CT-based radiomics research for discriminating the risk stratification of pheochromocytoma using different machine learning models: a multi-center study.

OBJECTIVES: The purpose of this study was to explore and verify the value of various machine learning models in preoperative risk stratification of pheochromocytoma. METHODS: A total of 155 patients diagnosed with pheochromocytoma through surgical pathology were included in this research (training cohort: n = 105; test cohort: n = 50); the risk stratification scoring system classified a PASS score of < 4 as low risk and a PASS score of ≥ 4 as high risk. From CT images captured during the non-enhanced, arterial, and portal venous phase, radiomic features were extracted. After reducing dimensions and selecting features, Logistic Regression (LR), Extra Trees, and K-Nearest Neighbor (KNN) were utilized to construct the radiomics models. By adopting ROC curve analysis, the optimal radiomics model was selected. Univariate and multivariate logistic regression analyses of clinical radiological features were used to determine the variables and establish a clinical model. The integration of radiomics and clinical features resulted in the creation of a combined model. ROC curve analysis was used to evaluate the performance of the model, while decision curve analysis (DCA) was employed to assess its clinical value. RESULTS: 3591 radiomics features were extracted from the region of interest in unenhanced and dual-phase (arterial and portal venous phase) CT images. 13 radiomics features were deemed to be valuable. The LR model demonstrated the highest prediction efficiency and robustness among the tested radiomics models, with an AUC of 0.877 in the training cohort and 0.857 in the test cohort. Ultimately, the composite of clinical features was utilized to formulate the clinical model. The combined model demonstrated the best discriminative ability (AUC, training cohort: 0.887; test cohort: 0.874). The DCA of the combined model showed the best clinical efficacy. CONCLUSION: The combined model integrating radiomics and clinical features had an outstanding performance in differentiating the risk of pheochromocytoma and could offer a non-intrusive and effective approach for making clinical decisions.

Development trends and knowledge framework in the application of magnetic resonance imaging in prostate cancer: a bibliometric analysis from 1984 to 2022.

Background: Prostate cancer (PCa) is the most common tumor of the male genitourinary system. With the development of imaging technology, the role of magnetic resonance imaging (MRI) in the management of PCa is increasing. The present study summarizes research on the application of MRI in the field of PCa using bibliometric analysis and predicts future research hotspots. Methods: Articles regarding the application of MRI in PCa between January 1, 1984 and June 30, 2022 were selected from the Web of Science Core Collection (WoSCC) on November 6, 2022. Microsoft Excel 2016 and the Bibliometrix Biblioshiny R-package software were used for data analysis and bibliometric indicator extraction. CiteSpace (version 6.1.R3) was used to visualize literature feature clustering, including co-occurrence analysis of countries, institutions, authors, references, and burst keywords analysis. Results: A total of 10,230 articles were included in the study. Turkbey was the most prolific author. The USA was the most productive country and had strong partnerships with other countries. The most productive institution was Memorial Sloan Kettering Cancer Center. Journal of Magnetic Resonance Imaging and Radiology were the most productive and highest impact factor (IF) journals in the field, respectively. Timeline views showed that "#1 multiparametric magnetic resonance imaging", "#4 pi-rads", and "#8 psma" were currently the latest research hotspots. Keywords burst analysis showed that "machine learning", "psa density", "multi parametric mri", "deep learning", and "artificial intelligence" were the most frequently used keywords in the past 3 years. Conclusions: MRI has a wide range of applications in PCa. The USA is the leading country in this field, with a concentration of highly productive and high-level institutions. Meanwhile, it can be projected that "deep learning", "radiomics", and "artificial intelligence" will be research hotspots in the future.

Efficacy and confounding factors of CT attenuation value differences in distinguishing acute and old vertebral compression fractures: a retrospective study.

