http://med-sci.cn/cgzcx/en/contents_list.asp?issue=202308 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.1674-8034.2023.08.001 ObjectiveTo develop Cox proportional hazards regression model for prediction of the recurrence risk of non-cardiogenic anterior circulation ischemic stroke during 2-year follow-up based on radiomic approach by extracting texture features from a symptomatic middle cerebral artery (MCA) plaque, and to further evaluate the developed model performance. Materials and MethodsIn our retrospective study from January 2019 to January 2020, a total of 82 eligible patients with first-ever ischemic stroke and middle cerebral artery >50% luminal stenosis underwent baseline intracranial high-resolution magnetic resonance imaging (HRMRI) followed up for 2 year when recurrent non-cardiogenic ischemic stroke in the territory of MCA served as an endpoint event were finally enrolled in current analyses. HRMRI-based radiomic features were manually extracted from an index MCA plaque using 3D-Slicer software package. Multivariable Cox regression analysis was used to develop the predicting model where multi-dimensional parameters were selected by LASSO (least absolute shrinkage and selection operator) regression analysis, for which the performance was further assessed with respect to its calibration, discrimination. ResultsOf which, 19 cases with endpoint events occurred during the 2-year follow-up period with 13.9 per 100 person-years of the recurrence rate of ischemic stroke. Multivariable Cox regression included top 4 parameters with nonzero coefficients defined by logλmin of LASSO regression (i.e., second-order texture feature, plaque hemorrhage, entropy and low density lipoprotein cholesterol). In the prediction model adjusting for baseline covariants, the gray level co-occurrence matrix was found to be the major contributor to the event endpoint [adjusted hazard ratio (aHR): 5.379, 95% confidence interval (CI): 1.716-16.859, P=0.004, weight=40.23%]. However, plaque hemorrhage (aHR: 2.226, 95% CI: 0.821-6.040, P=0.116, weight=20.86%), entropy (aHR: 1.324, 95% CI: 0.769-2.278, P=0.311, weight=16.13%) and low density lipoprotein cholesterol (aHR: 1.485, 95% CI: 0.877-2.516, P=0.142, weight=22.78%) just showed a trend towards significance. Additionally, the developed prediction model showed a good discrimination with a C-index of 0.8296 and good calibration. ConclusionsThe findings suggest that our developed prediction model can target a potential sub-population at high risk of recurrent ischemic stroke in which gray level co-occurrence matrix may account for the major contributing, although this must be confirmed in future. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.002 ObjectiveTo observe the effect of low-grade glioma (LGG) tumor and peritumoral white matter changes in the occurrence of glioma-associated epilepsy (GAE) by magnetic resonance diffusion white matter quantification analysis. Materials and MethodsThe clinical and imaging information of patients with LGG confirmed by pathology who underwent diffusion spectrum imaging (DSI) in the First Affiliated Hospital of Zhengzhou University from December 2018 to December 2020 was retrospectively analyzed. A total of 102 patients with WHO Ⅱ low-grade gliomas were enrolled, including 37 patients with preoperative GAE and 65 patients without preoperative GAE. Diffusion tensor imaging (DTI), neurite orientation dispersion and density imaging (NODDI) and mean apparent propagator (MAP) metrics. ITK snap was used to draw tumor and peritumoral regions of interest (ROI) were based on b=0 diffusion images. FAE was used to perform histogram feature extraction, volume calculation of interest, and morphological feature extraction. After single parameter analysis and collinear analysis, logistic regression models were constructed based on each diffusion model and ROIs, and the DeLong test was used to compare the performance of models. ResultsThere is a statistical difference in age between GAE groups (P=0.004). The incidence of GAE in patients with tumors located in the right hemisphere and trans hemisphere growth was lower than that in patients with tumors located in the left hemisphere, with a statistically significant difference (P=0.002). GAE predictive clinical-imaging model is constructed by age and hemispheric location of tumor, with AUC=0.779. The tumor and peritumoral volumes in the GAE group were significantly smaller than those in the non-GAE group (P＜0.05). There was no statistical difference in the morphological characteristics of the tumor area. The smaller the long and short diameters of the peritumoral area, the smaller the surface area, the more likely it is to be spherical, with higher incidence of GAE, and the difference is statistically significant (P＜0.05); the AUC value of constructing a GAE logistic regression model using morphological features of the peritumoral area can reach 0.730. Histogram features of quantitative parameters of diffusion models in tumor and peritumor areas with differences between GAE and non-GAE groups (P＜0.05), which including DTI_FA_Maximum, NODDI_ODI_90Percentile, MAP_NG_10Percentile. NODDI_ODI_90Percentile value of the peritumoral area in the GAE group was higher than that in the non-GAE group. The remaining features in tumor and peritumoral areas of GAE group were lower than the non-GAE group. Logistical models based on tumor area and peritumoral area showed no statistically significant difference in predictive performance of GAE, while the tumor area model had slightly higher performance than the peritumor area model. The combined model constructed by combining the features based on tumor area and peritumoral area have