients with combined type of ADHD from Peking University. After obtaining the brain T1W-MRI of each subject, the images were preprocessed to obtain standardized data. The voxel-level cortical thickness morphological features and corresponding radiomics features were extracted, and the reliability of the features was evaluated using the intra-class correlation coefficient (ICC). The nodal features of an individualized brain morphological network were constructed based on the radiomics sorted in descending order by the mean ICC of the whole brain and the Desikan-Killiany (DK) brain atlas, and the individualized brain morphological connections were characterized by feature distance similarity. The support vector regression (SVR) model was used to predict attention deficit symptoms, and the model performance was evaluated by leave-one-out cross-validation. ResultsThe correlation between attention deficit symptoms and the predicted values was r = 0.44 (P = 0.01). The predictive model showed that the brain region-related connections were centered on the right lateral occipital cortex and the right temporal transverse cortex. The findings of these significant brain regions and network connections support the hypothesis of abnormal prefrontal cortex function and default mode network abnormality in ADHD. ConclusionsIndividualized brain morphology networks based on cortical thickness radiomics features can effectively characterize the topological structure of the brain and have the potential to become a new imaging marker for assessing attention deficit. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.010 ObjectiveTo explore the microstructural characteristics of white matter fiber tracts in children with attention deficit hyperactivity disorder (ADHD) using diffusion kurtosis imaging (DKI) technology. Materials and MethodsThis study prospectively analysed 29 children with ADHD [mean age: (8.31 ± 1.25) years], newly diagnosed at our institution, and 27 healthy control (HC) children [mean age: (8.85 ± 1.21) years]. All participants underwent conventional magnetic resonance imaging (MRI) and DKI scans. Children in the ADHD group were additionally assessed using the Swanson, Nolan, and Pelham Ⅳ (SNAP-Ⅳ) scale. Tract-based spatial statistics (TBSS) were employed to analyze white matter tract alterations in ADHD. Post-processing yielded diffusion tensor and diffusion kurtosis parameters, which were compared between groups. Significant parameters were correlated with SNAP-Ⅳ scores. ResultsCompared to HC, ADHD children exhibited significantly reduced fractional anisotropy (FA) values (P < 0.05, corrected for multiple comparisons) in the splenium, genu, and body of the corpus callosum; bilateral posterior thalamic radiations (including optic radiations); right cingulum, right inferior longitudinal fasciculus (ILF), right inferior fronto-occipital fasciculus (IFOF), right posterior corona radiata, right external capsule, and right posterior limb of the internal capsule; left anterior/superior corona radiata, left anterior/posterior limbs of the internal capsule; and fornix. No significant differences were observed in other parameters (P > 0.05). Correlation analysis revealed negative associations between FA values in the splenium of the corpus callosum (r = -0.390, P = 0.018) and the right ILF/IFOF (r = -0.374, P = 0.023) with hyperactivity-impulsivity scores on the SNAP-Ⅳ. ConclusionsDKI reveals microstructural abnormalities in the brains of children with ADHD. Reduced FA in the splenium of the corpus callosum and right ILF and IFOF correlates with hyperactivity-impulsivity symptoms. DKI may provide novel insights into white matter abnormalities in ADHD. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.011 ObjectiveTo investigate the mediating roles and relative contributions of brain structure and function in the effects of aerobic exercise (jogging) on clinical symptoms in parents who lost their only child (PLOCs). Materials and MethodsThis prospective study prospectively collected the imaging data and general information of PLOCs registered at Yixing People^,s Hospital from March 2021 to July 2022. The study included 22 parents who engaged in jogging after the loss of their child, 40 who did not engage in jogging, and 40 healthy controls. All participants underwent MRI and completed the Clinician-Administered PTSD Scale (CAPS) along with other clinical symptom assessments. Mediation analysis was used to examine the mediating role of changes in brain structure and function in the improvement of clinical symptoms due to jogging, and to compare their relative contributions. ResultsCompared to the non-jogging group, the jogging group showed lower symptom scores in C (avoidance and numbing) and C7 (loss of future plans) (P = 0.001; P = 0.002). Reductions were observed in the gray matter volume (GMV) in the left globus pallidus, caudate, and right thalamus nucleus (P < 0.05, FDR corrected). The regional homogeneity (ReHo) values in the left inferior temporal gyrus, right supramarginal gyrus, and right precuneus were lower (P < 0.05, FDR corrected). Mediation analysis indicated that changes in right thalamus GMV and left temporal pole ReHo values both partially mediated the relationship between jogging and C7 symptoms [β = -0.270, 95% confidence interval (CI) was -0.667 to -0.007; β = -0.520, 95% CI was -1.339 to -0.057], with the latter showing a more substantial contribution (17.42% vs. 33.55%). ConclusionsAerobic exercise (jogging) may help alleviate clinical symptoms in PLOCs, and this effect may be mediated by changes in brain structure and function, with brain functional changes playing a more crucial role. