http://med-sci.cn/cgzcx/en/contents_list.asp?issue=202509 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.1674-8034.2025.09.001 ObjectiveBy analyzing neuroimaging features of corpus callosum (CC) subregions in children aged 1 to 12 years with autism spectrum disorder (ASD), this study summarizes structural alterations and dynamic developmental trajectories of CC subregions in ASD children. Materials and MethodsRetrospective data from 153 children diagnosed with ASD at their first visit of the Children^,s Brain Disease Diagnosis and Rehabilitation Center at the First Affiliated Hospital of Henan University of Chinese Medicine between June 2021 and December 2023 were collected. The children were divided into three groups based on their age at visit: the 1 to 3 years group (47 cases, 36 males, 11 females), the 3 to 6 years group (89 cases, 70 males, 19 females), and the 6 to 12 years group (17 cases, 14 males, 3 females). Subsequently, cranial magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI) scans were performed. Using DTI-related post-processing software, the fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were measured in the rostrum, genu, body, and splenium of the CC. The differences in DTI parameters were compared among ASD children by gender, age groups, and CC subregions. ResultsNo significant differences were observed in FA and ADC values between different genders across all age groups (P > 0.05). With increasing age, FA values gradually increased and ADC values gradually decreased in children with ASD. Correlation analysis of DTI parameters in the CC subregions with age revealed that: except for the rostrum of CC (r = -0.064, P = 0.433; r = -0.029, P = 0.727), FA values in the genu, body, and splenium of CC showed positive correlations with age (r = 0.335, P = 0.001; r = 0.350, P = 0.001; r = 0.264, P = 0.001), while ADC values showed negative correlations with age (r = -0.466, P= 0.001; r = -0.458, P = 0.001; r = -0.482, P = 0.001). Age-related differences in DTI parameters: the rostrum parameters of the CC showed no statistically significant differences among age groups (P > 0.05). For the genu, body, and splenium of the CC: statistically significant differences existed between the 1 to 3 years group and the 3 to 6 years group, as well as between the 1 to 3 years group and the 6 to 12 years group (P < 0.05). FA values in all CC subregions, along with ADC values in the genu and splenium, showed no statistically significant differences between the 3 to 6 years group and the 6 to 12 years group (P > 0.05). Analysis of DTI parameters in subregions of the CC in children with ASD revealed that FA values decreased sequentially in the splenium, genu, body, and rostrum of the CC, while ADC values increased sequentially in the genu, splenium, body, and rostrum of the CC. Regarding age-specific differences: Only the 1 to 3 years group showed statistically significant differences (P < 0.05) in FA and ADC values across all CC subregions. In the 3 to 6 years and 6 to 12 years groups, differences in some brain regions were not statistically significant (P > 0.05). ConclusionsDTI can systematically analyze structural alterations and dynamic evolution patterns in subregions of the CC in children with ASD, providing more objective and comprehensive neuroimaging evidence for clinical diagnosis and exploration of ASD pathological mechanisms. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.002 ObjectiveTo investigate the immediate effects of transcutaneous auricular vagus nerve stimulation (taVNS) at different stimulation frequencies on abnormal functional connectivity (FC) and functional network connectivity (FNC) within the triple network model (TNM) in patients with premenstrual syndrome (PMS). Materials and MethodsThis single-center, prospective study enrolled 56 patients with PMS and 67 healthy controls (HC). All participants underwent resting-state functional magnetic resonance imaging (rs-fMRI) and clinical assessments before and after taVNS. Independent component analysis (ICA) was used to extract FC and FNC of key brain networks within TNM, including the default mode network (DMN), executive control network (ECN), and salience network (SN). Based on the TNM network templates, two-sample t-tests were conducted to compare differences in FC and FNC between the PMS and HC groups, and correlation analyses were performed between the altered brain regions and clinical psychological scale scores. Furthermore, paired-sample t-tests were used to evaluate the modulatory effects of taVNS at different stimulation frequencies (2 Hz, 25 Hz, and sham stimulation) on abnormal FC and FNC within the TNM in PMS patients. ResultsCompared to HC, patients with PMS exhibited increased FC within the SN, specifically in the left inferior frontal gyrus (LIFG), insula, and superior temporal gyrus (STG), which positively correlated with emotional, physical, cognitive, and behavioral symptoms (r = 0.377-0.403, P < 0.05, FDR corrected). Altered FNC was also observed, with decreased SN-right ECN (rECN), increased SN-dorsal DMN(dDMN) and dDMN-rECN, the latter negatively associated with physical scores (r = -0.18, P < 0.05, FDR corrected). Following taVNS, both 2 Hz and 25 Hz significantly reduced FC within the SN in PMS patients (P < 0.05, FDR corrected). Specifically, 2 Hz mainly modulated the prefrontal cortex, LIFG, insula, and STG, while 25 Hz predominantly affected the STG. After sham stimulation (st-taVNS), no significant FC changes were observed within the SN in PMS patients. However, 25 Hz-taVNS led to a further decrease in FNC between the SN and rECN (P < 0.05, FDR corrected), whereas no significant changes were observed in the HC group. ConclusionsTaVNS modulates abnormal FC and FNC within TNM-related brain networks in PMS patients. The immediate modulation effects vary by stimulation frequency, indicating the potential need for personalized frequency selection based on individual symptoms. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.003 ObjectiveTo investigate the correlation between changes in cerebral perfusion patterns and serum immunological markers as well as cognitive assessments in patients with neuropsychiatric systemic lupus erythematosus (NPSLE) using three-dimensional arterial spin labeling (3D-ASL) technology, and to explore the pathogenesis of NPSLE from an imaging perspective. Materials and MethodsProspectively collected imaging, clinical serological, and cognitive assessment data from 37 NPSLE patients, 52 non-neuropsychiatric systemic lupus erythematosus (non-NPSLE) patients, and 39 healthy controls (HC) were analyzed. Differences in cerebral perfusion patterns were compared, and correlation analyses were conducted between cerebral blood flow (CBF) values in differentially perfused brain regions of NPSLE patients and serum/cognitive indicators. ResultsCBF differences revealed that the NPSLE group exhibited significantly higher CBF in the left middle temporal gyrus and left supramarginal gyrus compared to the non-NPSLE group (P < 0.001). Compared to the HC group, the NPSLE group showed increased CBF in the aforementioned regions as well as decreased CBF in multiple brain areas, including the right superior frontal gyrus (P < 0.001). Correlation analysis demonstrated that in NPSLE patients, CBF values in the left middle temporal gyrus were positively correlated with hemoglobin (r = 0.392, P = 0.037), red blood cell count (r = 0.437, P = 0.022), and visuospatial/executive function scores on the Montreal Cognitive Assessment (MoCA) (r = 0.358, P = 0.016), while negatively correlated with anxiety scale scores (r = -0.380, P = 0.015). CBF values in the left supramarginal gyrus were positively correlated with hemoglobin (r = 0.612, P = 0.016), hematocrit (r = 0.457, P = 0.016), and complement 3 (r = 0.538, P = 0.008), but negatively correlated with beck anxiety inventory (BAI) scores (r = -0.397, P = 0.040). ConclusionsPatients with NPSLE have abnormal cerebral perfusion patterns. These changes may play a key role in the pathophysiological process of neuropsychiatric symptoms in patients with NPSLE and are involved in the occurrence and development of its pathological mechanism. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.004 ObjectiveTo explore related factors affecting MRI evolution of recent small subcortical infarct (RSSI). Materials and MethodsA total of 105 patients with RSSI were admitted between August 2019 and September 2024. There were 78 males and 27 females, with mean age (62.77 ± 13.28) years. Clinical information of patients, images data of head MRI were collected. All patients were divided into cavitation group and no cavitation group [white matter hyperintensities (WMH) and disappearance] to analyze related factors affecting evolution with multivariate logistic regression analysis. ResultsSixty-three cases (60.00%) developed to cavities, 34 cases (32.38%) evolved into WMH, and 8 cases (7.62%) disappeared. There were significant differences in gender, initial diffusion-weighted imaging (DWI) infarct diameter and number of old lacunes lesion between the two groups (P < 0.05). In logistic regression analysis, initial DWI infarct diameter (OR = 1.394, P < 0.001) and number of old lacunes (OR = 1.455, P = 0.028) was an independent predictor of cavity formation. ConclusionsAbout 60% of RSSI develop to cavitation. All infarct lesions were reduced during follow-up. The RSSI showing larger diameter of infarct lesions and more number of old lacunar infarction have a greater possibility of cavitation. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.005 ObjectiveTo study the clinical application value of magnetic resonance angiography (MRA) combined with CT angiography (CTA) in the diagnosis of cerebral vascular malformations (CVMs) in children. Materials and MethodsA retrospectively analysis was conducted on the imaging data of 50 children with suspected CVMs who met the inclusion and exclusion criteria admitted to our hospital from January 2024 to December 2024. All children underwent digital subtraction angiography (DSA), CTA and MRA imaging examinations. The detection rates of CTA and MRA and their consistency with gold standard DSA were analyzed. The diagnostic efficacy of CTA, MRA and their combination in children with CVMs was analyzed. ResultsA total of 38 children with CVMs were detected by DSA. A total of 34 cases were detected by CTA, which was in good agreement with DSA (Kappa value was 0.606); and 33 cases were detected by MRA, and the consistency with DSA was moderate (Kappa value was 0.472); and 40 cases was detected by the combination of CTA and MRA, which was in good agreement with DSA (Kappa value was 0.767). The sensitivity and accuracy of CTA combined with MRA were higher than those of CTA or MRA alone (P < 0.05). ConclusionsIn the clinical diagnosis of CVMs in children, compared with the diagnoses of CTA or MRA alone, the diagnostic efficiency of MRA combined with CTA is superior. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.006 ObjectiveTo evaluate the value of susceptibility weighted imaging (SWI) in the diagnosis and differential diagnosis of neonatal craniocerebral injury. Materials and MethodsA retrospective collection of 900 neonates suspected of intracranial hemorrhage (ICH) after cranial ultrasound screening in Shanxi Children^,s Hospital was completed with conventional MRI and SWI. The chi-square test was used to compare the detection rates of conventional MRI and SWI for different hemorrhage sites. The chi-square test was used to evaluate the differences in different sites of ICH in preterm and term infants. Results(1) SWI has a higher detection rate of neonatal ICH than conventional MRI (24.0% vs. 19.8%, P < 0.05); (2) For hemorrhage in the lateral ventricles, cerebral cortex, cerebellar hemispheres, and subdural/epidural hemorrhages, SWI detects a greater number and extent of lesions than conventional MRI (P < 0.05); (3) The detection rate of hemorrhage lesions in germinal matrix and lateral compartment was higher in preterm infants than term infants, and the detection rate of hemorrhage lesions in subarachnoid space was lower than in term infants (P < 0.05). ConclusionsSWI is somewhat superior to conventional MRI in detecting neonatal ICH and identifying punctate white matter lesion and microhemorrhagic lesions. Therefore, it is recommended that conventional MRI combined with SWI sequences should be used for comprehensive diagnosis in order to improve the efficacy of detecting ICH lesions and provide an imaging basis for clinical diagnosis and treatment. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.007 ObjectiveTo investigate the feasibility of fluid-suppressed amide proton transfer-weighted imaging (FS-APTw) in suppressing the fluid components of post-treatment gliomas. Materials and MethodsFifty-nine patients with surgically and pathologically confirmed gliomas who received standard postoperative chemoradiotherapy were prospectively collected and underwent APTw and FS-APTw imaging. The APTw and FS-APTw maps were coregistered and fused with anatomical maps to extract APTw and FS-APTw values from fluid components, peri-lesion edema, and measurable enhancing regions of gliomas. Comparisons were made using the paired-samples t-test or Wilcoxon signed ranks test. Subgroup analyses were performed according to the characteristics of the fluid compartments on FLAIR and DWI images (Group 1, low signal on both FLAIR and DWI; Group 2, equal or high signal on FLAIR but low signal on DWI; Group 3, equal or high signal on FLAIR and high or mixed signals on DWI). The efficacy of fluid suppression among the three groups was compared using the Kruskal-Wallis test, followed by the post-hoc tests. ResultsAfter the use of fluid suppression, the FS-APTw values of the fluid and edema compartments were significantly decreased, as compared to their corresponding APTw values (1.37% vs. 1.67%, 1.03% vs. 1.14%) (both P < 0.05). No significant difference was found in enhancing compartments (2.16% vs. 2.22%, P = 0.404). The subgroup analysis showed notable differences in the baseline APTw values of the fluid compartments across the three groups. Specifically, Group 2 and 3 exhibited significantly higher APTw values (3.80% and 1.85%, respectively) compared to Group 1 (1.00%, P < 0.05). After fluid suppression, a marked decrease in FS-APTw values was observed in the fluid compartments of all three groups, as compared to their respective APTw values (P < 0.05 for all). Notably, the most pronounced reduction in APTw value was discerned in Group 2 (-1.23%). ConclusionsFS-APTw can effectively suppress the APTw signals of fluid components in post-treatment gliomas, while maintaining the signals of solid tissues. This fluid suppression effect was particularly pronounced in fluid compartments with protein-rich liquid, shown as equal or high signal intensities on FLAIR and low on DWI. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.008 ObjectiveTo develop a combined model based on multiparametric MRI, radiomics, and deep learning techniques to predict isocitrate dehydrogenase gene (IDH-1) mutation status with lower-grade gliomas (LGGs) in patients. Materials and MethodsClinical, imaging, and pathological data were retrospectively collected from patients with pathologically confirmed LGGs. Based on multiparametric MRI, a predictive model for IDH-1 mutation status was constructed by combining radiomic features and deep learning features extracted from the 2.5D-CrossFormer deep learning model. Through feature selection, application of machine learning algorithms, and integration with clinical variables, a clinical-radiomics-deep learning nomogram model was developed. ResultsA total of 186 patients were included, with 79 IDH-1-positive cases and 107 IDH-1-negative cases. A total of 10 530 radiomic features and 32 deep learning features were extracted. After screening and feature dimensionality reduction, 20 radiomics-deep learning features were retained. Among various machine learning models, the LightGBM-based deep radiomics model performed best, with an area under the curve (AUC) of 0.94 in the training group and 0.86 in the validation group. The nomogram model constructed by combining clinical variables achieved an AUC of 0.97 in the training group, significantly outperforming the radiomics model and clinical model, and also demonstrated good predictive performance in the validation group. ConclusionsBased on multiparametric MRI, radiomics, and deep learning techniques, this study successfully constructed a combined model incorporating intratumoral and peritumoral edema features to predict the IDH-1 mutation status in LGGs. This model exhibits high diagnostic accuracy and has the potential to provide important imaging evidence for the formulation of treatment plans and prognosis assessment in LGGs patients. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.009 ObjectiveTo explore the value of MRI histogram analysis in predicting the long-term efficacy of transsphenoidal neuroendoscopic surgery for growth hormone-secreting pituitary neuroendocrine tumors before and early after surgery. Materials and MethodsA total of 43 patients with growth hormone-secreting pituitary neuroendocrine tumors from June 2021 to June 2023 were retrospectively included, among which 22 achieved remission and 21 did not. Clinical, pathological, and laboratory data were evaluated and recorded. Histogram parameters were extracted and calculated based on the T2WI sequence. The differences in histogram parameters combined with preoperative and early postoperative clinical semantic features between the long-term remission and non-remission groups were analyzed, and their diagnostic performance was evaluated. ResultsIn the preoperative clinical semantic characteristics between the two groups, statistically significant differences were observed in cavernous sinus invasion (χ² = 5.495), acromegaly (χ² = 4.240), preoperative growth hormone (GH) levels (Z = -2.821), and serum insulin-like growth factor-1 (IGF-1) levels (t = -2.856) (all P < 0.05). Among the early postoperative clinical semantic features, immediate postoperative GH (Z = -3.681) and IGF-1 levels (t = 0.247) also demonstrated statistically significant differences (both P < 0.05). Regarding histogram parameters, significant differences were found in area (t = -2.716) and kurtosis (Z = -2.332) (both P < 0.05). The diagnostic performance of the model was evaluated using receiver operating characteristic (ROC) curve analysis and area under the curve (AUC). The predictive model incorporating MRI histogram parameters with preoperative and early postoperative clinical, pathological, and laboratory data exhibited optimal performance, achieving an AUC of 0.963. At a cutoff value of 0.447, the sensitivity and specificity were 95.2% and 86.4%, respectively. ConclusionsT2WI histogram parameters provide additional value in predicting the long-term efficacy of GH-PitNET postoperative outcomes. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.010 ObjectiveTo evaluate the application value of deep learning (DL)-based accelerated T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI) in head and neck tumors. Materials and MethodsThirty-five untreated patients with head and neck tumors were prospectively enrolled and underwent head and neck MRI standard (T1WI, T2WI-Dixon) and DL sequences (DL-T1WI, DL-T2WI-Dixon). Image quality was subjectively rated by two radiologists using a five-point scale for overall image quality, artifacts and lesion conspicuity. Objective image quality was assessed by calculation of signal-to-noise ratio (SNR) of muscle, fat and tumor and contrast-to-noise ratio (CNR) of tumor in standard and DL sequences by one radiologist. Scan time and image quality scores were compared between standard and DL sequences using Kruskal-Wallis test. ResultsDL-T1WI (89 s) and DL-T2WI-Dixon (101 s) sequences reduced 46% scan time compared to standard T1WI (164 s) and T2WI-Dixon (188 s) sequences, respectively. There were no significant difference in overall image quality, artifacts and lesion conspicuity between DL-T1WI, DL-T2WI-Dixon sequences and standard T1WI and T2WI-Dixon sequences (all P > 0.05). SNR of fat and tumor and CNR of tumor in DL-T1WI sequence were comparable with that in standard T1WI sequence (all P > 0.05), SNR of muscle, fat and tumor and CNR of tumor in DL-T2WI-Dixon sequence were comparable with that in standard T2WI-Dixon sequence (all P > 0.05). ConclusionsDL-based accelerated MRI sequences could effectively reduce scanning time in patients with head and neck tumors. Except for the SNR of muscle in DL-T1WI sequence, the remaining objective image quality metrics of DL sequences are comparable to those in standard sequences. Moreover, compared to standard T1WI and T2WI-Dixon sequences, DL-T1WI and DL-T2WI-Dixon sequences could maintain excellent subjective image quality. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.011 ObjectiveTo quantitatively assess correlations and differences in biventricular function between patients with heart failure with mildly reduced ejection fraction (HFmrEF) and heart failure with preserved ejection fraction (HFpEF) in the context of dilated cardiomyopathy (DCM) using cardiovascular magnetic resonance feature-tracking (CMR-FT), and to evaluate the predictive value of biventricular myocardial strain parameters for HFmrEF. Materials and MethodsClinical and imaging data of patients diagnosed with heart failure and previously confirmed DCM were retrospectively collected from the First Affiliated Hospital of Xinjiang Medical University between January 2021 and June 2024. Conventional cardiac magnetic resonance (CMR) parameters and myocardial strain parameters were obtained using the CVI42 software. Based on heart failure characteristics, patients were divided into HFmrEF and HFpEF groups. Baseline clinical data, basic CMR parameters, and global/segmental myocardial strain parameters were compared between the two groups. Elastic-net regularized regression and multivariate logistic regression were used to identify independent factors associated with HFmrEF in DCM patients and to construct receiver operating characteristic (ROC) curves. ResultsA total of 67 patients were enrolled, including 28 in the HFpEF group and 39 in the HFmrEF group. Significant differences were observed between the two groups in N-terminal pro-B-type natriuretic peptide (NT-proBNP), high-sensitivity troponin T (hs-TnT), and myoglobin levels (all P < 0.05). Compared with the HFpEF group, the HFmrEF group showed significantly reduced biventricular stroke volume index, left ventricular cardiac output, and cardiac index (all P < 0.05); significantly increased biventricular end-diastolic/end-systolic volume indices and myocardial mass index (all P < 0.05); and impaired global and segmental myocardial strain (all P < 0.05 except for LV apical longitudinal strain). Multivariate logistic regression analysis identified right ventricular stroke volume index (OR = 0.863, P = 0.008), left ventricular mid longitudinal strain (OR = 1.406, P = 0.004), and right ventricular mid longitudinal strain (OR = 1.110, P = 0.025) as independent predictors of HFmrEF. The area under the ROC curve of the three-variable combined model was the largest, at 0.896 (95% CI: 0.823 to 0.968). ConclusionsCMR-FT can accurately characterize myocardial strain in DCM patients with HFmrEF and HFpEF. Right ventricular stroke volume index, left ventricular mid longitudinal strain, and right ventricular mid longitudinal strain serve as independent predictors of HFmrEF in DCM patients, providing important assessment value for this patient subgroup. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.012 ObjectiveTo explore the value of non-contrast cardiac magnetic resonance (CMR) T1ρ mapping in evaluating myocardial fibrosis in patients with hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy (DCM). Materials and MethodsCMR images and clinical data of 63 patients clinically diagnosed with HCM and DCM in Zhongshan Hospital Affiliated to Fudan University from July 2023 to December 2023 were prospectively included. According to the manifestations of late gadolinium enhancement (LGE), the patients were divided into three groups: LGE (+), LGE (+-) and LGE (-) groups, they represent patients with obvious LGE areas, patients with suspicious LGE areas, and patients with no detected LGE areas in any myocardial segment respectively. At the same time, 20 healthy volunteers were included as the control group. The general information of the participants and the imaging data of CMR examinations were collected. The cardiac function indexes, native T1 and T1ρ values were compared between the patient group and the control group. Depending on whether the continuous variables conformed to normality, parametric tests (independent samples t-test) and non-parametric tests (Mann-Whitney test) were used for comparison respectively. In LGE (+) group, paired t-test or paired rank sum test was used to compare the non-LGE regions and LGE-positive regions of their own myocardium.In the patient and control groups, one-way ANOVA was used, and if there is a statistical difference (P < 0.05), post-hoc tests (such as Dunnett) were conducted for pairwise comparisons, or Kruskal-Wallis test was used, followed by Dunn^,s test to correct for multiple comparisons. ResultsThe heart rate of HCM patients was lower than that of the control group (P = 0.021). Compared with the control group, the left ventricular cardiac output of HCM and DCM patients was lower (P = 0.006, P < 0.001), while the left ventricular mass increased (all P < 0.001). In the LGE (+) group, the T1ρ and native T1 values in the LGE-negative regions were (55 ± 3) ms and (1046 ± 30) ms (compared with the control group, P = 0.009, P = 0.014), respectively. The myocardial T1ρ and native T1 values in the LGE-positive regions of both HCM patients and DCM patients were significantly increased, and there were statistically significant differences in the T1ρ and native T1 values between these regions and the non-LGE regions of their own myocardium as well as the myocardium of the control group (all P < 0.001). In the LGE (+-) group, the overall myocardial T1ρ and native T1 values were (57 ± 3) ms and (1070 ± 40) ms (compared with the control group, P = 0.032, P = 0.007), respectively. Then, for the patients in the LGE (-) group, the overall myocardial native T1 value was (1040 ± 30) ms , and the T1ρ value was (57 ± 2) ms (compared with the control group, P = 0.667, P < 0.001). The measurement of myocardial T1ρ and T1 values showed good intra-observer (ICC = 0.93/0.99, all P < 0.001) and inter-observer (ICC = 0.88/0.98, all P < 0.001) agreement. ConclusionsThe T1ρ mapping technique is a reliable tool for quantitatively detecting myocardial fibrosis. It can be used for non-contrast evaluation of myocardial fibrosis in HCM and DCM. It demonstrates particular utility in patients with diffuse fibrosis and outperforms native T1 mapping in evaluating early-stage lesions within LGE-negative myocardial regions. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.013 ObjectiveTo explore the application value of cardiac magnetic resonance tissue tracing (CMR-TT) and T1 mapping technology in the assessment of myocardial injury in patients with type 2 diabetes mellitus (T2DM). Materials and MethodsA total of 64 patients with T2DM and 32 healthy controls (HC) who underwent cardiac magnetic resonance examination in our hospital from December 2023 to April 2025 were prospectively collected. All cardiac magnetic resonance image data were imported into special software for analysis, and the myocardial strain parameters, biventricular function parameters and left ventricular T1 mapping parameters were obtained, and the above parameters were compared between the two groups by t-test, Mann-Whitney U test and chi-square test, and the association between myocardial structure, function and myocardial strain was analyzed by Pearson and Spearman correlation. ResultsIn the T2DM group, the left ventricular myocardial mass index (LVMI) and left ventricular remodeling index (LVRI) were increased (both P < 0.001), the global longitudinal peak strain in the left ventricle (LV GLS) and the global longitudinal strain of the right ventricle decreased (both P<0.05) and the absolute values of peak systolic strain rate of the left ventricle (LV PSSR), longitudinal LV PSSR and left ventricular peak strain rate (LV PDSR) in the T2DM group were decreased (all P < 0.019).The storage strain and cathete strain of the left atrium/right attrium (LA/RA) were decreased in patients with T2DM (both P < 0.001). The extracellular volume (ECV) of T2DM patients was higher than that of HC group (P < 0.001).Biventricular ejection fraction, end-systolic volume index were correlated with biventricular strain function (all P < 0.003). LVMI is correlated with the Global radial strain of the left ventricle (LV GRS), global circumferential strain of the left ventricle (LV GCS), LV GLS, circumferential LV PSSR, longitudinal LV PSSR, radial LV PDSR, circumferential LV PDSR, Longitudinal LV PDSR(all P < 0.021).The left ventricular end-diastolic volume index was correlated with LV GCS, LV GLS, circumferential LV PSSR, longitudinal LV PSSR, and circumferential LV PDSR (all P < 0.044). The right ventricular end-diastolic volume index was correlated with the global circumferential strain of the right ventricle (r = 0.331, P = 0.007). LVRI was correlated with LV GLS and longitudinal LV PDSR (both P < 0.01), and weakly correlated with radial LV PSSR (r = 0.266, P = 0.034). ConclusionsCompared with the control group, the whole heart myocardial strain of T2DM patients is reduced, and the ECV value is higher. Biventricular myocardial structure, function and myocardial strain are interrelated. CMR-TT and T1 mapping techniques can effectively detect diabetic myocardial injury. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.014 ObjectiveTo explore the value of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), time-signal intensity curve (TIC), and apparent diffusion coefficient (ADC) in predicting the expression status of hormone receptors (HR) in breast cancer, and to evaluate the correlation and predictive value between MRI features and hormone receptor status. Materials and MethodsThe MRI findings of 206 patients with pathologically confirmed invasive breast cancer from November 2019 to March 2025 were retrospectively analyzed, and the differences of morphological signs, TIC and ADC values in different HR expression states (HR+/HR-) breast cancer in DCE-MRI were analyzed. Multivariate regression analysis of univariate and multivariate MRI findings with statistical significance was performed, a logistic regression model was established, and ROC curve was drawn, evaluating the efficacy of MRI features in predicting the expression status of hormone receptors in breast cancer. ResultsThe morphological features (including tumor maximum diameter, margin, burr sign and enhancement mode), TIC and ADC in HR+ group and HR- group were statistically significant (P < 0.05), and HR+ was more likely to show maximum diameter ≤ 2 cm, blurred edges, burrs and uneven enhancement, and TIC was a type Ⅲ curve. And the average ADC value is lower than HR-. MRI morphological features (including maximum meridian, burr, enhancement mode) and ADC values can predict hormone receptor-positive and negative breast cancer, and the area under the curve (AUC) was 0.806 (95% CI: 0.747 to 0.864), 0.669 (95% CI: 0.593 to 0.744), the sensitivities were 74.1% and 75.3%, and the specificities were 74.4% and 51.2%, respectively, while the AUC for the combined predicting of HR expression status was 0.837 (95% CI: 0.784 to 0.890), the sensitivity, specificity and accuracy were 84.0%, 65.6%, respectively, among which MRI morphological features combined with ADC value had the highest value. ConclusionsThe combined model of MRI morphological features and ADC values has a good predictive value for the expression status of hormone receptors in breast cancer, thereby providing a basis for the formulation and adjustment of clinical treatment plans. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.015 ObjectiveTo investigate the value of an intratumoral and peritumoral deep learning radiomics (DLR) model based on dynamic contrast-enhanced MRI (DCE-MRI) for predicting neoadjuvant chemotherapy (NAC) response in triple-negative breast cancer (TNBC). Materials and MethodsThis retrospective study enrolled 161 TNBC patients from two medical centers who underwent pre-NAC DCE-MRI. Data from center 1 (n = 112) served as the training set, while data from center 2 (n = 49) constituted an external validation cohort. An additional public TICA dataset (n = 74) was used as an independent external test set. The study comprised three components: traditional radiomics, deep learning (DL), and integrated DLR modeling. Radiomic features were extracted from both the intratumoral region and the peritumoral rims at 3 mm, 5 mm, and 7 mm distances using PyRadiomics. Performance of seven classifiers, support vector machine (SVM), k-nearest neighbors (KNN), extreme gradient boosting (XGBoost), extra trees (ET), logistic regression (LR), random forest (RF), and naive bayes (NB), was evaluated across different feature models to identify the optimal classifier for constructing the radiomics model. DL was implemented with a 3D DenseNet-121 backbone. Following the training and prediction of the DL model, the generated deep learning score (DL_score) was fused with the radiomics features. XGBoost, selected as the optimal classifier, was then used to build the final DLR fusion model. Diagnostic performance was assessed via receiver operating characteristic (ROC) curves and area under the curve (AUC). Calibration curves evaluated model fit, while decision curve analysis (DCA) quantified clinical utility. ResultsThe optimal radiomics model was an ExtraTrees-based model combining intra-tumoral and peri-tumoral 3 mm region features, with AUC values of 0.847, 0.780, and 0.720 in the training, external validation, and external test sets, respectively. The DL model outperformed the radiomics model in identifying the response to NAC in TNBC patients, with AUC values of 0.865, 0.810, and 0.820 in the training, external validation, and external test sets, respectively. Compared to a single model, DLR further improved the discriminative ability, with AUC and accuracy of 0.917, 0.898, and 0.886, and 90.1%, 87.9%, and 86.5% in the training, external validation, and external test sets, respectively, demonstrating better clinical benefits and good calibration. ConclusionsThe DLR fusion model, integrating intratumoral and peritumoral deep-learning radiomic features derived from DCE-MRI, demonstrates potential clinical utility for predicting NAC response in TNBC patients. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.016 ObjectivesTo investigate the diagnostic value of habitat imaging based on diffusion spectrum imaging (DSI) for differentiating clear cell renal cell carcinoma (ccRCC) at different grades and fat-poor angiomyolipoma (fpAML). Materials and MethodsA prospective study was conducted on 59 patients, all of whom underwent multi-b-value diffusion-weighted imaging (DWI) examination (0 to 3000 s/mm²). The apparent diffusion coefficient (ADC), the fractional order calculus (FROC) model-related parameters, the diffusion coefficient (D), the tissue heterogeneity-related parameters (fractional order parameter β), and the microstructure quantity (μ) were measured. The mean diffusivity (MD) and mean kurtosis (MK) of the diffusion kurtosis imaging (DKI) model-related parameters, as well as the true diffusion coefficient (Dt), pseudo-diffusion coefficient (D^*), and perfusion fraction (f) of the intravoxel incoherent motion (IVIM) model-related parameters were also measured. Based on these parameters, the paired data of each voxel within the renal tumor region of all subjects were input into the K-means algorithm, and the renal tumors were classified into four habitats. The diagnostic value of these parameters for different grades of ccRCC and fpAML was analyzed using the receiver operating characteristic (ROC) curve. ResultsHabitat 1, characterized by low heterogeneity, high perfusion and low diffusion, and habitat 4, characterized by high heterogeneity, high perfusion and high diffusion, show differences in fpAML and ccRCC of different grades. Habitat 1 and 4 were statistically significant in differentiating fpAML and different grades of ccRCC (P < 0.05). Among them, the area under the curve (AUC) of habitat 1, 4 and their combination for differentiating fpAML and low-grade ccRCC was 0.90 [95% confidence interval (CI): 0.77 to 0.97], 0.84 (95% CI: 0.70 to 0.94), and 0.89 (95% CI: 0.76 to 0.97), respectively. The AUC of habitat 1 and their combination for differentiating fpAML and high-grade ccRCC was 0.68 (95% CI: 0.49 to 0.84) and 0.72 (95% CI: 0.53 to 0.87), respectively. The AUC of habitat 1, 4 and their combination for differentiating high and low grades of ccRCC was 0.73 (95% CI: 0.57 to 0.85), 0.76 (95% CI: 0.61 to 0.88), and 0.76 (95% CI: 0.60 to 0.87), respectively. ConclusionsThe use of habitat imaging based on DSI shows the potential of non-invasive diagnosis of fpAML and different grades of ccRCC in clinical practice, and has high accuracy. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.017 ObjectiveTo evaluate the diagnostic performance of susceptibility-weighted imaging (SWI) in differentiating benign from malignant Bosniak ⅡF-Ⅲ renal cystic lesions. Materials and MethodsThis retrospective analysis included 38 Bosniak ⅡF-Ⅲ lesions including 17 benign, and 21 malignant cases with pathological confirmation. Two radiologists independently evaluated conventional magnetic resonance imaging (MRI) sequences (T1WI/T2WI) and SWI features through hemorrhagic foci, microvascular density, and ITSS area ratio. Statistical analysis incorporated Cohen^,s Kappa, Mann-Whitney U test, and binary logistic regression to develop a composite predictive model. Receiver operating characteristic (ROC) curve analysis was performed to compare the diagnostic performance of individual ITSS parameters and the combined predictor. ResultsConventional MRI (T1WI/T2WI) signal characteristics showed no significant intergroup differences (P > 0.05). Substantial interobserver agreement was observed for dominant ITSS structures and ITSS area ratios (Kappa value of 0.72 and 0.74), with excellent agreement for the number of intratumoral hemorrhagic lesions and the number of intratumoral vessels (Kappa value of 0.90 and 0.84). Malignant lesions demonstrated significantly higher the number of intratumoral hemorrhagic lesions (P = 0.02), the number of intratumoral vessels (P < 0.01), and ITSS area ratio (P < 0.01). Compared to individual parameters, the composite predictive model achieved superior diagnostic performance [AUC = 0.943, 95% (confidence interval, CI): 0.877 to 1.000] in the number of intratumoral hemorrhagic lesions (AUC = 0.695, 95% CI: 0.520 to 0.869), the number of intratumoral vessels (AUC = 0.868, 95% CI: 0.757 to 0.980), and ITSS area ratio (AUC = 0.877, 95% CI: 0.771 to 0.983). ConclusionsComprehensive analysis of susceptibility signals on SWI provides valuable information for differentiating benign and malignant Bosniak ⅡF-Ⅲ renal cystic lesions, offering reliable imaging evidence for clinical decision-making. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.018 ObjectiveTo explore the clinical value in predicting high/low grade bladder carcinoma and muscle layer invasive bladder carcinoma (MIBC) and non-myoinvasive bladder carcinoma (NMIBC) with various parameter values of intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI). Materials and MethodsIncluding 57 patients with bladder cancer, all of them underwent preoperative bladder multi-parameter magnetic resonance imaging (mp-MRI) (including T2WI, IVIM-DWI, and DWI). The patients were classified into high-grade and low-grade bladder cancer groups based on the degree of differentiation of bladder cancer; and into MIBC and NMIBC groups based on whether the bladder cancer had infiltrated the muscle layer of the bladder wall. Two physicians independently delineated region of interest (ROI) for the bladder cancer lesions on IVIM-DWI images (b = 800 s/mm²), calculated the intra-group correlation coefficients (ICC), and evaluated the repeatability and consistency of the measurement results. The apparent diffusion coefficient (ADC), true diffusion coefficient (D), perfusion fraction (f), and perfusion-related diffusion coefficient (D^*) were analyzed. Establish a binary logistic regression model, combining the parameter values to calculate the prediction values for different grades of bladder cancer and whether there is muscle invasion, both independently and in combination. Use the area under the receiver operating characteristic curve (AUC) to evaluate the diagnostic value of IVIM-DWI in predicting high-grade, low-grade, and muscle-invasive bladder cancer. The comparison of AUCs uses the DeLong test. ResultsThe ADC, D, and f values measured by different physicians and the same physician on three occasions showed good repeatability (ICC range 0.916 to 0.991). The ADC, D, and f values for high-grade bladder cancer group were (1.403 ± 0.575) × 10^-3 mm²/s, (7.276 ± 5.895) × 10^-3 mm²/s, and 0.490 ± 0.203, all lower than those of the low-grade bladder cancer group [ADC, D, and f values were (1.810 ± 0.288) × 10^-3 mm²/s, (19.522 ± 6.274) × 10^-3 mm²/s, and 0.873 ± 0.174; P < 0.001]; the ADC, D, and f values for MIBC group were (1.382 ± 0.334) × 10^-3 mm²/s, (9.686 ± 9.069) × 10^-3 mm²/s, and 0.543 ± 0.261, all lower than those of the NMIBC group [ADC, D, and f values were (1.822 ± 0.445) × 10^-3 mm²/s, (18.116 ± 6.490) × 10^-3 mm²/s, and 0.842 ± 0.193; P < 0.001]. After DeLong testing, in both high-grade and low-grade bladder cancer groups, the independent predictive efficacy of ADC, D, and f was lower than that of their combined use (AUC = 0.774, 0.822, 0.801, 0.869), with statistically significant differences (P = 0.018, 0.027, 0.028); in MIBC and NMIBC groups, the independent predictive efficacy of ADC, D, and f was also lower than that of their combined use (AUC = 0.568, 0.595, 0.623, 0.671), with statistically significant differences (P = 0.009, 0.034, 0.024). ConclusionsThe combined predictive efficacy of ADC, D and f values of IVIM-DWI for high-grade and low-grade bladder cancer, MIBC and NMIBC was higher than that of independent predictive efficacy. It can be used to preoperative prediction of bladder cancer grade and presence or absence of bladder wall muscle infiltration. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.019 ObjectiveTo identify the intrinsic imaging phenotype based on multi parameter magnetic resonance imaging (mpMRI) radiomics clustering analysis of prostate cancer, in order to evaluate the invasiveness of prostate cancer and predict prognosis. Materials and MethodsWe retrospectively collected preoperative mpMRI and clinical pathological data of 185 patients with pathologically confirmed prostate cancer from January 2022 to January 2024. The mpMRI includes ZOOMit diffusion weighted imaging (ZOOMit-DWI), ZOOMit apparent diffusion coefficient (ZOOMit-ADC), T2WI, and T2WI fat suppression (T2WI-FS). Cluster analysis was performed based on extracting radiomics features from mpMRI to obtain clustering subtypes. Chi square test was used to analyze the categorical variables Gleason score, significant prostate cancer (sigPCA), P504S, lymph node metastasis (LNM), prostate-specific antigen (PSA), visible cancer thrombus in the vasculature, perineural invasion (PNI), and Ki-67 in clinical pathological variables. Independent sample t-test was used to evaluate the age and prostate volume (PV) of clinical continuous variable data, exploring the relationship between clinical pathological variables and subtypes. Is there a significant difference or association. ResultsTwo cluster subtypes were obtained. Cluster 1 was associated with a higher incidence of clinically significant prostate cancer (92.857%) and lymph node metastasis (21.429%). There was a statistically significant difference between cluster 1 and cluster 2, with P values of 0.024 (sigPCA) and 0.028 (LNM), respectively. There was a statistically significant difference in Gleason scores between the two subtypes of clusters (P = 0.035). In cluster 1, the proportion of (4 + 3) scores was the highest, at 32.143%, followed by (3 + 4) and (5 + 4) scores, both at 15.476%. In Cluster 2, the proportion of (3 + 4) scores is the highest, at 26.733%. The proportions of (3 + 5) and (5 + 3) for the two subtypes of clusters are 2.381% and 1.980%, respectively. There was no statistically significant difference in the incidence of PNI between the two clusters (P = 0.754). ConclusionsThe potential imaging phenotype of multi parameter MRI in prostate cancer is associated with the incidence of sigPCA and LNM, which symbolize high invasiveness and poor prognosis. This can help evaluate the prognosis of PCA patients and provide individualized treatment after risk stratification. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.020 ObjectiveTo explore the application value of the intratumoral and peritumoral apparent diffusion coefficient (ADC) histogram of the whole tumor in preoperative prediction of lymphovascular space invasion (LVSI) in patients with early cervical cancer. Materials and MethodsA retrospective analysis was conducted on 150 patients with stage ⅠB-ⅡA1 cervical cancer confirmed by postoperative pathology, and were divided into LVSI-positive (n = 45) and LVSI-negative (n = 105) groups according to postoperative pathological results. All patients underwent pelvic MRI before surgery, and the region of interest (ROI) were manually delineated layer by layer along the largest edge of the tumor on the ADC axial image, with the peritumoral region being uniformly expanded outward. Whole-volume ADC histogram analysis was performed for intratumoral region, intratumoral-2 mm peritumoral region and intratumoral-4 mm peritumoral region, respectively. Difference in clinicopathologic characteristics and ADC histogram parameters between the two groups were analyzed, and establish a joint parameter model. Receiver operating characteristic (ROC) curve was used to evaluate the diagnostic efficiency of each parameter and the combined parameter model in preoperative prediction of LVSI status in patients with early cervical cancer, the area under the curve (AUC), optimum cutoff value, sensitivity and specificity were calculated, and the AUC values of each parameter and the joint model were statistically compared using the DeLong test. ResultsIn the intratumoral ADC histogram analysis, ADC_max, ADC_mean, ADC₅₀, ADC₇₅, ADC₉₀, ADC₉₅, ADC_stdev, ADC_variance and ADC_kurtosis in the LVSI-positive group were significantly lower than in the LVSI-negative group (P < 0.05). Intratumoral ADC histogram parameters in predicting LVSI status of early cervical cancer, ADC_max, ADC₉₀ and ADC₉₅ had the best diagnostic efficacy, with AUC of 0.747, 0.756 and 0.776, respectively (P < 0.05). The AUCs of the combined parameter models for intratumoral, intratumoral+2 mm peritumoral, and intratumoral+4 mm peritumoral regions were 0.830, 0.710, and 0.673, respectively. DeLong test revealed that the AUC of the intratumoral combined model was significantly higher than that of the intratumoral+2 mm peritumoral (P < 0.05) and intratumoral+4 mm peritumoral combined models (P < 0.05); no significant difference was found between the AUCs of the intratumoral+2 mm peritumoral and intratumoral+4 mm peritumoral combined models (P > 0.05). ConclusionsADC histogram based on whole tumor volume has potential value in preoperative predicting LVSI status in patients with early cervical cancer, among which ADC_max, ADC₉₀ and ADC₉₅ are the most promising predictive parameters, and peritumoral region cannot increase the diagnostic efficiency of ADC histogram. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.021 ObjectiveTo analyze the association between patellofemoral osteoarthritis (PFOA) and tibiofemoral osteoarthritis (TFOA), and analyze their potential common relationships. Materials and MethodsProspectively recruited 101 subjects (56 males, 45 females) who visited Joint Surgery Department of the Affiliated Hospital of Zunyi Medical University and underwent same-day weight-bearing knee X-ray and ipsilateral knee MRI examinations in the Radiology Department. Based on the Kellgren-Lawrence grading (KLG) system for knee osteoarthritis (KOA) on weight-bearing X-rays, the subjects were divided into four groups: KLG 0-1, KLG 2, KLG 3, and KLG 4. According to the Recht MRI grading system for cartilage damage, patellofemoral joint (PFJ) cartilage injuries were classified into four groups: Recht Ⅰ, Recht Ⅱ, Recht Ⅲ and Recht Ⅳ. MRI T2 mapping was used to measure the T2 values of patellar cartilage. Pearson or Spearman correlation analysis was employed to evaluate the relationships both between the MRI quantitative parameter (T2 values) and age, body mass index (BMI), PFJ Recht grading, KOA KLG and the correlation between PFJ Recht grading and KOA KLG. ResultsHigher T2 values were associated with older age (r = 0.47, P < 0.001). Higher T2 values correlated with more severe PFJ Recht grading (r = 0.86, P < 0.001). Higher T2 values were also linked to more advanced KOA KLG (r = 0.47, P < 0.001). Additionally, more severe PFJ Recht grading was associated with higher KOA KLG (r = 0.41, P < 0.001). ConclusionsA moderate correlation was identified between PFOA and TFOA, indicating potential shared characteristics between the two conditions, this association serves as radiographic support for total knee protective therapy. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.022 ObjectiveTo explore dynamic contrast-enhanced MRI (DCE-MRI) multi-parametric and radiomics data in constructing an evaluation model for inflammatory activity in axial spondyloarthritis, providing a reference for clinical diagnosis and treatment. Materials and MethodsThis study enrolled 93 patients clinically diagnosed with axial spondyloarthritis, who were classified into the active and inactive inflammatory groups based on the Ankylosing Spondylitis Disease Activity Score (ASDAS). All participants underwent DCE-MRI scans of the sacroiliac joints. The Omni-Kinetics post-processing software was utilized to measure quantitative permeability parameters, quantitative perfusion parameters, and semi-quantitative parameters in the region of interest (ROI) of the sacroiliac joints. ITK-SNAP software delineated three-dimensional volume of interest (VOI) encompassing the bilateral sacral and iliac bone surfaces. Radiomic features were extracted from these VOIs utilizing the Artificial Intelligent Kit (A.K.) software. Subsequently, predictive models for different levels of axial spondyloarthritis inflammatory activity were constructed employing cross-validation and the least absolute shrinkage and selection operator (LASSO) method. These models included the Spondyloarthritis Research Consortium of Canada (SPARCC) scoring model, DCE-MRI multi-parameter combined model, and DCE-MRI radiomics integrated model. The models were validated through receiver operating characteristic (ROC) curve analysis, with their performance evaluated based on the area under the curve (AUC), accuracy, sensitivity, and specificity. Results(1) The AUC (95% CI) for SPARCC was 0.697 (0.589 to 0.805). (2) Among DCE-MRI parameters, efflux rate constant (K_ep), time to peak (TTP), and extravascular extracellular volume fraction (V_e) showed significant differences (P < 0.05 by t-test or U-test), with AUCs (95% CI) of 0.628 (0.505 to 0.751), 0.648 (0.535 to 0.761), and 0.630 (0.511 to 0.749), respectively. The combined DCE-MRI parameter model achieved an AUC of 0.712 (0.600 to 0.823). (3) Radiomics features of DCE demonstrated AUCs (95% CI) ranging from 0.617 (0.489 to 0.746) to 0.889 (0.826 to 0.953), with the DCE-MRI combined radiomics model achieving an AUC of 0.951 (0.910 to 0.992). (4) The DCE-MRI radiomics integrated model^,s diagnostic performance was superior to the DCE-MRI multi-parameter combined model and SPARCC scoring model (AUC: 0.951 vs. 0.712; 0.951 vs. 0.697; both P < 0.001). ConclusionsThe DCE-MRI radiomics integrated model demonstrated significantly better performance than the DCE-MRI multi-parameter combined model and the SPARCC scoring model in assessing inflammatory activity in axial spondyloarthritis, providing a novel reference for its clinical management. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.023 ObjectiveTo evaluate the hippocampal injury at different time points during early brain injury (EBI) period following the induction of a subarachnoid hemorrhage (SAH) model in rats. Materials and MethodsFrom January to December 2023, 72 Sprague-Dawley (SD) rats were divided into a Sham operation group (Sham) (36 rats) and a SAH group (36 rats). Each group was further divided into six subgroups based on postoperative time points of 3, 6, 12, 24, 48, and 72 hours (6 rats per subgroup). The SAH model was established using the twice cisterna magna blood injection method, with the Sham group receiving an equal volume of normal saline instead of blood. After modeling, whole-brain MRI scanning was performed at each time point to measure the apparent diffusion coefficient (ADC) values of the hippocampus. Neurological function was evaluated using the modified Garcia scoring scale. The severity of SAH was evaluated using a conventional grading scale. Hippocampal tissues were subjected to hematoxylin-eosin (HE) staining to observe pathological changes, and the expression level of interleukin-1β (IL-1β) was detected by enzyme-linked immunosorbent assay (ELISA). ResultsCompared with the Sham group, the hippocampal ADC values in the SAH group showed significant differences at 12 and 24 hours (P < 0.05). Compared with the Sham group, in each SAH group, the neurological function scores of rats were