http://www.med-sci.cn/cgzcx/en/contents_list.asp?issue=202503 zh-cn An RSS feed for Chinese Journal of Magnetic Resonance Imaging http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.001 ObjectiveTo explore gender differences in morphological change patterns among different subgroups of Parkinson^,s disease (PD) patients. Materials and MethodsHigh-resolution T1-weighted magnetic resonance imaging and clinical scale data were collected from a total of 278 participants in the Parkinson^,s disease Progression Marker Initiative database. Using a cutoff score of 5 on the rapid eye movement sleep behavior disorder (RBD) screening questionnaire, patients were classified into PD-pRBD (≥ 5 points) and PD-nonRBD (< 5 points) groups. The final sample included 93 PD-pRBD patients, 114 PDnonRBD patients, and 71 healthy controls (HC). The Computational Anatomy Toolbox 12 (CAT12) tool was utilized to gather data on gray matter volume (GMV) and cortical morphological metrics. Subsequently, individual-level morphological similarity networks were constructed based on these cortical metrics. Finally, the topological properties of the network were analyzed using graph theoretic methods. ResultsIn the PD-pRBD group, GMV in the frontal and temporal lobes of males was lower than that of females, while the gyrification index (GI) in the frontal lobes was lower in females than in males (P = 0.024). However, the GI of the frontal lobe in males was lower than that in females within the PDnonRBD group (P = 0.009). Network analyses based on graph theory revealed that male PD-pRBD patients displayed lower network information integration compared to female patients, particularly in terms of the global properties of fractal dimension (FD) networks. Moreover, in the PD-pRBD group, male patients showed a strong correlation between morphological network metrics and cognitive performance as measured by the delayed memory score of Hopkins Verbal Learning Test-Revised (HVLT-R) memory scores and the Montreal Cognitive Assessment (MoCA) (all P < 0.05). ConclusionsPD patients with and without RBD exhibit significant sex-specific patterns at both the morphological and network levels. Moreover, the sex differences between males and females in the PD-RBD group are more extensive than those in the nonRBD group, and these differences are further associated with cognitive function. This finding emphasizes the importance of considering gender differences in the diagnosis and treatment of PD-pRBD patients. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.002 ObjectiveTo use quantitative susceptibility imaging (QSM) technology to study the brain iron difference between children with autism spectrum disorder (ASD), so as to provide new imaging markers for the pathophysiology and early diagnosis of ASD. Materials and MethodsThirty children with ASD were included as the experimental group and 30 normal children as the control group. After collecting clinical data and scales, all the children were scanned and processed with QSM sequence to obtain quantitative maps, and the area of interest was outlined manually to obtain the magnetization rate values. The differences in magnetization rate values between the two groups were compared, and their correlation with the Gesell Developmental Scale scores was analyzed. The diagnostic efficacy was assessed by plotting the Receiver operating characteristic (ROC) curve, and the ASD group was divided into mild-moderate and severe groups according to the Childhood Autism Rating Scale (CARS), and the magnetization rates between mild-moderate and severe groups and the control group were further compared. The differences in magnetization rate values between the mild-moderate, severe, and normal control groups were further compared. ResultsCompared with healthy children, the magnetic susceptibility values in the frontal white matter, left temporal white matter, red nucleus, substantia nigra, and dentate nucleus of children with ASD were significantly lower (P < 0.05). Correlation analysis revealed a positive correlation between the magnetic susceptibility value of the left frontal white matter and language scores, as well as between the magnetic susceptibility value of the right red nucleus and fine motor scores in children with ASD (P < 0.05). ROC curve analysis showed that the AUC value of the right dentate nucleus was the highest at 0.752 (95% confidence interval: 0.627 to 0.878), with a sensitivity of 76.7%, a specificity of 73.3%. The intergroup comparison based on ASD severity indicated significant differences in the magnetic susceptibility values of the right frontal white matter, left temporal white matter, right dentate nucleus, and left red nucleus between the normal control group and the mild-moderate group. The magnetic susceptibility value of the right dentate nucleus showed significant differences between the normal control group and the severe group. ConclusionsThe brain iron content in multiple regions of children with ASD is lower than that of typically developing children and is correlated with their clinical symptoms and severity, which has clinical significance. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.003 ObjectiveTo explore the central pathological mechanisms underlying the transition from acute to chronic neck pain using imaging transcriptomics. Materials and MethodsFrom March 2023 to May 2024, 86 patients with acute neck pain and 89 patients with chronic neck pain were recruited from Shanxi University of Chinese Medicine and Shanxi Provincial Hospital of Acupuncture and Moxibustion. Using a 3.0 T MR scanner to collect resting-state functional magnetic resonance imaging (rs-fMRI) data based on blood oxygen level-dependent signals from the subjects, with the hypothalamus as the seed region for whole-brain functional connectivity (FC) analysis. Imaging transcriptomics analysis was performed using the Allen Human Brain Atlas (AHBA) transcriptomic dataset, and partial least squares (PLS) regression was employed to investigate regional changes in brain functional activity and gene expression between chronic and acute pain patients. Gene enrichment analysis was conducted using Metascape to reveal the central cross-scale mechanisms during the chronicification of neck pain. ResultsCompared to acute neck pain, patients with chronic neck pain exhibited increased FC values in the left hypothalamus and right orbital superior frontal gyrus, and decreased FC values in the left hypothalamus and right middle frontal gyrus (Voxel-level P < 0.01, Cluster-level P < 0.05). Additionally, this study found that the PLS1 model explained 43.43% of the variance, and the PLS1weighted gene expression profile was positively correlated with the case-control t-map space (Pearson^,s r = 0.491, P < 0.05). Enrichment analysis revealed that PLS1+ genes were closely associated with cellular components such as "glutamatergic synapse" and biological processes like "synaptic signaling," while PLS1‍- genes were closely related to cellular components like "intermediate filament cytoskeleton," molecular functions such as "DNA binding transcription activator activity," and biological processes including "regulation of growth hormone secretion". ConclusionsA cross-scale analysis based on imaging transcriptomics has uncovered comprehensive changes in brain functional activity, gene expression, and cell composition during the chronicification of neck pain. This indicates that the chronicification of neck pain is a multi-level and multi-scale interactive process involving molecular levels, cellular structures, and brain network functions. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.004 ObjectiveTo construct a prediction model for recurrence of intracranial atherosclerotic stroke patients by combining the radiomic features of intracranial culprit plaques in three-dimensional high-resolution vessel wall imaging with MRI (3D-HRVWI) and intraplaque hemorrhage (IPH). This can help clinically target targeted interventions for high-risk populations to reduce the risk of future stroke recurrence. Materials and MethodsA total of 296 stroke patients who underwent HRVWI examination from November 2021 to August 2023 were retrospectively collected, and the imaging features of culprit plaques were measured in the non-contrast sequence T1WI and enhanced sequence CE-T1WI images of 296 patients, and the plaques were delineated, the radiomics features were extracted, and the feature correlation analysis and feature screening based on L1 regularization (linear models penalized with the L1norm, L1 Based) screened radiomics features, and all data were randomly divided into training group and test group in a 7 : 3 ratio. In the training group, the radiomics features of the screened responsible plaques were used to construct a radiomics model for predicting stroke recurrence, and the radiomics features and IPH were used to construct a combined model, and the performance was evaluated in the test group. The receiver operating curve (ROC) and area under the curve (AUC) were used to evaluate the predictive performance of each model, and the DeLong test was used to compare the differences between AUC, and finally the nomogram visualization model was established. ResultsThe mean age of the participants was 66 years, including 207 male participants (69.9%) and 89 female participants (30.1%), of whom 58 (19.6%) had recurrent stroke. IPH (OR = 8.577, 95% CI: 4.374 to 16.818) was an independent risk factor for stroke recurrence among the clinical features and radiographic features of the culprit plaques. The radiomics features of 2153 culprit plaques were extracted from the CE-T1WI and T1WI sequences, respectively, and after feature screening, 4 radiomics features were retained in the CE-T1WI sequence data and 6 radiomics features were retained in the T1WI sequence data. In the training group, the AUC was 0.757 (0.693 to 0.814) for IPH, 0.770 (0.707 to 0.826) for radiomics features, and 0.866 (0.811 to 0.909) for the combined model. In the test group, the AUC was 0.750 (0.647 to 0.836) for IPH, 0.819 (0.723 to 0.892) for radiomics features, and 0.880 (0.794 to 0.939) for the combined model. The results of DeLong^,s test showed that the combined model outperformed the IPH model in the training group and the test group (P < 0.05). ConclusionsThe 3D-HRVWI radiomics features of intracranial culprit plaque combined with IPH have good efficacy in predicting recurrence in patients with intracranial atherosclerotic stroke, which is better than the independent IPH model. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.005 ObjectiveTo evaluate the glymphatic system with diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) and enlarged perivascular space (EPVS) changes, and to explore its relationship with cognitive function, and then to find influence factors for cognitive changes in cerebral small vessel disease (CSVD) patients. Materials and MethodsThe clinical data of CSVD group (n = 81) and control group (n = 20) were prospectively collected. According to the neuropsychological evaluation results, the CSVD patients were divided into normal cognition group (CSVD-NC, n = 27), mild cognitive impairment group (CSVD-MCI, n = 31), and vascular dementia group (CSVD-VaD, n = 23). The routine MRI sequence, diffusion tensor imaging (DTI) were performed in all. The grade of EPVS of basal ganglia (BG) and centrum semi-ovale (CSO) was quantitatively scored, and ALPS-index was measured. The differences of EPVS and ALPS-index were compared among the four groups. Correlation analysis and multiple linear regression was further analyzed between ALPS-index, EPVS and cognitive function. ResultsCompared with the HC group, the ALPS-index were decreased in the CSVD-NC group, CSVD-MCI group and CSVD-VaD group, the BG-EPVS and CSO-EPVS score were increased in the CSVD-MCI group and the CSVD-VaD group (P < 0.05). The BG-EPVS was negatively correlated with ALPS index, MoCA score and MMSE score in CSVD patients (r = -0.291, P = 0.008; r = -0.342, P = 0.002; r = -0.309, P = 0.005), and ALPS-index was positively correlated with MoCA score and MMSE score (r = 0.226, P = 0.043; r = 0.225, P = 0.044). Multiple linear regression analysis showed that BG-EPVS was independent factor of MoCA score and MMSE score (β = -0.294, P = 0.009; β = -0.274, P = 0.017). ConclusionsThe ALPS index and EPVS changes could be used to evaluate glymphatic system in CSVD patients, which were closely related to cognition impairment. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.006 ObjectiveTo compare the feasibility and value of three-dimensional double echo steady state (3D-DESS) sequence and three-dimensional flip-angle evolution short-tau inversion (3D-SPACE-STIR) sequence in the display of mandibular nerve extracranial segment in patients with nasopharyngeal carcinoma. Materials and MethodsThe image data of 36 patients with nasopharyngeal carcinoma meeting the inclusion and exclusion criteria were retrospectively analyzed, with a total of 72 mandibular nerves. The paired sample t test and paired sample Wilcoxon sign rank sum test were used to compare the subjective scores and objective parameters of the display quality of the two sequences, including the signal intensity ratio (SIR), signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the images. Paired Chi-square test was used to compare the mandibular nerve involvement between the two sequences in nasopharyngeal carcinoma. ResultsBoth 3D-DESS and 3D-SPACE-STIR sequences had a higher display rate of mandibular nerve, and the subjective quality scores of both showed a good performance, which were 4 and 3 points, respectively, with no statistical significance (P = 0.052); 3D-SPACE-STIR sequence SIR_N/M was better than 3D-DESS sequence, 3.55 and 1.40, respectively, and the difference was statistically significant (P < 0.001). SNR_M of 3D-DESS sequence was better than that of 3D-SPACE-STIR sequence (13.68 and 8.00, respectively), and the difference was statistically significant (P = 0.002). A total of 24 mandibular nerves were involved by tumors. 3D-DESS sequences showed better relationship between nerves and tumors (95.83% vs. 12.50%), nerve continuity (87.50% vs. 37.50%) and nerve morphology (70.83% vs. 29.17%) of tumor segments than 3D-SPACE-STIR sequences, and the differences were statistically significant (P < 0.001, P < 0.001, P = 0.004). The 3D-SPACE-STIR sequence showed that the nerve thickening in non-tumor segment was better than that in 3D-DESS sequence (70.83% vs. 41.67%), and the difference was statistically significant (P = 0.042). Conclusions3D-DESS and 3D-SPACE-STIR have similar display rates on the extracranial segment of mandibular nerve, but 3D-DESS is based on better resolution of the relationship between nerve and surrounding structure and low field of view value, and its clinical application is worthy of promotion. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.007 ObjectiveTo evaluate the performance of radiomics in predicting axillary lymph node metastasis (ALNM) including sentinel lymph node metastasis (SLNM) in breast cancer by Meta analysis. Materials and MethodsA systematic search was conducted in the electronic databases PubMed, Embase, Web of Science, Cochrane Library, CNKI, and Wanfang database for relevant studies published between January 1, 2018 and February 23, 2024. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) was used to assess the quality of the included studies. The diagnostic odds ratio (DOR), sensitivity, specificity, and summary receiver operating characteristic (SROC) curve were calculated to evaluate the diagnostic value of imagingomics for ALNM, including SLNM, in breast cancer patients. Spearman correlation coefficients were calculated to assess threshold effects, and meta-regression and subgroup analyses were performed to explore possible causes of heterogeneity. ResultsA total of 22 studies involving 4230 patients were included in the meta-analysis, summarizing the overall diagnostic accuracy of imaging-omics detection of ALNM, including SLNM: DOR was 34 [95% confidence interval (CI): 21 to 54]; sensitivity 87% (95% CI: 85% to 89%); specificity 76% (95% CI: 75% to 78%); The area under the curve (AUC) of the SROC curve was 0.92, and Q^* was 0.86; The positive likelihood ratio was 5.30 (95% CI: 3.70 to 7.60); The negative likelihood ratio was 0.17 (95% CI: 0.13 to 0.22). Meta-analysis showed that there was significant heterogeneity among the included studies, and there was no evidence of threshold effect. ConclusionsOur results suggest that imagingomics has good diagnostic performance in predicting ALNM, including SLNM, in breast cancer. Therefore, we recommend this method as a clinical method for preoperative identification of ALNM and SLNM. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.008 ObjectiveTo develop and validate intratumoral and multiregion peritumoral radiomics models based on contrast-enhanced magnetic resonance imaging (CEMRI) for predicting pathological differentiation in hepatocellular carcinoma (HCC) patients. Materials and MethodsA total of 213 HCC patients diagnosed between January 2020 and July 2023 at the Third Affiliated Hospital of Soochow University was included in the retrospective study, comprising 62 poorly differentiated HCC (pHCC) and 161 non-poorly differentiated HCCs (npHCC). The HCCs were randomly divided into training (149 patients, 156 HCCs) and validation (64 patients, 67 HCCs) cohorts at a 7∶3 ratio. The ITK-SNAP software delineated the region of interest (ROI) on arterial, portal vein, and delayed phase images, while PyRadiomics software extracted 3045 radiomic features. Feature selection was carried out using Spearman rank correlation, least absolute shrinkage and selection operator (LASSO), and maximum relevance-minimum redundancy (mRMR) approaches, followed by support vector machine algorithm to build Intratumoral, 5 mm peritumoral (Peri_5mm), 10 mm peritumoral (Peri_10mm), and Intratumoral + 10 mm peritumoral (IntraPeri) models. The predictive performance of these models was assessed using the area under the curve (AUC) of receiver operating characteristic and decision curve analysis (DCA). ResultsThe Intratumoral, Peri_5mm, Peri_10mm, and IntraPeri models consisted of 10, 17, 11, and 12 features, respectively. In the Intratumoral model, the AUC values for predicting pHCC in the training and validation cohorts were 0.92 and 0.93, respectively. The Peri_10mm model exhibited higher AUCs compared to the Peri_5mm model: 0.88 versus 0.82 in the training cohort and 0.90 versus 0.85 in the validation cohort. The IntraPeri model demonstrated superior performance with AUC values of 0.95 and 0.95 in the training and validation cohorts, respectively. DCA suggested that the Intratumoral, Peri_5mm, and Peri_10mm models provided notable clinical benefits, with the IntraPeri model being the most optimal. ConclusionsThe IntraPeri model based on CEMRI can accurately predict HCC differentiation and has good clinical benefits. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.009 ObjectiveTo develop a nomogram combining clinical biomarkers and MRI features to predict vessels encapsulating tumor clusters (VETC) in hepatocellular carcinoma (HCC). Materials and MethodsRetrospective analysis of clinical and imaging data of 213 patients with surgical pathologically confirmed HCC, and the patients were divided into training and validation cohorts in a ratio of 7∶3 according to chronological order, and the differences in clinical, pathological and imaging features between the two groups were compared. Univariate and multivariate logistic regression analysis were used to analyze the independent risk factors, including clinical biomarkers and imaging features for VETC in training cohort. Nomogram for predicting VETC were developed based on the results of regression analysis, and this nomogram was validated using the validation cohort. ResultsOne hundred and forty-eight patients were included in the training cohort and 65 patients in the validation cohort, and there was no statistical difference in clinical, pathological and imaging features between the two groups. In the logistic regression analysis, AFP > 400 ng/mL, larger tumor diameter, greater number of tumors, non-smooth tumor margin and presence of intra-tumoral artery were the independent risk factors for predicting VETC. The C index of the nomogram developed based on the above factors was 0.825 and 0.817 in the training and validation cohort, respectively. ConclusionsThe nomogram developed by clinical biomarkers and MRI features has good accuracy in predicting VETC and can directly visualize the probability of VETC, which can facilitate personalized treatment plans. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.010 ObjectiveTo explore the value of combining whole-tumor apparent diffusion coefficient (ADC) histogram parameters with imaging biomarkers in predicting perineural invasion (PNI) and lymphovascular invasion (LVI) in rectal adenocarcinoma. Materials and MethodsA retrospective analysis was conducted on the preoperative clinical and magnetic resonance imaging (MRI) data of 102 patients with pathologically confirmed rectal adenocarcinoma. Based on pathological results, patients were divided into two groups: the PNI/LVI-positive group (with either or both PNI and LVI positive) and the PNI/LVI-negative group (both PNI and LVI negative). Using FireVoxel software, regions of interest (ROIs) were delineated to obtain ADC histogram parameters of the primary tumor, including ADC mean (ADC-mean), standard deviation, coefficient of variation, entropy, skewness, and the 1st, 5th, 10th, 25th, 50th, 75th, 90th, 95th, and 99th percentiles of ADC (ADC-1%, ADC-5%, ADC-10%, ADC-25%, ADC-50%, ADC-75%, ADC-90%, ADC-95%, ADC-99%). Differences in ADC histogram parameters, MRI assessment extramural venous invasion (mrEMVI) status, tumor location, mrT stage, and mrN stage between the PNI/LVI-positive and negative groups were analyzed. Parameters with statistically significant differences (P < 0.05) were selected through univariate analysis and used to construct a multivariate logistic regression model (ADC histogram model). Additionally, non-histogram parameters that were also statistically significant (P < 0.05) in univariate analysis were included in a multivariate logistic regression analysis to establish a combined predictive model. The predictive performance of the ADC histogram model and the combined model was evaluated using receiver operating characteristic (ROC) curve analysis, and the DeLong test was used to compare the differences in the area under the curve (AUC) between the models. ResultsSignificant differences were observed between the PNI/LVI-positive and negative groups in ADC-mean, standard deviation, ADC-1%, ADC-75%, ADC-95%, ADC-99%, and mrEMVI (P < 0.05). Among these continuous variables, ADC-99% had the highest diagnostic performance (AUC, sensitivity, and specificity were 0.835, 77.1%, and 86.6%, respectively). The combined model, constructed using ADC-mean, standard deviation, ADC-1%, ADC-75%, ADC-95%, ADC-99%, and mrEMVI, had an AUC, sensitivity, and specificity of 0.918, 89.6%, and 82.9%, respectively, outperforming the histogram model (AUC = 0.898) and individual whole-tumor ADC histogram parameters (AUC = 0.670 to 0.835). In addition to the combined model and the histogram model, there were statistically significant differences between the two models and the histogram parameters (P < 0.05). ConclusionsWhole-tumor ADC histogram parameters and imaging biomarkers (mrEMVI) can be used to predict the neurovascular status of rectal adenocarcinoma preoperatively. The predictive value is higher when both are combined. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.011 ObjectiveThis study aimed to develop a model based on deep transfer learning (DTL) features from preoperative multiparametric magnetic resonance imaging (MRI) to predict lymphovascular space invasion (LVSI) status in patients with endometrial carcinoma (EC). Materials and MethodsA retrospective analysis was conducted on clinical information and preoperative MRI images of 187 EC patients who were surgically and pathologically confirmed in our hospital from February 2016 to July 2023. The patients were randomly divided into a training set (131 patients) and a test set (56 patients) in a 7∶3 ratio. Regions of interest were delineated on axial T2-weighted imaging, diffusion-weighted imaging, apparent diffusion coefficient (ADC) maps, and contrast-enhanced T1-weighted imaging, manually. Subsequently, 12 DTL models were established using ResNet50, ResNet101, and DenseNet121 networks. Fusion models were then established using three decision-level fusion methods: mean, maximum, and minimum, with the best model selected as the final DTL model. A clinical model was established after screening clinical features through univariate and multivariate logistic regression analysis, and a DTL-clinical combined model was developed using logistic regression incorporating DTL and clinical features. The receiver operating characteristic curve was used to assess the diagnostic performance of the models for LVSI in EC patients, the area under the curve (AUC) was compared using the DeLong test. The calibration curve was used to analyze the goodness of fit of the models, and the decision curve was used to explore the clinical applicability of the models. ResultsIn the test set, the ResNet101 model based on the ADC images showed the highest AUC value of 0.850 [95% confidence interval (CI): 0.736 to 0.963] for diagnosing LVSI in EC patients. The fusion model established using the mean fusion method had the highest AUC value of 0.932 (95% CI: 0.868 to 0.996) in the test set, representing the best DTL model. Logistic