the highest performance, with a statistically significant difference compared to model based on the peritumor area (P=0.02) only. In the combined model, tumor features account for the majority, and the tumor DTI_MD_10Percentiles has the highest OR value, which positively correlated with the occurrence of GAE. The highest OR value among all models is the NODDI_ODI_Mean based on peritumoral feature, which is positively correlated with the occurrence of GAE. The MAP model has slightly higher performance than the DTI and NODDI models based on the individual diffusion models in the tumor area and peritumor area. There was no statistically significant difference in the predictive performance of GAE among clinical-imaging model, peritumoral morphological model, and combined models based on diffusion parameters. ConclusionsQuantitative analysis of white matter is a promising way to predict the occurrence and mechanism of GAE. White matter damage in the tumor area, accompanied by increased dispersity or relatively intact white matter in the peritumoral area, increases the risk of GAE. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.003 ObjectiveTo predict the recurrence of glioma after surgery through preoperative conventional magnetic resonance imaging signs, so as to help clinicians planning more accurate surgical resection range before surgery. Materials and MethodsThis study is a retrospective study, involving 123 patients with postoperative recurrence of glioma confirmed by pathology in two centers, all of whom have complete preoperative and postoperative MRI images of recurrence. Two radiologists established a plane rectangular coordinate system with the center of the preoperative and postoperative glioma as the midpoint, thus dividing the tumor into four quadrants, respectively evaluating the MR imaging signs of the four quadrants before surgery and whether the quadrant recurred after surgery, and performing interrater reliability (IRR) analysis on the two radiologists; 18 MRI manifestations of Visually Accessible Rembrandt Images (VASAIR) signs were selected as the predictive index variables. The binary logistic regression is used as a classifier to build the prediction model, and the cross-validation method is used to verify the prediction ability of the model, where the training set∶validation set=3∶1; Select meaningful variables to establish a nomogram, and use concordance index curve and decision curve analysis (DCA) to verify. ResultsOne hundred and twenty three patients were divided into four quadrants, a total of 492 quadrants. They were randomly divided into training set (129 non-recurrent quadrants and 240 recurrent quadrants) and validation set (43 non-recurrent quadrants and 80 recurrent quadrants). There were statistically significant differences in the enhancement quality (P=0.03), unenhanced diameter line (P＜0.01), deep white matter invasion (P=0.02), unenhanced area crosses midline (P=0.04), ependymal invasion (P＜0.01), the T1WI/fluid-attenuated inversion-recovery (FLAIR) (P=0.02). Further establish logistic regression model. The area under the receiver operating characteristic (ROC) curve in the training set is 0.7642 (P＜0.05), and the Kappa value is 0.38. The area under the ROC curve in the validation set data is 0.8493 (P＜0.05), and the Kappa value is 0.56. ConclusionsEnhancement quality, unenhanced diameter line, deep white matter invasion, unenhanced area crosses midline, ependymal invasion, and T1WI/FLAIR in the VASAIR feature concentration can predict glioma recurrence and recurrence site (quadrant) before surgery, which is helpful for neurosurgeons to make surgical plans. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.004 ObjectiveTo explore the value of multi-sequence radiomics features and clinical related parameters in predicting isocitrate dehydrogenase 1 (IDH1) gene mutations in gliomas. Materials andMethodsA total of 81 patients with gliomas confirmed by histopathology and containing IDH1 gene mutation status information were analyzed retrospectively. Five types of images of T2WI, T1WI, diffusion weighted imaging (DWI), apparent diffusion coefficient (ADC), and contrast enhancement MRI (CE-MRI) were applied for radiomics feature extraction. Each sequence can extract 107 radiomics features. The above features were subjected to single factor rank sum test, correlation analysis, and least absolute shrinkage selection operator (LASSO) dimensionality reduction screening. Multivariate logistic regression was used to establish various sequence models and multiple sequence fusion models for the remaining features, including T2WI model, T1WI model, DWI model, ADC model, CE-MRI model, and multiple sequence radiomics model. Finally, a combined model is established by combining the Radscores output from the multi sequence radiomics model with the clinical multivariate model. The above models used receiver operating characteristic (ROC) curves to analyze the predictive performance of each model, and compared the differences in area under the curve (AUC) using DeLong non parametric tests. In addition, decision curve analysis (DCA) was used to evaluate the clinical benefits of multiple sequence radiomics models and combined models in identifying IDH1 gene mutation status. ResultsThe combined model showed the best performance in predicting IDH1 gene mutations in gliomas (AUC: 0.928). The AUC values of multiple sequence radiomics models were higher than those of T2WI, DWI, and ADC models (0.865 vs. 0.752, 0.656, 0.631, P＜0.05, respectively); The AUC value of the combined model was higher than that of T2WI, T1WI, T1 enhanced, and multi sequence radiomics models (0.928 vs. 0.752, 0.827, 0.829, 0.865, P＜0.05, respectively); However, there was no statistically significant difference in AUC values between the combined model and the clinical model (0.928 and 0.880, respectively, P＞0.05). The decision curve analysis showed that the combined model had higher clinical benefits in identifying IDH1 gene mutations with sequence radiomics models. ConclusionsThe combination of multi-sequence radiomics features, clinical and MRI imaging features has important value in preoperative differentiation of IDH1 gene mutations in gliomas. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.005 ObjectiveTo investigate the value of a convolutional neural network model based on multi-sequence MRI to predict the promoter methylation status of O⁶-methylguanine-DNA-methyltransferase (MGMT) in glioma. Materials and MethodsRetrospective analysis of clinical and MRI data of 161 patients with glioma confirmed by surgical pathology from November 2015 to June 2022 at Ningxia Medical University General Hospital, including 80 cases of MGMT promoter methylation type and 81 cases of unmethylated type. T2WI, T2 fluid-attenuated inversion recovery (T2-FLAIR) and contrast enhanced T1‐weighted imaging (CE-T1WI) of preoperative MRI were collected, and regions of interest (ROI) were outlined after preprocessing of all images. The images were randomly divided into training and validation sets according to 7∶3 after labeling. A 34-layer-residual neural network (ResNet34) was used to build T2WI, T2-FLAIR, enhanced T1WI and multiple sequence fusion models T2-net, T2f-net, TC-net and TS-net, respectively, to predict the methylation status of MGMT promoters. The area under the receiver operating characteristic (AUROC), area under the precision-recall curve (AUPRC), accuracy, specificity and sensitivity were used to assess model efficacy, and the predictive power was compared between models by DeLong test. ResultsAll four prediction models T2-net, T2f-net, TC-net, and TS-net had good prediction efficacy, and the AUROC values of TS-net were higher than those of T2-net, T2f-net, and TC-net (training set: 0.930 vs. 0.859, 0.877, 0.920; validation set: 0.910 vs. 0.812, 0.840, 0.854). The AUPRC values of TS-net were higher than those of T2-net, T2f-net, and TC-net (training set: 0.912 vs. 0.860, 0.864, 0.908; validation set: 0.896 vs. 0.796, 0.826, 0.839). The AUROC values of TS-net in the validation set were all higher than those of T2-net, T2f-net, and TC-net, and the differences were all statistically significant. In addition, the differences in the training set were statistically significant compared with T2-net and T2f-net (DeLong test, P＜0.05). ConclusionsConvolutional neural network models based on multi-sequence MRI fusion can accurately and non-invasively predict the MGMT methylation status of glioma, which is superior to single-sequence models and provides a reliable basis for guiding clinical treatment decisions and assessing the prognosis of glioma patients. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.006 ObjectiveTo analyze clinical and imaging features of acute ischemic stroke (AIS) in patients with Trousseau^,s syndrome, and to explore the prognostic factors. Materials and MethodsA retrospective analysis was conducted on 65 patients with malignant tumor complicated with Trousseau^,s syndrome and their clinical and imaging features were summarized. Based on their modified Rankin Scale (mRS) score, patients were divided into good prognosis group (mRS＜3, n=35) and poor prognosis group (mRS≥3, n=30) . The features of the two groups were compared using independent sample t‑test, Mann‑Whitney U test or χ² test. Indices with statistical significance in univariate binary logistic regression analysis were included in multivariate binary logistic regression analysis for prognostic analysis. Besides, patients were divided into asymptomatic cerebral infarction group (n=25) and symptomatic cerebral infarction group (n=40), and multivariate binary logistic regression analysis was performed to explore the independent risk factors of symptomatic cerebral infarction. ResultsPatients with Trousseau^,s syndrome showed increased level of D-dimer, small and numerous infarcts in different vascular supply areas. Compared with the good prognosis group, the poor prognosis group showed higher proportion of three territory sign in MRI, greater number of cerebral infarctions, higher National Institutes of Health Stroke Scale (NIHSS) score at admission, higher level of D-dimer, lower level of hemoglobin (Hb) and red blood cell count (P＜0.05). Multivariate analysis showed that increased D-dimer (OR=5.094, 95% CI: 1.726-15.039) was an independent prognostic factor for poor prognosis in such patients. Infarct volume was an independent risk factor for symptomatic cerebral infarction (OR=1.227, 95% CI: 1.025-1.047). ConclusionsPatients with AIS complicated with Trousseau^,s syndrome were characterized by hypercoagulation status and small and numerous infarcts in different vascular supply areas; increased D-dimer level is an independent prognostic risk factor and infarct volume was an independent risk factor for symptomatic cerebral infarction. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.007 ObjectiveTo analyze the cerebral characteristics of patients with pruritus by resting-state functional magnetic resonance imaging (rs-fMRI). Materials and MethodsForty-two patients with eczema were recruited in the observation group, and 44 healthy subjects were included in the control group. The visual analogue scale (VAS), 12-Item Pruritus Severity Scale (12-PSS), Pittsburgh Sleep Quality Index and Self-rating Anxiety Scale were recorded in the observation group. The different values of fraction amplitude of low frequency fluctuation (fALFF) were compared between the observation and control group through rs-fMRI program. Subsequently, functional connectivity (FC) was analyzed using the brain regions with significant differences between groups as seed points. ResultsThere were no significant differences in average age, sex ratio, years of education between the two groups (P＞0.05). Compared with the control group, the fALFF values of the left precentral gyrus, left postcentral gyrus, left supplementary motor area, and left anterior cingulate cortex in the observation group were increased. The FC values between the left precentral gyrus and bilateral inferior temporal gyrus, bilateral fusiform gyrus, bilateral hippocampus, bilateral middle temporal gyrus, bilateral inferior occipital gyrus, and bilateral lingual gyrus were decreased in the observation group. The FC values between left supplementary motor area and bilateral inferior temporal gyrus, bilateral superior temporal gyrus, right hippocampus and right insula were decreased in the observation group. The 12-PSS score was positively correlated with fALFF value of left precentral gyrus (r=0.59, P＜0.01) and left postcentral gyrus (r=0.52, P＜0.01), and was positively correlated with VAS score (r=0.33, P＜0.05) in the observation group. ConclusionsThe spontaneous activity of the left somatosensory and motor area are abnormally increased in patients with pruritus, and there is a correlation between given cerebral regions and clinical scales, which provides potential neurobiological markers for the future study of pruritus. The synchronous decrease of FC between left somatosensory motor area and occipital or temporal lobe is another important brain network feature in patients with pruritus. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.008 ObjectiveTo compare the image quality and application value of water-fat separation (Dixon) and spectral attenuated inversion recovery (SPAIR) MRI fat suppression techniques in thyroid-associated ophthalmopathy (TAO). Materials and MethodsThis cross-sectional retrospective study included 73 TAO patients (146 eyes) who underwent orbital MRI including Dixon and SPAIR fat suppression sequences from 2020 to 2022. We compared the fat suppression effect, artifacts, noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) and diagnostic efficacy of TAO activity between T2WI-Dixon aqueous phase and T2WI-SPAIR images. ResultsIn terms of the overall image quality of MRI, the proportion of T2WI-Dixon aqueous phase images superior to T2WI-SPAIR images was higher (67.1% vs. 13.7%). The fat suppression effect, artifact score and noise of T2WI-Dixon aqueous phase were significantly lower than those of T2WI-SPAIR images (Z: 6.632-10.473, P＜0.001), SNR and CNR were significantly higher than those of T2WI-SPAIR images (Z: -8.792--10.482, P＜0.001). The correlation between extraocular muscle/white matter signal intensity ratio (EOM-SIR), lacrimal gland/white matter signal intensity ratio (LG-SIR) and clinical activity score (CAS) in T2WI-Dixon water phase images was higher than that in T2WI-SPAIR image (0.613 vs. 0.520; 0.282 vs. 0.251). The AUC of EOM-SIR and LG-SIR of T2WI-Dixon water phase in evaluating TAO activity was higher than that of T2WI-SPAIR image (0.872 vs. 0.772; 0.673 vs. 0.604), the difference was statistically significant (Z=4.070, P＜0.001; Z=2.174, P=0.030). ConclusionsThe image quality of T2WI-Dixon was better than that of T2WI-SPAIR. In addition, T2WI-Dixon water has a stronger ability to evaluate the inflammatory edema of TAO retrobulbar tissue than T2WI-SPAIR images, and has higher diagnostic efficiency for TAO diagnosis and activity staging. Dixon technique provides an excellent fat suppression technique for MRI examination of TAO patients. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.009 ObjectiveTo explore the value of whole volume histogram features from synthetic MRI (SyMRI) in the diagnosis of benign and malignant breast tumors. Materials and MethodsClinical and imaging data of 186 patients with breast lesions were retrospective collected from October 2019 to November 2021. All patients were confirmed by puncture biopsy and/or surgical pathology, and preoperative conventional MRI and SyMRI scans were performed. The independent samples t-test or Mann-Whitney U test was chosen to compare quantitative parameters between benign and malignant breast lesions. Multivariate logistic regression model was developed based on the univariate result, and the corresponding ROC curves were obtained with AUC, sensitivity and specificity. ResultsA total of 150 patients with breast lesions (166 lesions in total) were enrolled. Multivariate analysis showed that T1-90th (P=0.002), T1-entropy (P=0.001) and proton density-kurtosis (P=0.014) were independent predictors for the differentiation of benign and malignant breast lesions. The AUC of differentiating benign and malignant breast lesions by multivariate logistic regression model was 0.89 (95% CI: 0.84-0.94), with the sensitivity of 85.15% and the specificity of 81.97%. ConclusionsThe whole volume histogram parameters from SyMRI can provide a basis for accurate diagnosis of identifying benign and malignant breast lesions. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.010 ObjectiveTo investigate the clinical value of imaging radiomics features based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) combined with apparent diffusion coefficient (ADC) in predicting the expression level of Ki-67 in breast cancer. Materials and MethodsMRI images of 234 patients with breast cancer confirmed by pathology from December 2018 to December 2021 were retrospectively analyzed. According to postoperative immunohistochemical results, the tumors were divided into the Ki-67 high expression group (n=180) and low expression group (n=54). 