PURPOSE: To evaluate the influence of various factors on CT attenuation values (HUs) of acute and old fracture vertebra, and to determine the efficacy of HU differences (△HUs) in the differentiation of the two type of fractures. MATERIALS AND METHODS: A total of 113 acute and 71 old fracture vertebrae confirmed by MRI were included. Four HUs measured at the mid-sagittal, upper 1/3 axial, mid-axial, and lower 1/3 axial planes of each vertebra were obtained. The △HUs between fracture vertebra and its control counterpart was calculated. Receiver operating characteristic (ROC) curve analysis was used and the areas under the ROC curve (AUC) were calculated to evaluate the efficacy of HUs and △HUs. To evaluate the effect of height reduction, region, age and gender on HUs and △HUs, one-way analysis of variance, Pearson correlation analysis and t-test were used. RESULTS: The HUs and △HUs at the upper 1/3 axial plane achieved the highest AUCs of 0.801 and 0.839, respectively. The HUs decreased gradually from Thoracic to Lumbar in control group of acute fracture. While no significant differences were found in the HUs among the 3 localizations in both fracture groups (all P > 0.05). The HUs were negatively correlated with age in all groups. The HUs of male were significantly higher than female patients in all groups (all P < 0.05). While △HU was not significantly different between males and females (all P > 0.05). CONCLUSION: The vertebral HUs at the upper 1/3 axial plane are more likely to identify acute fractures. △HUs were beneficial in eliminating interfering factors.

Clinical study on the use of advanced magnetic resonance imaging in lupus nephritis.

OBJECTIVES: To investigate the correlation between the histopathology of the kidney and clinical indicators in patients with lupus nephritis (LN) using magnetic resonance imaging (MRI). METHODS: A total 50 female participants were enrolled in the study. Thirty patients with LN were divided into types 2, 3, 4, and 5, according to their pathological features. The control group consisted of 20 healthy female volunteers. Serum creatinine, C3, C1q, and anti-ds-DNA were measured. Conventional MRI, DTI, DWI, and BOLD scanning was performed to obtain the FA, ADC, and R2* values for the kidney. RESULTS: Compared with the control group, FA and the ADC were decreased in patients with LN, while the R2* value was increased (P < 0.05). The overall comparison of the SLEDAI (Activity index of systemic lupus erythematosus) score, total pathological score, AI, and serum creatinine C3 showed that these were significantly different between the two groups (P < 0.05). FA and the ADC were negatively correlated with urinary, blood ds-DNA, and serum creatinine and positively correlated with C1q (P < 0.05). The R2* value was positively correlated with urinary NGAL, blood ds-DNA, and serum creatinine (P < 0.05). FA and the ADC were negatively correlated with the SLEDAI score, total pathological score, AI, CI, nephridial tissue C3, and C1q. The R2* value was positively correlated with the SLEDAI score, total pathological score, AI, CI, nephridial tissue C3, and C1q (P < 0.05). CONCLUSIONS: MRI examination in female patients with LN was correlated with pathologic test results, which may have clinical significance in determining the disease's severity, treatment, and outcome.

Multiparametric MRI Radiomics for the Early Prediction of Response to Chemoradiotherapy in Patients With Postoperative Residual Gliomas: An Initial Study.

PURPOSE: To evaluate whether multiparametric magnetic resonance imaging (MRI)-based logistic regression models can facilitate the early prediction of chemoradiotherapy response in patients with residual brain gliomas after surgery. PATIENTS AND METHODS: A total of 84 patients with residual gliomas after surgery from January 2015 to September 2020 who were treated with chemoradiotherapy were retrospectively enrolled and classified as treatment-sensitive or treatment-insensitive. These patients were divided into a training group (from institution 1, 57 patients) and a validation group (from institutions 2 and 3, 27 patients). All preoperative and postoperative MR images were obtained, including T1-weighted (T1-w), T2-weighted (T2-w), and contrast-enhanced T1-weighted (CET1-w) images. A total of 851 radiomics features were extracted from every imaging series. Feature selection was performed with univariate analysis or in combination with multivariate analysis. Then, four multivariable logistic regression models derived from T1-w, T2-w, CET1-w and Joint series (T1+T2+CET1-w) were constructed to predict the response of postoperative residual gliomas to chemoradiotherapy (sensitive or insensitive). These models were validated in the validation group. Calibration curves, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA) were applied to compare the predictive performances of these models. RESULTS: Four models were created and showed the following areas under the ROC curves (AUCs) in the training and validation groups: Model-Joint series (AUC, 0.923 and 0.852), Model-T1 (AUC, 0.835 and 0.809), Model-T2 (AUC, 0.784 and 0.605), and Model-CET1 (AUC, 0.805 and 0.537). These results indicated that the Model-Joint series had the best performance in the validation group, followed by Model-T1, Model-T2 and finally Model-CET1. The calibration curves indicated good agreement between the Model-Joint series predictions and actual probabilities. Additionally, the DCA curves demonstrated that the Model-Joint series was clinically useful. CONCLUSION: Multiparametric MRI-based radiomics models can potentially predict tumor response after chemoradiotherapy in patients with postoperative residual gliomas, which may aid clinical decision making, especially to help patients initially predicted to be treatment-insensitive avoid the toxicity of chemoradiotherapy.