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.012 ObjectiveFor the first time, this study combines the fractional amplitude of low-frequency fluctuation (fALFF) method with voxel-based morphometry (VBM) technique to systematically investigate the brain structural-functional coupling characteristics in end-stage renal disease (ESRD) patients and their associative mechanisms with cognitive dysfunction. Materials and MethodsProspectively, 57 ESRD patients and 45 healthy control were recruited. Both groups underwent cranial 3D-T1 structural imaging, resting-state functional magnetic resonance imaging (rs-fMRI) scanning, and cognitive function assessment tests [including Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Trail Making Test A (TMT-A)]. The fALFF maps and gray matter volume (GMV) maps of the two groups were obtained. By calculating the ratio of fALFF to GMV for each voxel, the structure-function coupling (fALFF/GMV) maps were generated. The differences between the two groups were compared, and a Pearson correlation analysis was conducted between the fALFF/GMV values of the brain regions with significant differences and the cognitive scores. ResultsCompared with the healthy control group, in ESRD patients, the fALFF/GMV values increased in the bilateral hippocampi, putamina, middle temporal gyri, cerebellar Cere8 regions, as well as the right amygdala, olfactory cortex, parahippocampal gyrus, left lenticular pallidum, fusiform gyrus, and cerebellar Cere7b region. The fALFF/GMV values decreased in the bilateral medial superior frontal gyri and inferior parietal lobules (P < 0.001, corrected by FDR). There was a significant negative correlation between the total score of the MMSE and the fALFF/GMV values in the left putamen and left lenticular pallidum. A significant negative correlation was observed between the total score of MoCA and the fALFF/GMV values in the left putamina. There was a significant positive correlation between the TMT-A and the fALFF/GMV values in the bilateral medial superior frontal gyri (P < 0.05, corrected by FDR). ConclusionsESRD patients exhibit a significant phenomenon of structural-functional decoupling in multiple relevant brain regions within the default network and the executive control network, which is closely associated with the degree of cognitive impairment in these patients. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.013 ObjectiveTo develop a predictive model based on hippocampal MRI radiomics to assess whether the level of cognitive function in type 2 diabetes mellitus (T2DM) patients belongs to the category of cognitively normal (CN), mildly cognitively impaired (MCI) or dementia (Dem). Materials and MethodsClinical data and MRI imaging data of 140 T2DM patients were retrospectively collected, and they were classified into CN group, MCI group and Dem group according to the Montreal Cognitive Assessment Beijing version (MoCA-B) score, and randomly assigned to the training set (n = 98) and the test set (n = 42) according to the ratio of 7∶3 in order to validate the performance of the model. The right and left hippocampus regions of interests (ROIs) were outlined using the uAI Research Portal (uRP), the radiomics features were extracted, and the MRI radiomics features were used to construct a machine learning (ML) model using twelve classifiers, and the confusion matrix to evaluate the classification model performance. The optimal cutoffs and tuning parameters are explored in the training data, and the models are further evaluated in the experimental data. the optimal algorithm is determined by comparing the area under the curve (AUC) of each classifier. ResultsFrom the original 2313 omics features of hippocampal MRI, ten key features were selected using the K-best selection method. Subsequently, the SelectKBest algorithm was applied to identify two optimal features. When twelve classifiers were employed for training in the CN, MCI, and Dem groups, the quadratic discriminant analysis (QDA) algorithm demonstrated the best performance among the classifiers. The AUC values for each group in the training set were 0.869, 0.854, and 0.893, respectively, while the AUCs for each group in the validation set were 0.819, 0.779, and 0.811, respectively. ConclusionsThe MRI-based QDA model demonstrates significant potential in predicting cognitive dysfunction among patients with T2DM. When compared to various algorithms within CN group, MCI group, and Dem group, the QDA algorithm exhibits superior performance. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.014 ObjectiveTo investigate the relationship between brain glymphatic function and sleep status in insomnia disorder (ID) patients, and to investigate whether repetitive transcranial magnetic stimulation (rTMS) can improve sleep status and brain glymphoid function in patients with ID. Materials and MethodsPatients were divided into a real rTMS treatment group (n = 28), a sham rTMS treatment group (n = 9), and a healthy control group matched for age and gender (n = 20). All subjects underwent neuropsychological and sleep scale assessments, polysomnography (PSG), and MRI. Diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) was used to evaluate glymphatic function. Continuous variables were compared among three groups using one-way ANOVA or Kruskal-Wallis test, with post-hoc analysis conducted using Tukey or Dunn^,s multiple comparison test. Categorical variables were analyzed using the χ² test. The intraclass correlation coefficient was used to assess the consistency of DTI-ALPS index measurements among different observers. Univariate and multivariate linear regression analyses were used to evaluate the correlations between DTI-ALPS and clinical data, sleep assessment