1906 radiomics features were extracted form the first phase of the DCE-MRI by semi-automatic separation method. Using intraclass correlation coefficient (ICC), the linear correlation analysis and the least absolute shrinkage and selection operator (LASSO), four features were selected to construct the radiomics model. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic effectiveness of the radiomics, average ADC values and combined models. Calibration curves and decision curves were used to evaluate the clinical usefulness of the predictive model. ResultsA total of 1906 features were extracted from the tumor body, 207 features were excluded by ICC analysis, 1626 features were excluded by linear correlation analysis, and the remaining 73 features were selected by LASSO dimensionality reduction to select 4 optimal omics features. Four radiomics features and the average ADC values were significantly different between two groups (P＜0.05). Radiomics model, the average ADC value and the combined model predicted that the area under the curve (AUC) of Ki-67 high expression were 0.820, 0.676 and 0.856, respectively, with statistically significant differences each other (P＜0.05). The combined model had the best predictive efficiency for Ki-67 expression, and its AUC, sensitivity and specificity were 0.856, 88.3% and 74.1%, calibration curves and decision curves showed that the combined model had clinical application value. ConclusionsThe combined model which constructed by the images radiomics features based on DCE-MRI and the average ADC values has high efficacy in predicting Ki-67 expression in breast cancer.The combined model is superior to the radiomics model and the average ADC value. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.011 ObjectiveTo explore the relationship between MRI enhanced features, pathological basis and prognosis of intrahepatic mass-forming cholangiocarcinoma (IMCC). Materials and MethodsThe data of intrahepatic cholangiocarcinoma (ICC) patients admitted to Aviation General Hospital of China Medical University from January 2008 to June 2020 were retrospectively analyzed. A total of 69 cases of IMCC were included in the study including 62 males and 7 females. According to the different MRI enhancement methods of IMCC, they were divided into 3 groups: progressive enhancement group (25 cases), peripheral enhancement group (24 cases) and rich blood supply group (20 cases). The MRI findings and pathological characteristics of the three groups were analyzed, and the survival rate of the three groups was compared. ResultsThe enhancement degree of MRI was related to the proportion of tumor cells, fibrous tissues, and other tissues ( such as mucus and tissue necrosis). The median survival time of all patients was 35.9 months (95% CI: 17.4-54.6), the 1-year cumulative survival rate was 75.9%, the 3-year cumulative survival rate was 46.7%, and the 5-year cumulative survival rate was 23.4%. The 1-year, 3-year, and 5-year survival rates of patients in progressive enhancement group were 75.0%, 41.3%, and 20.6%, those in peripheral enhancement group were 71.8%, 47.9%, and 18.0%, and those in rich blood supply group were 83.3%, 51.3%, and 32.1%, respectively. Log rank test showed that there was no significant difference in survival rate among the three groups (χ²=1.117, P=0.572). There was no significant difference in 1-year survival rate among the three groups (F=0.50, P=0.616), and no significant difference in 3-year survival rate (F=0.632, P=0.725). There was a significant difference in 5-year survival rate (F=5.93, P=0.007). The rich blood supply group was superior to the progressive reinforcement group (P=0.013) and the peripheral reinforcement group (P=0.001). ConclusionsDifferent enhancement modes of MRI in IMCC have their corresponding pathological basis. There may be a correlation between different reinforcement methods and prognosis, and blood supply rich ICC may have an advantage in 5-year survival rate. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.012 ObjectiveTo investigate the correlation between the microsatellite instability (MSI) status and each parameter of diffusion kurtosis image (DKI) in rectal cancer, and to provide imaging detection indicators for evaluating the MSI status before and after rectal cancer treatment. Materials and MethodsEighty eight patients with a pathologically definite diagnosis of rectal cancer were included for analysis. All patients underwent MRI examination within one week before radical resection of rectal cancer surgery. The examination sequence contained DKI imaging. The obtained data were imported into the dedicated software to acquire DKI parameters such as mean kurtosis (MK), axial kurtosis (Ka), radial kurtosis (Kr), mean diffusion (MD), axial diffusion (Da), radial diffusion (Dr), fractional anisotropy (FA), and postoperative pathobiological characteristics. These parameters were used for statistical analysis. Intra-class correlation coefficient was used to evaluate the measurement consistency between two observers. The Kolmogorov-Smirnov test was to assess the normal distribution of DKI parameters. Spearman correlation coefficient was employed to examine the correlation between each quantitative parameter of DKI and MSI and microsatellite stability (MSS). Spearman correlation coefficient was used to compare the correlation between each quantitative parameter of DKI and MSI and MSS. The ROC curve analysis was performed to analyze each parameter of DKI associated with the presence of MSI to observe its value in