OBJECTIVE: To investigate the liver T1rho values for detecting fibrosis, and the potential impact of fatty liver on T1rho measurements. MATERIALS AND METHODS: This study included 18 healthy subjects, 18 patients with fatty liver, and 18 patients with liver fibrosis, who underwent T1rho MRI and mDIXON collections. Liver T1rho, proton density fat fraction (PDFF) and T2* values were measured and compared among the three groups. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the T1rho values for detecting liver fibrosis. Liver T1rho values were correlated with PDFF, T2* values and clinical data. RESULTS: Liver T1rho and PDFF values were significantly different (p < 0.001), whereas the T2* (p = 0.766) values were similar, among the three groups. Mean liver T1rho values in the fibrotic group (52.6 ± 6.8 ms) were significantly higher than those of healthy subjects (44.9 ± 2.8 ms, p < 0.001) and fatty liver group (45.0 ± 3.5 ms, p < 0.001). Mean liver T1rho values were similar between healthy subjects and fatty liver group (p = 0.999). PDFF values in the fatty liver group (16.07 ± 10.59%) were significantly higher than those of healthy subjects (1.43 ± 1.36%, p < 0.001) and fibrosis group (1.07 ± 1.06%, p < 0.001). PDFF values were similar in healthy subjects and fibrosis group (p = 0.984). Mean T1rho values performed well to detect fibrosis at a threshold of 49.5 ms (area under the ROC curve, 0.855), had a moderate correlation with liver stiffness (r = 0.671, p = 0.012), and no correlation with PDFF, T2* values, subject age, or body mass index (p > 0.05). CONCLUSION: T1rho MRI is useful for noninvasive detection of liver fibrosis, and may not be affected with the presence of fatty liver.
Body Mass Index ; Fatty Liver* ; Fibrosis ; Healthy Volunteers ; Humans ; Liver Cirrhosis* ; Liver* ; Magnetic Resonance Imaging ; Prospective Studies* ; Protons ; ROC Curve

Objective:To identify the potential intracranial inflammation in neuromyelitis optica spectrum disorders(NMOSD) patients without supratentorial MRI lesions using quantitative susceptibility mapping (QSM).Methods:Seventy NMOSD patients and 35 age- and gender-matched healthy controls (NC) underwent QSM, 3D-T 1, diffusion MRI from Beijing Tiantan Hospital during June 2019 to June 2021. Susceptibility was compared among NMOSD patients with acute attack (ANMOSD), NMOSD patients in chronic phase (CNMOSD) and NC. The correlation between susceptibility in several brain regions and the cerebrospinal fluid levels of inflammatory makers were analyzed. Results:NMOSD patients showed different susceptibility in several brain regions including bilateral hippocampus, precuneus, right cuneus, putamen, superior parietal and inferior temporal ( P<0.001) and the posr-hoc showed it is higher than normal. Compared to CNMOSD patients, the ANMOSD patients showed increased susceptibility in the cuneus (0.009 ± 0.004 vs. 0.005 ± 0.004, P<0.05). There was significant positive correlations between susceptibility and CSF levels of sTREM2 which reflect the active of microglial cells ( r = 0.494, P<0.05). Conclusions:Despite the absence of supratentorial lesions on MRI, increased susceptibility suggests underlying inflammation in the cerebral cortex in both patients with ANMOSD and CNMOSD, and some of them are obviously related to inflammatory markers in CSF. QSM sequence can be used to explore the potential inflammation in NMOSD patients without obvious supratentorial lesions.

Objective:To investigate the efficacy of a machine learning model based on radiomics of brain lesions on T 2WI in differentiating multiple sclerosis (MS) from neuromyelitis optica spectrum disorders (NMOSD). Methods:Totally 223 MS and NMOSD patients who were treated from January 2009 to September 2018 in Beijing Tiantan Hospital Affiliated to Capital Medical University, Donghu Branch of the First Affiliated Hospital of Nanchang University, Tianjin Medical University General Hospital, and Xuanwu Hospital of Capital Medical University were analyzed retrospectively, and according to the proportion of 7∶3, 223 patients were completely randomly divided into training set (156 cases) and test set (67 cases). A total of 74 patients with MS and NMOSD who were treated in Huashan Hospital Affiliated to Fudan University and China-Japan Friendship Hospital of Jilin University from January 2009 to September 2018 and in Xianghu Branch of the First Affiliated Hospital of Nanchang University from March 2020 to September 2021 were collected as an independent external validation set. All patients underwent brain cross-sectional MR T 2WI, radiomics features were extracted from T 2WI, and features were selected by max-relevance and min-redundancy and least absolute shrinkage and selection operator algorithms. Then various machine learning classifier models (logistic regression, decision tree, AdaBoost, random forest or support vector machine) were constructed to differentiate MS from NMOSD. The area under curve (AUC) of receiver operating characteristics was used to evaluate the performance of each classifier model in the training set, test set and external validation set. Results:Based on multi-center T 2WI, a total of 11 radiomics features related to the discrimination between MS and NMOSD were extracted and classifier models were constructed. Among them, the random forest model had the best efficiency in distinguishing MS from NMOSD, and its AUC values for distinguishing MS from NMOSD in the training set, test set and external validation set were 1.000, 0.944 and 0.902, with specificity of 100%, 76.9% and 86.0%, and sensitivity of 100%, 92.1% and 79.7%, respectively. Conclusion:The random forest model based on the radiomic features of T 2WI of brain lesions can effectively distinguish MS from NMOSD.