scales, and PSG indicators. Paired sample t-tests were used to assess changes in sleep status and DTI-ALPS indices before and after treatment in the real/sham rTMS groups. ResultsThe DTI-ALPS index was lower in all ID patients than in healthy controls before rTMS treatment (F = 13.08, P < 0.001). N2 sleep duration (β = 0.01, 95% CI: 0.01 to 0.01, P = 0.034; β = 0.01, 95% CI: 0.01 to 0.01, P = 0.034) and arousal index (β = -0.01, 95% CI: -0.02 to -0.01, P < 0.001; β = -0.01, 95% CI: -0.01 to -0.01, P = 0.002) were independently associated with the DTI-ALPS index in both the minimally adjusted model Ⅰ and the strictly adjusted model Ⅱ in all ID patients. After rTMS, the sleep questionnaire results, total sleep time, N2 sleep time, arousal index, and DTI-ALPS index in the true stimulation group compared to the sham stimulation group reflected that rTMS treatment improved sleep status as well as brain glymphatic function in ID patients. ConclusionsN2 sleep time and arousal index were independently correlated with brain glymphatic function, and rTMS treatment could improve brain glymphatic function and sleep status of insomnia patients. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.015 ObjectiveTo explore the brain regions and their neurobiomarkers that are vulnerable to brain activation and brain volume in patients with social isolation (SI). This study has been PROSPERO (https://www.crd.york.ac.uk/prospero) web site registration, registration code for CRD42024628028. Materials and MethodsChinese literature was searched using the Chinese search terms "social isolation" "brain" and "magnetic resonance imaging" in Chinese databases including China National Knowledge Infrastructure, Wanfang, VIP, and the Chinese Biomedical Literature Database. English literature was searched using the English search terms "social isolation" "brain," and "magnetic resonance imaging" in English databases including PubMed, Web of Science, Cochrane Library, and Embase. Synonyms for both Chinese and English search terms were expanded, and the search period spanned from the establishment of the databases to September 2024. Literature was included according to strict inclusion and exclusion criteria. Activation likelihood estimation (ALE) was employed, and the single dataset module of Ginger ALE software was utilized to conduct an integrated analysis of the abnormal brain regions in SI patients compared to healthy controls (HCs) in previous studies. ResultsA total of 10 studies from 10 articles were included (488 SI patients and 476 HCs), encompassing task-based functional magnetic resonance imaging (task-based fMRI) and structural magnetic resonance imaging (sMRI) studies. The meta-analysis results of task-based fMRI studies revealed that SI patients exhibited increased brain activation in the left parahippocampal gyrus and left amygdala compared to HCs (voxel volume 1640 mm³, P < 0.001), with no regions showing decreased brain activation. The meta-analysis results of sMRI studies indicated that SI patients had increased brain volume in the right lentiform nucleus, right lateral globus pallidus, and right fusiform gyrus compared to HCs (voxel volume 912, 840 mm³, P < 0.001), and decreased brain volume in the left fusiform gyrus, right inferior frontal gyrus, and right middle occipital gyrus (voxel volume 480, 448, 448 mm³, P < 0.001). ConclusionsThis study, through ALE Meta-analysis, identified that the brain regions susceptible to activation in SI patients are the left parahippocampal gyrus and the left amygdala. The brain regions susceptible to volume changes include the left fusiform gyrus, the right inferior frontal gyrus, the right middle occipital gyrus, the right lentiform nucleus, the right lateral globus pallidus, and the right fusiform gyrus. These findings contribute to a deeper understanding of the neural mechanisms underlying the impact of social isolation on the brain and provide a theoretical basis for preventing the negative effects of social isolation. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.016 ObjectiveTo explore the resting state functional magnetic resonance imaging (rs-fMRI) characteristics of patients with depression after unilateral basal ganglia stroke, and clarify the application value of rs-fMRI in the clinical diagnosis and treatment of patients with depression after unilateral basal ganglia stroke. This study is registered at the Chinese Clinical Trial Registry (No. ChICTR1800016263). Materials and MethodsForty-five stroke patients were included, and based on the results of the 17 item Hamilton Depression Scale (HAMD-17), 23 subjects were included in the post-stroke depression (PSD) group and 22 subjects were included in the non post-stroke depression (NPSD) group. Regional homogeneity (ReHo) and amplitude of low frequency fluctuations (ALFF) analysis methods were used to compare the brain regions with significant differences in ReHo and ALFF values between the two groups of subjects. ResultsCompared with the PSD group, the NPSD group showed an increase in ReHo values in the left temporal polar gyrus and left anterior cingulate gyrus (t = 5.442 8, 3.507 9; P < 0.05), a decrease in ReHo values in the left fusiform gyrus and left anterior cingulate gyrus (t = -3.552 8, -4.112 4; P < 0.05), and an increase in ALFF values in the right fusiform gyrus, right inferior temporal gyrus, left posterior cingulate gyrus, right cingulate gyrus, right superior marginal gyrus, left anterior central lobule, and right anterior cingulate gyrus (t = 3.514 9, 3.277 5, 4.610 2, 3.734 3, 4.218 9, 3.854 2, 4.342 9, 3.964 4; P < 0.05). The difference between the groups was statistically significant (P < 0.05, FWE correction). ConclusionsThe default mode network, sensory motor network, and social network of patients with unilateral basal ganglia depression show ReHo and ALFF changes in some brain regions. These imaging features may be key to the occurrence and development of the disease, clarifying the application value of rs-fMRI in the clinical diagnosis and treatment of patients with unilateral basal ganglia depression after stroke. This will help in the early diagnosis and intervention treatment of PSD. This will provide reference for exploring the brain function research and disease prognosis related to the development mechanism of PSD. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.017 ObjectiveTo explore the role of arterial spin labeling (ASL) imaging in the focal assessment of febrile seizures (FS). Materials and MethodsThe clinical and imaging data of FS children admitted to pediatric Neurology Department of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology from January 2022 to July 2024 were retrospectively analyzed. Two senior radiologists performed blind reading of ASL cerebral blood flow images (ASL-CBF) and gave visual scores. ResultsA total of 24 eligible FS patients were included in this study, of which 3 patients showed no perfusion abnormality in ASL of the whole brain (1 point, 12.5%), and 21 patients (87.5%) showed abnormal perfusion, all with reduced perfusion, including 20 patients (83.3%) involved in the temporal region. Electroencephalogram (EEG) was performed in 22 of the 24 patients within 24 hours of FS onset, and there was no significant difference in ASL-CBF visual scores between normal and abnormal EEG groups (P = 0.698). According to the clinical symptomatology and EEG results, 6 of the 24 patients were classified as simple FS (25.0%), 14 as complex FS (58.3%), and 4 as persistent FS (16.7%); Among them, 4 patients with FS persistence showed bilateral asymmetry reduction (ASL-CBF score was 3 points). There were significant differences between the three groups in the time from FS onset to ASL collection, whether there was recurrence within 24 hours, and ASL-CBF visual score (P < 0.05). Post hoc analysis indicated that although there was no significant difference in ASL-CBF visual score between complex FS and FS persistence group (χ² = 5.143, P = 0.162), the ASL-CBF visual score between simple FS and FS persistent state group showed significant difference (χ² = 10.000, P = 0.019). Compared with simple FS, there were no significant differences in gender, age, time from FS onset to ASL collection, previous FS history, familial FS history, and EEG in patients with complex FS group (P > 0.05), except for whether there was recurrence within 24 hours (χ² = 0.008, P = 0.008), which was basically consistent with clinical practice. ConclusionsThe underlying lesion of febrile seizures may be located in the temporal region, and the continuous state of FS showed asymmetric reduction of bilateral perfusion. ASL techniques are helpful in evaluating focal problems in FS, especially in those cases that had normal EEG and negative structural MRI. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.018 ObjectiveTo explore the characteristics of cerebral blood flow (CBF) in children with global developmental delay (GDD) using three-dimensional pseudo-continuous arterial spin labeling (3D-pCASL) imaging technology and further investigate the correlation between the Gesell Developmental Scale and CBF. Additionally, the diagnostic efficacy of CBF values in identifying GDD was assessed. Materials and MethodsA total of 45 children diagnosed with GDD and 42 healthy control (HC) children were included in this study. All participants underwent magnetic resonance imaging of the brain, including 3D-pCASL scanning. Quantitative CBF pseudocolor maps were obtained through post-processing, and differences in CBF values between the two groups were analyzed. Perform correlation analysis between the CBF values with statistically significant intergroup differences and the Gesell Developmental Scale scores in children with GDD , and the diagnostic efficacy of CBF values in identifying GDD children was evaluated using receiver operating characteristic (ROC) curves. ResultsCompared with the control group, CBF values in the bilateral temporal lobe white matter, left temporal lobe gray matter, and bilateral frontal lobe white matter were reduced (P < 0.05). Correlation analysis revealed that CBF values in the right and left temporal lobe white matter were positively correlated with the language developmental quotient (r = 0.477, P = 0.001; r = 0.513, P < 0.001), CBF values in the right frontal lobe white matter were positively correlated with the personal-social developmental quotient (r = 0.347, P = 0.019), and CBF values in the left frontal lobe white matter were positively correlated with the fine motor developmental quotient (r = 0.577, P < 0.001). CBF values in multiple brain regions demonstrated good diagnostic efficacy in distinguishing GDD children (AUC > 0.7). Conclusions3D-pCASL technology can non-invasively assess abnormalities in cerebral blood flow in children with GDD. The study found that reduced CBF values may be associated with impairments in multiple developmental abilities in children with GDD, and CBF values have good diagnostic value in identifying GDD. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.019 ObjectiveTo characterize the heterogeneity of adult-type diffuse gliomas using local habitat entropy based on multimodal magnetic resonance imaging and to develop and validate a comprehensive model for predicting isocitrate dehydrogenase molecular status. Materials and MethodsA retrospective collection and analysis were performed on data obtained from the Affiliated Hospital of Inner Mongolia Medical University, the University of California, San Francisco and The Cancer Genome Atlas, encompassing a total of 533 subjects. Six types of conventional magnetic resonance images (T2, T1, T2-FLAIR, T1-CE, DWI, and ADC) were used for further image preprocessing. The preprocessing pipeline included N4 bias field correction, super-resolution reconstruction based on a migration model, isotropic resampling, and image normalization. An improved nn-Unet was employed to automatically segment tumor regions, followed by manual confirmation and correction. Habitat local entropy values were obtained for the entire lesion area, using a 3 × 3 × 3 matrix considering the size of the region of interest. During this process, global image discretization was performed according to the entire cohort, meaning the discretization histogram was based on the actual maximum and minimum values of the cohort, and finally adjusted to 32 Bins with equal interval Bin width. K-means was used to generate habitats based on T1-CE and T2-FLAIR matching, with the number of cluster centers ranging from 2 to 5. Then, 16 first-order features of different habitat subregions were extracted from all modalities. The UCSF public database served as the training set, and internal validation was performed using 10-fold cross-validation. The remaining two databases were used as independent test sets. A multi-pipeline approach (240 basic pipelines) was used to construct machine learning models. Feature selection and hyperparameter tuning were performed through cross-validation. The diagnostic performance of the models was evaluated using receiver operating characteristic (ROC) curves, and the DeLong test was used to compare model differences. The deviation between the model predictions and actual results was visualized using calibration curves. Decision curve analysis was employed to determine the clinical net benefit. ResultsWhen the number of cluster centers was set to 2 or 3, the corresponding Calinski-Harabasz indices were 95 080 and 100 379, respectively, the Silhouette coefficients were 0.477 and 0.422, and the Davies-Bouldin indices were 0.741 and 0.810. Since the results for cluster centers of 4 and 5 were suboptimal, subsequent analyses were conducted only for clusters 2 and 3. All three multimodal models (whole lesion area, cluster 2, and cluster 3) demonstrated excellent diagnostic performance, with AUC values ranging from 0.942 to 0.974 in the training set and from 0.739 to 0.864 in the test sets. Specifically, when the cluster number was 2, the sensitivity was higher in both independent test sets (95.2% and 80.0%, respectively). Conversely, when the cluster number was 3, the specificity was higher in both independent test sets (72.2% and 89.2%, respectively). The calibration curves and decision analysis curves for all three models indicated high and similar predictive consistency and clinical applicability. ConclusionsLocal habitat entropy based on multimodal MRI provides valuable information on the heterogeneity of adult-type diffuse gliomas. The combined application of local features and habitat analysis offers new insights and methods for the non-invasive assessment of various pathological abnormalities. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.020 ObjectiveTo investigate the diagnostic value of radiomics models based on multimodal MRI in differentiating autoimmune encephalitis (AE) from herpes simplex virus encephalitis (HSE). Materials and MethodsA retrospective collection was conducted for patients with acute or subacute autoimmune encephalitis (AE) and herpes simplex encephaliti (HSE) confirmed by cerebrospinal fluid or serological tests at Huashan Hospital Affiliated to Fudan University between January 2013 and July 2024. Patients were randomly divided into training and independent test sets at a ratio of 8∶2. T2-fluid attenuated inversion recovery (T2-FLAIR), T1-weighted imaging (T1WI), and diffusion weighted imaging (DWI) data were collected. All T2-FLAIR hyperintense lesions were manually delineated. Pyradiomics was employed to extract radiomic features, followed by feature selection using the least absolute shrinkage and selection operator (LASSO) algorithm and correlation analysis. The random forest (RF), support vector machine (SVM) and K-nearest neighbor (KNN) models were established; the model parameters were optimized via five-fold cross-validation, and the models were validated on the independent test set. Diagnostic performance was evaluated by AUC, sensitivity, specificity, and accuracy of ROC curves. ResultsThe study totally included 117 AE cases and 110 HSE cases. There were 182 patients including 810 lesions in the training set and 45 patients including 215 lesions in the test set, there were respectively 22, 10, 15, and 12 features being selected for the multimodal, T2-FLAIR, DWI, and T1WI models. The AUCs of RF models based on multimodal, T2-FLAIR, DWI, and T1WI were 0.884, 0.841, 0.775, and 0.799 respectively in the training set. The corresponding AUCs were 0.805, 0.809, 0.696, and 0.737 in the test set, with accuracies of 74.9%, 73.5%, 67.0%, and 67.4% respectively. The AUCs of SVM models based on multimodal, T2-FLAIR, DWI, and T1WI were 0.831, 0.820, 0.780 and 0.816 respectively in the training set. The corresponding AUCs were 0.792, 0.807, 0.696 and 0.728 in the test set, with accuracies of 74.9%, 76.7%, 68.8% and 68.8% respectively. The AUCs of KNN models based on multimodal, T2-FLAIR, DWI, and T1WI were 0.850, 0.806, 0.760 and 0.766 respectively in the training set. The corresponding AUCs were 0.805, 0.809, 0.712 and 0.734 in the test set, with accuracies of 74.0%, 73.0%, 67.9% and 71.2% respectively. The multimodal and T2-FLAIR-based RF, SVM and KNN models exhibited significantly higher AUCs than the DWI-based model (P < 0.05). There were no significant differences in the AUC values of the RF, SVM, and KNN models based on different MRI modalities in the test set. ConclusionsThe radiomics RF, SVM and KNN models based on multimodal MRI and T2-FLAIR sequence achieved a high diagnostic performance in distinguishing AE from HSE, assistting clinicians making diagnoses in a non-invasive method and helpful for the early formulation of clinical decisions. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.021 ObjectiveTo assess the diagnostic value of unenhanced MRI with diffusion kurtosis imaging (DKI) in differential diagnosis between thyroid benign and malignant nodules. Materials and MethodsA total of 96 consecutive patients, each with a single thyroid nodule, were included in this study. Among these, 39 nodules were histopathologically confirmed as benign, while 57 were malignant. All patients underwent thyroid MRI, which included T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and DKI. Two radiologists independently evaluated the images, measuring the signal intensity ratio (SIR) on T1WI and T2WI, the apparent diffusion coefficient (ADC) from DWI, as well as mean diffusivity (MD) and mean kurtosis (MK) from DKI. Additionally, morphological features of the nodules were assessed. Univariate and multivariate logistic regression analyses were performed to determine the predictive value of these imaging parameters and morphological features for malignancy. To assess the robustness of the logistic regression model, a 5-fold cross-validation approach was applied. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic effectiveness of the continuous variables that were statistically significant in the multivariate analysis. ResultsThe characteristics of thyroid malignant nodules including lower ADC values (P < 0.001), lower MD values (P < 0.001), higher MK values (P < 0.001), younger age (P < 0.001), smaller tumor size (P < 0.001), solid component (P < 0.001), and irregular margins (P < 0.001). Multivariate analysis further revealed that lower MD values (odds ratio = 5.046; P = 0.001), smaller tumor size (odds ratio = 3.817; P = 0.001), and irregular margins (odds ratio = 84.876; P < 0.001) were independent risk factors for thyroid malignant nodules. The combined model yielded an average area under the ROC curve of 0.968 in 5-fold cross-validation, with a sensitivity of 91.4%, specificity of 84.6%, accuracy of 88.7%, and an F1 score of 0.904 at the optimal cutoff value of 0.42. ConclusionsMD values derived from DKI, combined with morphological features can provide imaging diagnostic basis for the preoperative differential diagnosis between thyroid benign and malignant nodules. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.022 ObjectiveTo investigate the value of cardiac magnetic resonance feature tracking (CMR-FT) in evaluating non-dilated left ventricular cardiomyopathy (NDLVC). Materials and MethodsA retrospective analysis was performed on 50 patients with NDLVC who underwent CMR examination in Beijing Anzhen Hospital from January 2022 to November 2024, defined as the NDLVC group, and another 25 patients without cardiovascular disease and no abnormal CMR manifestations in the same period were selected as the control group. Left ventricular 3D global strain parameters of all subjects were obtained by CVI42 post-processing software. The receiver operating characteristic (ROC) curve was utilized to assess the diagnostic efficacy of strain parameters in distinguishing the NDLVC group from the control group. ResultsCompared with the control group, LVESV and LV mass at ED in NDLVC patients increased, and the difference was statistically significant (P < 0.05); While the absolute values of LVSV, LVCO, LVEF, GRS (3D), GCS (3D) and GLS (3D) decreased, and the difference was statistically significant (P < 0.05). ROC curve analysis showed that the critical values for GRS (3D), GCS (3D) and GLS (3D) identification of the two groups were 35.34%, -16.88% and -10.55%, and the sensitivity was 72.00%, 46.00% and 58.00%, respectively. The specificity was 72.00%, 96.00%, 76.00%, and AUC was 0.778, 0.713, 0.706, respectively. ConclusionsCMR-FT technique has certain diagnostic value for NDLVC patients, GRS (3D) sensitivity is higher, GCS (3D) specificity is higher. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.023 ObjectiveTo evaluate the influence of left ventricular myocardial fibrosis on left and right ventricular function in patients with coronary artery disease (CAD) based on cardiac magnetic resonance feature tracking (CMR-FT) technique, and to explore the biventricular interaction. Materials and MethodsA total of 50 patients diagnosed with CAD and 32 healthy subjects from September 2023 to March 2025 were retrospectively analyzed. The left and right ventricle conventional cardiac function parameters, global strain parameters, including circumferential peak strain (PCS), radial peak strain (PRS) and longitudinal peak strain (PLS), and left ventricular late gadolinium enhancement percentage (%LGE) were measured after post-processing. According to the median value of %LGE, CAD patients were divided into %LGE ≥ 6% group and %LGE < 6% group. One-way analysis of variance, non-parametric test, Chi-square test and other statistical methods were used to compare the differences in clinical data and CMR parameters between the control group and different patient groups. Spearman or Pearson correlation coefficient was used to analyze the correlation between right ventricular function and left ventricular function as well as myocardial