predicting MSI. The DeLong test was utilized to compare the statistical differences in the AUC of each parameter. P values less than 0.05 were considered statistically significant. ResultsThe correlation coefficient values between MSI and the DKI parameters were as follows: 0.258 (95% CI: 0.122-0.386) for Da, 0.346 (95% CI: 0.191-0.476) for Dr, -0.276 (95% CI: -0.421--0.118) for Ka, and -0.260 (95% CI: -0.383--0.139) for MK. There was indeed a weak positive correlation observed between MSI and Da as well as Dr, while a weak negative correlation was found between Ka and MK. However, no significant correlation was observed between MSI and MD, FA, or Kr (P＞0.05). The AUC values for Da, Dr, Ka, and MK in diagnosing MSI in rectal cancer were 0.759 (95% CI: 0.654-0.865), 0.847 (95% CI: 0.749-0.945), 0.777 (95% CI: 0.651-0.902), and 0.758 (95% CI: 0.665-0.856), respectively. The corresponding cut-off values were 0.65, 0.68, 0.55, and 0.70. ConclusionsThere is a correlation between MSI status and DKI parameters in rectal cancer, and they have some predictive value for it. This correlation is expected to make DKI parameters an optional method for predicting MSI status. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.013 ObjectiveTo compare the value between the radiomics models based on zoomed imaging technique with parallel transmission diffusion weighted imaging (ZOOMit DWI) and readout segmentation of long variable echo-trains (RESOLVE) DWI for the diagnosis of clinically significant prostate cancer (csPCa). Materials and MethodsA total of 168 patients were included in this retrospective study, including 83 cases of csPCa and 85 cases of non-csPCa. The patients were grouped randomly into a training set (n=117) and a test set (n=51) in a ratio of 7∶3. Optimal radiomics features were selected by using Pearson correlation coefficient (PCC) method, analysis of variance (ANOVA) and least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation in the training set. Logistic regression was used to develop the models. The single sequence radiomics models were built to predict csPCa including ZOOMit DWI, ZOOMit apparent diffusion coefficient (ADC), RESOLVE DWI and RESOLVE ADC. The bi-parametric MRI (bpMRI) radiomics models was built combining DWI sequence with better diagnostic performance and T2-weighted imaging (T2WI). The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic performance of the radiomics models. The DeLong test was performed to statistically compare areas under the curve (AUC). ResultsIn the test group, ZOOMit DWI had higher AUC than RESOLVE DWI (0.917 vs. 0.851, P=0.022); ZOOMit ADC had higher AUC than RESOLVE ADC, of borderline statistical significance (0.948 vs. 0.871, P=0.052). The bpMRI radiomics models was established based on T2WI, ZOOMit DWI and ZOOMit ADC. The AUC of the bpMRI radiomics model was 0.937 in the test set, which was significantly higher than that of prostate-specific antigen (PSA) (0.792, P=0.012). ConclusionsThe radiomics models based on the ZOOMit DWI sequence had better diagnostic performance for csPCa than those based on the RESOLVE DWI sequence. The bpMRI radiomics model combined ZOOMit DWI sequence and T2WI showed great diagnostic value for csPCa. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.014 ObjectiveTo develop and test a model for predicting the efficacy of concurrent chemoradiotherapy for stage ⅡB-Ⅲ cervical cancer based on quantitative MRI, pathology, and blood cell parameters. Materials and MethodsFrom March 2020 to June 2022, clinical data from 151 cervical cancer patients at the North China University of Science and Technology Affiliated Hospital were analyzed retrospectively, and data from 93 cervical cancer patients at the Hebei General Hospital for Veterans were used for model external validation. To screen for risk factors associated with the efficacy of concurrent chemoradiotherapy, least absolute shrinkage and selection operator (LASSO) regression was used. The value of a risk factor model for the efficacy of concurrent chemoradiotherapy was evaluated using the consistency index (C-index), calibration curve, mean absolute error (MAE), and decision curve analysis (DCA). ResultsLASSO regression analysis revealed that elevated volume transport constant (K^trans), apparent dispersion coefficient (ADC), and perfusion-related volume fraction (f) were independent factors for objective remission (OR) following concurrent chemoradiotherapy. High International Federation of Gynecology and Obstetrics (FIGO) staging, lymph node metastasis, elevated extravascular extracellular volume ratio (V_e), elevated slow ADC (D), and elevated monocyte to lymphocyte ratio (MLR) were all independent risk factors for OR after concurrent chemoradiotherapy. The C-index and MAE of model F (which included FIGO staging, lymph node metastasis, K^trans, V_e, ADC, D, f, and MLR) were 0.984 and 0.033, respectively, which were higher than those of model S (which included K^trans, V_e, ADC, D, and f; 0.940, 0.020) and model T (which included FIGO staging, lymph node metastasis, and MLR; 0.897, 0.020). The calibration curves showed that the calibration curve for model S overlapped with the ideal curve slightly better than the calibration curves for models F and T. Over the entire risk threshold range, the DCA showed that model F had a higher net benefit than model S and model T. Model F had a higher C-index (0.996) than model S (0.942) and model T (0.917) and a lower MAE (0.017) than model S (0.043) and model T (0.043), according to the model validation results. The calibration curves showed that the calibration curves for model F and model S overlapped with the ideal curve more closely than model T. Over the entire risk threshold range, the DCA showed that model F had a higher net benefit than model S and model T. ConclusionsFIGO staging, lymph node metastasis, K^trans, V_e, ADC, D, f, and MLR are associated with the efficacy of concurrent chemoradiotherapy for stage ⅡB-Ⅲ cervical cancer. The model based on the above indicators can aid in predicting the efficacy of concurrent chemoradiotherapy, and its efficacy is higher than that of the MRI-only quantitative parameter model and the pathology and blood cell parameter model. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.015 ObjectiveTo explore the diagnostic value of machine learning model based on MRI T2WI imaging features for placenta accrete spectrum disorders (PAS). Materials and MethodsThe imaging data of 130 patients who underwent MRI examination and later caesarean section due to suspected placenta accretion were retrospectively analyzed. According to the postoperative results of caesarean section, MRI T2WI images were used to extract the imaging features of the layers with and without placenta accretion. The data were divided into a training set (n=91) and a validation set (n=39) by stratified sampling in a ratio of 7∶3. Five machine learning methods were adopted: logistic regression (LR), support vector machine (SVM), random forest (RF), decision tree (DT) and K nearest neighbor (KNN) for modeling and classification diagnosis. The hyperparameters of the machine learning model were determined by five-fold cross-validation. Receiver operating characteristic (ROC) curve was adopted to evaluate the prediction efficiency of the model, calculated the area under the curve (AUC), accuracy, sensitivity and specificity, and verified each model in the verification set. In addition, in addition to comparing the diagnostic effectiveness of different machine learning models with that of imaging diagnostic doctors, calibration curves were used to analyze the model efficacy and decision curve analysis (DCA) was used to evaluate clinical practicability. ResultsPlacenta accreta was confirmed in 56 patients after cesarean section, and in 74 patients without placenta accreta. Based on 1688 omics features included in the image preprocessing, 9 image omics features were selected for the construction of the model after least absolute shrinkage and selection operator (LASSO), Selectbest and REF processing. Five kinds of classifier models in the validation set (LR AUC=0.96, SVM AUC=0.97, RF AUC=0.99, DT AUC=0.87, KNN AUC=0.96) had higher diagnostic efficacy for placenta acta than that of imaging doctors (AUC=0.86). Calibration curves show that the calibration degree of RF models is best in the verification set. When the threshold value of validation set DCA is 0.0-0.6, the clinical net benefit of RF, SVM, KNN and LR models is greater than that of DT models. ConclusionsThe machine learning model based on MRI T2WI image features can accurately distinguish the presence or absence of placenta accretion, and its diagnostic efficacy is obviously better than that of physicians^, visual analysis. In addition, compared with the five models, the RF machine learning model has better performance in the diagnosis of placenta accreta. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.016 ObjectivesTo explore the value of combination of N4 bias field correction, histogram-matching (HM) normalization and ComBat harmonization to reduce the "scanner effect" of radiomics features from brain MRI. Materials and MethodsThree-dimensional T1 weighted image (3D-T1WI) and diffusion tensor imaging (DTI) of the brain was performed in 23 healthy volunteers with three MRI scanners (Philip 1.5 T, Philip 3.0 T, GE 3.0 T). Computational Anatomy toolbox (Cat 12) and FMRIB^,s software library (FSL) were used for preprocessing. Then, N4 bias field correction and HM normalization were performed on the preprocessed T1WI and DTI. Finally, LIFEx software was used to extract radiomics features of gray and white matter and then Combat harmonization were carried out. The Shapiro-Wilk test was used to exam the normality and the analysis of variance (ANOVA) and Tukey honestly significant difference (Tukey-HSD) test were used to compare the radiomics features of the three scanners, and the Bartlett spherical test was used to estimate whether the variance was uniform. The differences between scanners in the number of radiomics features and numerical statistical distribution in each processing were qualitatively and quantitatively evaluated. ResultsA total of 10 males and 13 females were enrolled. There was no significant difference in age (t=1.090, P=0.316), education years (t=-0.638, P=0.574) and CES-D score (t=-0.670, P=0.510) between the males and females (P＞0.05). In the original images acquired by the three MRI scanners, the distribution range and peak value of the intensity histogram were not aligned. When N4 bias field correction was performed using 5-level (50 iterations) full mask, the intensity variation coefficient of brain tissue among the three scanners was the lowest. N4 correction sharpened the intensity peak, HM normalized and aligned each intensity peak, and Combat harmonization further aligned the image intensity distribution range and peak of the three MRI. The process (N4 bias field correction, HM normalization and ComBat harmonization) had the same influence on T1WI and DTI sequences. Through the combination of N4 correction, HM normalization and Combat harmonization, the percentage of radiomics features with differences between scanners was reduced from 88.6% (70/79) before bias field correction to 3.8% (3/79) after ComBat harmonization. At the same time, the percentage of radiomics features with differences between VOI of gray and white matter increased from 43.0% (34/79) before bias field correction to 84.8% (67/79) after ComBat harmonization. ConclusionsThe combination of N4 bias field correction, HM normalization and ComBat harmonization can effectively eliminate the "scanner effect" of the brain structural MRI and thereby help to incorporate multi-center MRI data across scanners. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.017 ObjectiveTo optimize frequency offset of T1WI spectral pre-saturation with inversion recovery (SPIR) fat suppression sequence and improve the image quality of T1WI SPIR fat suppression sequence for thoracic spine. Materials and MethodsRetrospective subjective and objective image quality evaluation for 9 T1WI SPIR fat suppression sequences with different frequency offsets (40, 60, 80, 100, 120, 127, 140, 160, 180 Hz) were performed in 36 children with thoracic spine MRI. Two senior radiologists used a 4-point scale to score subjective image quality of the 9 T1WI SPIR sequences, including the homogeneity of vertebral and back fat suppression, and overall image quality. The signal to noise ratio (SNR) of the vertebra, cerebrospinal fluid, intervertebral disc, and spinal cord, and the contrast to noise ratio (CNR) among the them were calculated. Kruskal-Wallis non-parametric test was used to evaluate subjective image quality scores, and one-way analysis of variance was used to evaluate SNR and CNR, LSD used for pairwise comparison among groups. Results(1) There were significant differences in the homogeneity of the vertebral fat suppression (F=168.49, P＜0.001), homogeneity of the back fat suppression (F=96.10, P＜0.001) and overall image quality (F=27.11, P＜0.001) among 9 T1WI SPIR sequences with different frequency offsets. The 4 sequences, with the frequency offsets of 40, 60, 80 and 100 Hz, showed best homogeneity of the vertebral, and there were no significant differences. The 5 sequences, with frequency offsets of the 100, 120, 127, 140 and 160 Hz, showed best homogeneity of the back fat suppression, and there were no significant differences. The 2 sequences, with the frequency offsets of 80 and 100 Hz, showed best overall image quality, and there were no significant differences. (2) There were significant differences in the SNR of vertebra (F=2.83, P＜0.05) and CNR between vetebra and cerebrospinal fluid (F=2.67, P＜0.05) among 6 T1WI SPIR sequences with different frequency offsets. The 3 sequences, with the frequency offset of 40, 60, 80 Hz, had no significant differences in the SNR of vertebra and CNR between vetebra and cerebrospinal fluid, and T1WI SPIR sequence with frequency offset of 40 Hz had higher the SNR of vertebra and CNR between vertebra and cerebrospinal fluid than that of T1WI SPIR sequences with frequency offsets of 100, 120, 127 Hz. There were no significant difference in the SNR and CNR of the remaining studied tissues for the 6 sequences. ConclusionsWhen the frequency offsets were 80 and 100 Hz, the overall image quality of T1WI SPIR sequence of thoracic spine could be improved. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.020 ^,s disease (PD) is a complex progressive neurodegenerative disease with a high prevalence and low diagnostic accuracy. In recent years, a variety of neuroimaging methods based on texture analysis and radiomics had become the focus of PD research. In this paper, the research status and application prospects of the above fields are reviewed, aiming at providing new ideas for neuroimaging research of PD, and then providing more accurate imaging support for clinical diagnosis and treatment of PD. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.021 ^,s disease (PD) is a prevalent neurodegenerative condition characterized by progressive motor dysfunction that can lead to reduced daily activities, diminished quality of life, and eventual disability. To evaluate the underlying structural changes in PD patients, three dimensions T1-weighted imaging (3D-T1WI) and diffusion-tensor imaging (DTI) in structural MRI are utilized to provide a quantitative analysis of microstructural changes in gray matter and white matter, respectively. In recent years, research utilizing 3D-T1WI has implicated tremor in the tremor-dominant (TD) PD subtype, particularly in the hippocampus, parafascicular and ventral intermediate nucleus of the thalamus, pallidumand and cerebellum. DTI has revealed white matter changes in TD subtype patients mainly affecting the inferior fronto-occipital fasciculi, inferior longitudinal fasciculi, corticospinal tract and left dentatorubrothalamic tract, which closely related to the point of origin and treatment of tremor symptoms. This paper presents a review of the imaging principles and analysis methods utilized in the study of TD subtypes with 3D-T1WI and DTI, offering a foundation and guidance for future research. ]]> Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.023 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.024 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.025 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.026 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.027 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.028 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.029 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.030 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.031 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.032 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.033 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.034 Sun,20 Aug 2023 00:00:00 GMT http://med-sci.cn/cgzcx/en/en_articlexml.asp?doi=10.12015/issn.1674-8034.2023.08.035 Sun,20 Aug 2023 00:00:00 GMT