decreased, the SAH severity scores were increased, and the expression of IL-1β was elevated (all P < 0.05). Compared with the Sham group, the expression of IL-1β in rats of each SAH group was elevated (P < 0.05). Compared with the Sham group, the SAH group exhibited increased neuronal density, cellular crowding, and neuronal degeneration around the pathological lesions. The ADC value of the hippocampus was negatively correlated with the IL-1β protein concentration and the SAH severity score (all P < 0.05, r = -0.695, r = -0.624), and positively correlated with the neurological function score (P < 0.05, r = 0.568); the IL-1β protein concentration was negatively correlated with the neurological function score (P < 0.05, r = -0.419) and positively correlated with the SAH severity score (P < 0.05, r = 0.568); the SAH severity score was negatively correlated with the neurological function score (P < 0.05, r = -0.680). ConclusionsDuring the EBI phase after SAH, the decrease in ADC values of the rat hippocampus, the increase in IL-1β protein concentration, the decrease in neurological function scores and the increase in the SAH severity scores were more pronounced at 12 and 24 hours after SAH modeling. The period of 12 to 24 hours after SAH may be the peak period for hippocampal injury, and the hippocampal ADC values could provide new therapeutic ideas and directions for clinical practice. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.024 ObjectiveTo explore the optimized scanning scheme for compressed sensing golden-angle radial sparse parallel sequence (CS-GRASP) and evaluate the application value of the optimized sequence in liver dynamic contrast-enhanced MRI (DCE-MRI) for patients with poor breath-holding. Materials and MethodsThis retrospective analysis was conducted on 46 patients with poor breath-holding capability who underwent dynamic contrast enhanced-magnetic resonance imaging of the liver at our hospital from March 2021 to October 2023, including 21 patients in the unoptimized CS-GRASP group and 25 patients in the optimized group. Signal intensity (SI) of the liver and erector spinae, standard deviation (SD), and the mean standard deviation of image background noise (SD noise) were measured at the levels of the main hepatic portal vein and its left and right branches during non-contrast, early arterial, and late arterial phases. The signal-to-noise ratio (SNR), contrast-noise ratio (CNR), and coefficient of variation (CV) of SI for liver CS-GRASP images of both groups were calculated. Subjective scoring was conducted for image noise, the severity of streak artifacts, image quality, and clarity of liver structures in the left and right liver lobes. ResultsDuring plain scanning, arterial early phase, and arterial late phase, the SNR and CNR of images of the left and right lobes of the liver in the CS-GRASP optimized group was higher than those in the unoptimized group, and the CV values were lower than that of the unoptimized group; the difference in CNR of the arterial early phase of the right lobe of the liver was not statistically significant between the two groups (P > 0.05), while the difference of the rest of the parameters were statistically significant (P < 0.05). In the CS-GRASP optimized group, scores for image noise and streak artifacts of the left and right liver lobes, scores for image clarity, and overall image quality were higher than those in the non-optimized group, with statistically significant differences (P < 0.05). ConclusionsThe optimized CS-GRASP sequence can improve image quality and reduce streak artifacts, making it a better alternative for patients with poor breath-holding during liver contrast-enhanced MRI. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.025 ObjectiveTo investigate the current application status of magnetic resonance imaging (MRI) equipment in the discipline construction of radiology departments within public hospitals, providing a reference for further promoting the high-level development of medical imaging disciplines. Materials and MethodsFrom October 16 to December 31, 2023, an electronic questionnaire was used to collect basic information about the radiology departments and fundamental details and efficiency indicators of MRI equipment from 210 Tier Ⅲ public hospitals in Jiangsu Province. Departments designated as "National, Provincial, or Municipal Key Clinical Specialties and/or Disciplines," or those included/nominated in the "Fudan University National Specialty Comprehensive Ranking or East China Regional Radiology Reputation Ranking" were defined as Outstanding Departments (n = 52). Others were defined as Regular Departments (n = 158). Descriptive analysis was performed using Python and SPSS software to compare differences in equipment status and efficiency indicators between the two types of departments. ResultsA total of 189 valid questionnaires were collected, accounting for 90.00% of the total Tier Ⅲ public hospitals in the province, including 52 Outstanding Departments and 137 Regular Departments. Outstanding Departments demonstrated superior performance in the average number of devices, brand variety, installation time, and number of physicians, technicians, and nurses (P < 0.05). However, there was no significant difference in equipment origin or magnetic field strength between the two groups (P > 0.05). In terms of efficiency, Outstanding Departments had higher daily average examinations per scanner, daily total examinations, and annual outpatient and inpatient examinations (P < 0.05), but lower proportions of outpatients and inpatients referred for MRI examinations (P < 0.05). ConclusionsThe configuration of MRI equipment is closely associated with the level of medical imaging discipline construction. Outstanding Departments not only have more personnel but also deployed more diverse and numerous devices earlier, undertaking a higher volume of examinations. Regular Departments showed homogeneity in equipment field strength and origin, yet exhibited relatively higher referral rates for outpatient and inpatient MRI examinations. High-level development of the imaging discipline should be promoted by integrating personnel training and equipment optimization, leveraging smart healthcare technologies to enhance quality and efficiency. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.026 Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.027 ^, functional recovery, mental and physical health, and long-term prognosis. In recent years, resting-state functional magnetic resonance imaging (rs-fMRI), a noninvasive and repeatable brain imaging technique, has been increasingly utilized to reveal spontaneous neural activity and functional connectivity patterns. This approach provides important neuroimaging evidence for understanding the underlying neural mechanisms of PSD and holds potential for early diagnosis and prognosis prediction. This review systematically summarizes the major findings, emerging analytical methods, and clinical implications of rs-fMRI in PSD research. It also discusses the limitations of existing studies and proposes future research directions, aiming to provide new insights in the mechanistic study and clinical management of PSD. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.028 Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.029 Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.030 Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.031 ^, long-term outcomes. ]]> Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.032 Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.033 Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.034 Sat,20 Sep 2025 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.09.035 Sat,20 Sep 2025 00:00:00 GMT