fibrosis, and multiple linear regression analysis was further performed. ResultsCompared with the %LGE < 6% group, left ventricle global PCS, PRS, PLS, ejection fraction (EF) and right ventricle global PLS were decreased, while left ventricular end-diastolic volume (EDV), end-diastolic volume index (EDVi), end-systolic volume (ESV), end-systolic volume index (ESVi), left ventricular mass (LVM), left ventricular mass index (LVMi) and %LGE were increased in the %LGE ≥ 6% group (all P < 0.05). Correlation analysis showed that right ventricle global PCS and PLS were positively correlated with left ventricular global PCS, PRS, PLS and EF, and negatively correlated with left ventricular EDVi, ESVi and %LGE (r = -0.762 to 0.731, all P < 0.05). Linear regression analysis showed that right ventricular global PCS was independently and positively correlated with left ventricular global PLS (β = 0.356, P = 0.011), and independently and negatively correlated with left ventricular ESVi (β = -0.362, P = 0.010); Right ventricular global PLS was independently and positively correlated with left ventricular global PRS (β = 0.291, P = 0.022), and independently and negatively correlated with %LGE (β = -0.344, P = 0.003). ConclusionsEven though there are no significant changes in the conventional cardiac function parameters of the right ventricle, the left and right ventricular strains in CAD patients are impaired after left ventricular myocardial fibrosis. In addition, the right ventricular strain is independently and positively correlated with the left ventricular strain, and independently and negatively correlated with the degree of left ventricular myocardial fibrosis. CMR-FT can provide earlier and more accurate information for ventricular function impairment in CAD patients, and provide a new basis for early treatment. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.024 ObjectiveTo explore the differences in diagnostic performance of varying peritumoral region (PTR) extents for breast non-mass enhancement (NME) lesions by utilizing the advantages of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) combined with intratumoral region (ITR) and PTR radiomics. Materials and MethodsData of 168 patients from September 2021 to September 2024 were included in this retrospective study. They were randomly divided into training set (n = 117) and validation set (n=51) according to 7∶3. Based on DCE-MRI images, ITK-SNAP software was used to manually outline the ITR of the lesion and automatically expand the PTR. The least absolute shrinkage and selection operator (LASSO) was used to extract radiomics features in the tumor and in the extended area of 3 mm, 4 mm and 5 mm around the tumor. LASSO was used to select features and construct imaging models for ITR, PTR 3 mm, PTR 4 mm and PTR 5 mm. The optimal PTR model and ITR model were combined to form the optimal radiomics model. Clinical characteristics were added and a clinical model was developed using multivariate logistic regression analysis. Finally, the clinical model, the optimal radiomics model, and the combined model incorporating clinical features and optimal radiomics features were evaluated. The calibration curve and decision curve analysis (DCA) were used to evaluate the performance of the model, and Shapley additive explanations (SHAP) diagram was used to explain the performance of the model. ResultsThe ITR-PTR 4 mm radiomics model was found to have the best area under the curve (AUC) (training set: 0.822, validation set: 0.782) for constructing the combined model. In the clinical model, only the type of time signal intensity curve (TIC) was found to be significantly positively correlated with benign and malignant lesions by multivariate analysis (r = 0.681, P < 0.001). The AUC of the final combined model in the training set reached 0.912. In the validation set, the AUC was 0.806. DCA curve showed that the combined model had the highest clinical efficacy and was close to the diagonal in the calibration curve, so the fitting effect and generalization ability of the combined model were better. ConclusionsThe study found that the combined model combining radiomics features and clinical features can effectively distinguish benign and malignant NME lesions of undetermined nature in breast MRI, which provides a new reference for clinical diagnosis. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.025 ObjectiveTo explore the value of Shapley additive explanations (SHAP) interpretable machine learning models based on high-resolution enhanced delayed-phase magnetic resonance imaging in preoperatively predicting histologic grade of non-special type invasive breast cancer. Materials and MethodsRetrospectively collected the clinical-pathological-imaging data of 154 patients with invasive breast carcinoma of no special type from January 2019 to December 2023. Based on pathological biopsy results, Grade Ⅰ and Ⅱ were classified into the low-grade group, while Grade Ⅲ was classified into the high-grade group. They were randomly divided into a training group of 107 cases and a validation group of 47 cases in a 7∶3 ratio. 