regression analysis indicated that age was an independent risk factor for LVSI. The DTL-clinical combined model had an AUC of 0.934 (95% CI: 0.871 to 0.997) in the test set, with significantly better diagnostic performance than the clinical model [AUC: 0.554 (95% CI: 0.436 to 0.671), P < 0.001] and no statistical difference compared to the DTL model (P = 0.909). The combined model demonstrated good fit in both the training and test sets (Hosmer-Lemeshow test: P = 0.814 and 0.402, respectively) and offered greater clinical net benefit. ConclusionsThe DTL model based on preoperative multiparametric MRI, as well as the combined model integrating DTL features with clinical features, can effectively predict the LVSI status of EC patients, outperforming clinical models. DTL demonstrates excellent performance on our small-sample EC MRI data, providing important clinical assistance for preoperative LVSI prediction. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.012 ObjectiveTo explore the predictive value of radiomics models based on apparent diffusion coefficient (ADC) in evaluating the myometrial invasion depth of endometrial carcinoma (EC), providing a reliable evidence for clinicians to formulate treatment plans. Materials and MethodsRetrospective analysis of 155 patients with EC who underwent preoperative pelvic MR examination and were confirmed by pathology after operation from January 2016 to December 2023 in Beijing Luhe Hospital (superficial myometrial invasion = 114, deep invasion = 41), and randomly divided into training set (n = 124) and validation set (n = 31) in a 4∶1 ratio. The ITK-SNAP software was used to delineate the tumor regions layer by layer on the ADC maps, and the radiomics features were extracted, the extracted features were normalized. Pearson correlation coefficients (PCC) and least absolute shrinkage and selection operator (LASSO) were used to reduce features dimensionality, and the importance of the screened radiomics features was ranked according to the weight coefficient, the top 10 features were used to build radiomics models using three algorithms: logistic regression (LR), random forest (RF), and gradient boosting machine (GBM). The models were validated on the validation set. The performance of three radiomics models were evaluated by the receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA). The AUC values were compared using the DeLong test. ResultsThe AUC values of the LR, RF, and GBM models in predicting the invasion depth of endometrial carcinoma were 0.780 (95% CI: 0.762 to 0.804), 0.860 (95% CI: 0.846 to 0.879), and 0.860 (95% CI: 0.843 to 0.877), respectively. The AUC values of the RF and GBM were the highest and equal. The DeLong test showed that there was a statistically significant difference in AUC values between LR, RF, and GBM models (P = 0.017, 0.023), while there was no statistically significant difference in AUC values between RF and GBM models (P = 3.310). The calibration curve and DCA curve show that all three models have good fit and clinical practicality. ConclusionsThe radiomics models based on ADC map have good value in predicting the invasion depth of EC. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.013 ObjectiveTo explore the ability of a MRI based imaging histologic model to identify placenta percreta (PP). Materials and MethodsA retrospective study was conducted to collecting data from 80 cases of pregnant women who underwent placental MRI scanning and MRI indications pointing to PP in the Department of Radiology of the Third Affiliated Hospital of Zhengzhou University from January 2021 to December 2023, with surgical findings as the standard, including 48 cases of PP and 32 cases of non-PP. The region of interest was manually outlined on the axial, coronal and sagittal T2WI sequences, and the features of imaging histology were extracted. All patients were randomly divided into training and test sets in the ratio of 7∶3. The extracted imaging histology features were firstly subjected to Z-score regularization, then feature screening by t test, followed by calculation of Pearson correlation coefficients, and finally the least absolute shrinkage and selection operator algorithm was used. Selection operator algorithm for screening and dimensionality reduction of the features of the histology, and calculate the radiomics score. The optimal algorithm was selected from 7 different machine learning algorithms and used to construct an radiomics model. Univariate logistic regression analysis was performed on both clinical data and radiomics scores, revealing statistically significant differences. Subsequently, factors demonstrating significant differences were incorporated into multivariate analysis to identify independent risk factors (clinical information was used to construct clinical models). These factors were then visualized to construct a predictive combined model (nomogram). The receiver operating characteristic curve was plotted, and the efficacy of the model was compared by the indicators of area under the curve (AUC), sensitivity, specificity, and accuracy, and the calibration curve was used to evaluate the calibration degree