3D Slicer was used to delineate the lesion edges and extract radiomics features. Features were screened through multifactorial analysis. Radiomics feature models were established using Random forest (RF) and logistic regression, while clinical models, radiology models, and a combined radiology-clinical-radiomics feature model were developed using logistic regression. The performance of the model was evaluated using the area under the receiver operating characteristic curve (AUC) and accuracy, while model comparison was conducted using DeLong test. SHAP analysis was used to visualize the contribution and importance of features in the model. ResultsThere were significant differences in progesterone receptor (PR), tumor boundary, Ki-67 and estrogen receptor (ER) between low-grade group and high-grade group (P < 0.05). The AUC of the combined model based on radiology-clinical-radiomics features for preoperative prediction of the histological grade of invasive breast cancer was relatively good, with AUC values of 0.807 (95% CI: 0.723 to 0.891) in the training group and 0.890 (95% CI: 0.795 to 0.984) in the validation group. Among the two independent radiomics feature models, the logistic radiomics model showed no obvious overfitting, with AUC values of 0.750 (95% CI: 0.655 to 0.846) in the training group and 0.801 (95% CI: 0.667 to 0.936) in the validation group. The AUC of the clinical model and the radiology model in the training group were 0.661 (95% CI: 0.551 to 0.771) and 0.600 (95% CI: 0.493 to 0.706), respectively, and in the validation group were 0.789 (95% CI: 0.645 to 0.933) and 0.708 (95% CI: 0.565 to 0.850), respectively. ConclusionsThe joint model showed good efficacy in preoperatively predicting histologic grade of non-special type invasive breast cancer, providing guidance for preoperative treatment of breast cancer patients in clinical practice. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.026 ObjectiveTo explore the value of diagnosing Luminal and non-Luminal breast cancer (BC) based on multiparameter MR habitat imaging analysis. Materials and MethodsA retrospective analysis was conducted on 216 cases of breast diseases treated at Ma^,anshan People^,s Hospital from December 2019 to May 2024. These cases were confirmed by puncture biopsy or surgical pathology, including 147 cases of Luminal-type BC and 69 cases of non-Luminal-type BC. The patients^, ages ranged from 26 to 85 years old, with an average age of (54.8 ± 10.9) years. The 216 patients were randomly divided into a training set and a validation set at a ratio of 7∶3. All patients underwent multi-parameter magnetic resonance imaging (mpMRI) scans. Image preprocessing was performed on the T2WI sequence, small field diffusion weighted imaging (ZOOMit-DWI) sequence in mpMRI, as well as the PEI, TTP, WASHIN, and WASHOUT sequences obtained from the analysis of the dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) sequence. Radiomics features were extracted from each functional parameter map. Cluster analysis was carried out through the Gaussian mixture model (GMM), and the clustering results were evaluated by the Silhouette coefficient. Finally, six groups of habitat images were generated, with each group containing three sub - regions. After image preprocessing, 1197 and 3591 omics features were extracted from the original and habitat sub-region images respectively. Redundant features were removed using robust normalization, Z-score standardization, min-max normalization, F test, the least absolute shrinkage and selection operator (LASSO) algorithm, and 10-fold cross-validation. Four features were selected from the clinical data to construct a clinical model. In the radiomics part, 4, 5, 10, 6, 11, and 9 features were selected, and six groups of radiomics models were established. Then, logistic regression was used to screen the radiomics models and clinical features to establish a combined model. In the habitat radiomics part, 14, 13, 19, 4, 14, and 13 features were selected, and six groups of habitat radiomics models were established. Logistic regression was used to screen the habitat radiomics models, intratumor heterogeneity score (ITH-score), and clinical features to establish a combined mode. The thresholds, sensitivities, specificities, accuracies, negative predictive values, and positive predictive values of the clinical model, radiomics model, habitat radiomics model, and their respective combined models were calculated respectively. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was calculated to analyze the diagnostic efficacy of each model. The DeLong test was used to compare the differences between groups pairwise, and decision curve analysis (DCA) was further used to evaluate the net benefit of the models. ResultsIn the training set and validation set, there were statistically significant differences in ER, PR, Ki-67 and WHO grade between Luminal and non-Luminal BC (P < 0.05). In both the radiomics and habitat radiomics models, the Combine model had the best prediction performance. The AUC values of the Combine model in the training set and validation set were 0.967 and 0.798 respectively, and the optimal model was the MLP model among them. The AUC values of the Combine model in the training set and validation set were 0.969 and 0.910 respectively, and the optimal model was the linear_SVM model among them. Comparatively, the performance of the linear_SVM model was significantly better than that of the MLP model. ConclusionsAnalysis based on multi-parameter MR habitat imaging can diagnose Luminal-type and non-Luminal-type BC relatively accurately, which is helpful for the clinical diagnosis, treatment and management of BC. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.028 Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.029 Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.030 Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.031 Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.032 ^, understanding and diagnosis of CCM and ultimately improve the quality of life and prognosis of CCM patients. ]]> Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.033 Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.034 Tue,20 May 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.05.035 Tue,20 May 2025 00:00:00 GMT