of the model, and the decision curve analysis was used to assess the effectiveness of the model. The calibration curve was used to evaluate the calibration degree of the model, and the decision curve analysis was used to assess the clinical utility value of the model. ResultsThe multivariate analysis identified two independent risk factors: parity and radiomics score. Parity demonstrated a protective effect with an odds ratio of 0.272 [95% confidence interval (CI): 0.151 to 0.492], while the radiomics score showed a strong positive association with an exceptionally high odds ratio of 1 934.105 (95% CI: 118.985 to 31 445.149). The AUC values for the imaging histology model and the clinical model in the training set were 0.948 (95% CI: 0.884 to 1.000) and 0.723 (95% CI: 0.596 to 0.850), respectively, and in the test set were 0.828 (95% CI: 0.601 to 1.000) and 0.676 (95% CI: 0.474 to 0.878). The AUC value of the imaging histology-clinical model in the training set was 0.962 (95% CI: 0.906 to 1.000). The results of DeLong test showed that there were significant differences in the training set, both between the clinical model and the imaging histology model as well as between the clinical model and the imaging histology-clinical model (P < 0.05), but the differences between the imaging histology model and the imaging histology-clinical model were not statistically significant (P > 0.05). Both the radiomics model and the radiomics-clinical model had good calibration and clinical application value in the test set. ConclusionsImaging histology-clinical modeling has better diagnostic efficacy and can be used as a modality for the identification of PP. It provides a reliable foundation for clinicians in determining the timing and method of pregnancy termination, thereby aiding in the formulation of informed clinical decisions. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.014 ObjectiveR2^* value and proton density fat fraction (PDFF) of lumbar spine in middle-aged and elderly women were measured by mDixon-Quant technique to evaluate iron deposition and fat content in bone marrow, explore the correlations between R2^*, PDFF and bone mineral density (BMD) and evaluate the diagnostic value for osteoporosis (OP). Materials and MethodsA total of 105 middle-aged and elderly women underwent lumbar spine MRI using the mDixon-Quant sequence. R2^* values and PDFF of L1 to L5 vertebrae were measured, and BMD of L1 to L5 vertebrae were obtained after post-processing of lumbar spine quantitative computed tomography (QCT) scans. General clinical data were recorded. The trend of changes in values of R2^*, PDFF, and BMD of L1 to L5 vertebrae were evaluated. Partial correlation analysis was performed to assess associations among R2^*, PDFF, and BMD. The diagnostic efficacy of R2^* and PDFF on OP was evaluated using receiver operating characteristic (ROC) curve analysis and DeLong^,s test was used to compare the diagnostic efficacy. ResultsR2^* value and BMD gradually decreased from lumbar 1 to lumbar 5 vertebrae (P_trend < 0.05), and the PDFF gradually increased (P_trend < 0.05). After age adjustment, lumbar R2^* demonstrated a negative correlation with PDFF (r = -0.227, P = 0.020), and a positive correlation with lumbar BMD (r = 0.332, P < 0.001). The area under curve (AUC) of R2^* and PDFF diagnosis for OP were 0.792 and 0.702, respectively, difference showed no statistical significance (P = 0.07). The AUC of R2^* combined with PDFF in diagnosis of OP was 0.804, 95% confidence interval (95% CI) was (0.702 to 0.865), sensitivity was 81.8% and specificity was 73.5%, and the diagnostic efficacy was better than that of PDFF (P = 0.01). R2^* was associated with BMD and PDFF. ConclusionsThere were physiological gradient changes in R2^*, PDFF and BMD of lumbar spine in middle-aged and elderly women. The efficacy of R2^* in diagnosis of OP was comparable to that of PDFF, R2^* combined with PDFF achieved the highest diagnostic efficacy on OP. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.016 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.017 ^,s understanding of the involvement of the Hb in the pathogenesis of depression, but also helped to improve the diagnosis, treatment and prognosis of depression. In this paper, we provide a concise overview of the MRI, segmentation and imaging changes of the Hb in depression, discuss the challenges of current research, and provide new ideas for early diagnosis and personalized treatment strategies for depression. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.018 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.019 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.020 ^,s disease (PD) is a late-onset neurodegenerative disorder with an increasing incidence year by year. The dysregulation of iron homeostasis in the brain is regarded as one of the significant pathological features of PD. In recent years, remarkable progress has been made in the field of neuroimaging, which has opened up new vistas for the clinical diagnosis of PD and the in-depth exploration of its pathological mechanisms. Against this backdrop, quantitative susceptibility mapping (QSM) technology, as an advanced magnetic resonance imaging modality, utilizes the phase information in magnetic resonance imaging to achieve an accurate quantitative assessment of iron content in brain tissue. This paper reviews the imaging principle of QSM technology and its potential value and application prospects in the diagnosis of PD, aiming to provide new insights for PD diagnosis. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.021 ^,s lymphatic system, examines factors modulating its activity, and evaluates current detection methodologies.Furthermore, we synthesize emerging evidence highlighting the involvement of this system in the pathogenesis of various CNS disorders, with the goal of adwancing neuroimaging strategies and therapeutic interventions for these conditions. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.022 ^, cognitive function. This review reviews the GS-related studies of MS and focuses on its relationship with cognitive impairment, in order to provide new insights into the pathophysiological process, diagnosis, and treatment of MS. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.023 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.024 ^, quality of life and life safety. Therefore, accurately classifying brain tumors is of crucial importance for formulating targeted treatment plans and evaluating the prognosis of patients. In recent years, the rapid development of deep learning technology has opened up new avenues in the field of medical image analysis, and the deep residual network (ResNet) and its derived variants have demonstrated excellent performance in image classification tasks, bringing new breakthroughs in brain tumor MRI classification. In this paper, the optimization strategy of the network model based on deep residual networks in brain tumor MRI classification is discussed in depth, firstly, the development of deep residual networks is introduced, followed by a detailed analysis of the current applications of deep residual networks and their derived variants on brain tumor MRI images. Finally, the current challenges faced in this field are pointed out, and the future research directions are prospected, aiming to provide comprehensive references and ideas for related research, and to promote the further development and application of deep residual networks in brain tumor MRI classification, so as to improve the accuracy and efficiency of brain tumor diagnosis, and to provide more powerful support for clinical treatment. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.025 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.026 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.027 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.028 129Xe magnetic resonance imaging (HP ¹²⁹Xe MRI) in the diagnosis, classification, and efficacy evaluation of CTD-ILD. Moreover, it explores the limitations of current technologies and future development directions, aiming to provide new ideas for optimizing the clinical management of CTD-ILD. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.029 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.030 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.031 ^,s health. Early screening, diagnosis, and treatment can significantly improve the survival rate of breast cancer patients. In recent years, with the rapid development of big data and computer learning algorithms, artificial intelligence (AI) technology has been widely used in the field of medical imaging research, making disease diagnosis more efficient and accurate. Magnetic resonance imaging (MRI) has high resolution for soft tissues and is also sensitive in diagnosing breast lesions, and is widely used in clinical practice. A series of achievements have been made in the application of AI technology to process breast MRI data, which has improved the accuracy of breast cancer diagnosis to varying degrees. The purpose of this review is to summarize the latest application and research progress of AI technology in the diagnosis and treatment of breast cancer, so as to provide a valuable reference for clinicians to apply AI technology in the diagnosis and treatment of breast cancer, and help promote the further development of this field. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.032 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.033 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.034 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.035 ^,s awareness and understanding of HCC, particularly early-stage HCC, and to provide valuable guidance for radiologists and clinicians in accurately assessing, making treatment decisions, and prognostic evaluations for HCC patients. Furthermore, it seeks to offer researchers a more comprehensive comparative perspective to help more patients benefit from clinical diagnosis and treatment. ]]> Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.036 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.037 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.038 Thu,20 Mar 2025 00:00:00 GMT http://www.med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2025.03.039 Thu,20 Mar 2025 00:00:00 GMT