Interrupted or Uninterrupted Oral Anticoagulants in Patients Undergoing Atrial Fibrillation Ablation.

BACKGROUND AND PURPOSE: The safety and efficacy of uninterrupted, minimally interrupted (one dose skipped) or completely interrupted (24 h skipped) oral anticoagulant therapy in patients with atrial fibrillation (AF) ablation are poorly defined. We conducted a network meta-analysis to explore the effect of interrupted or uninterrupted oral anticoagulants in patients with AF undergoing ablation. METHODS: The Cochrane Library, PubMed and Embase databases were systematically searched for studies comparing uninterrupted, minimally interrupted or completely interrupted non-vitamin K antagonist oral anticoagulants (NOACs) with continuous or interrupted warfarin in patients undergoing AF ablation. RESULTS: Twelve randomized clinical trials (RCTs) with a total of 5597 patients with AF undergoing catheter ablation were included. For thromboembolism, minimally interrupted NOACs (OR 0.03, 95% CI 0.01-0.35), uninterrupted NOACs (OR 0.04, 95% CI 0.01-0.23) and continuous VKAs (OR 0.05, 95% CI 0.01-0.21) were better than interrupted warfarin. The risk of total bleeding appeared higher in the completely interrupted NOAC group compared with the minimally interrupted NOACs (OR 2.74, 95% CI 1.18-6.37), uninterrupted NOACs (OR 2.15, 95% CI 1.05-4.38) and uninterrupted warfarin (OR 2.04, 95% CI 1.02-4.08). To reduce the risk of total bleeding, minimally interrupted NOACs (OR 0.15, 95% CI 0.08-0.27), uninterrupted NOACs (OR 0.19, 95% CI 0.14-0.42) and uninterrupted warfarin (OR 0.24, 95% CI 0.15-0.39) were better than interrupted warfarin. In the event of major bleeding, there was no significant difference in the interrupted NOAC, uninterrupted NOAC, interrupted VKA and uninterrupted VKA groups. CONCLUSIONS: These three NOAC strategies may have similar safety and efficacy in terms of thromboembolism and major bleeding complications. The total bleeding risk of completely interrupted oral anticoagulants is higher than that of uninterrupted and minimally interrupted NOACs. For thromboembolism, minimally interrupted NOACs, uninterrupted NOACs and continuous VKAs were better than interrupted warfarin.

Altered whole-brain gray matter volume in primary angle closure glaucoma patients: a voxel-based morphometry study.

The study aimed to compare the whole-brain gray matter volume (GMV) and white matter volume (WMV) difference between primary angle closure glaucoma (PACG) patients and health controls (HCs) using a voxel-based morphometry method. A total of 27 patients with PACG (17 males and 10 females) and 27 HCs (17 males and 10 females), closely matched for age and education, were enrolled in the study. All subjects underwent magnetic resonance imaging (MRI) scans. The MRI data were processed using SPM8 software in voxel-based morphometry 8 toolbox. The relationship between the mean GMV values of brain regions and the clinical features including psychological testing and mean retinal nerve fiber layer (RNFL) thickness in PACG groups were analyzed by using Pearson correlation. Compared with HCs, PACG patients showed significantly decreased GMV values in the left cerebellum posterior lobe (CPL), right extra-nuclear, and right superior temporal gyrus. In contrast, PACG patients showed significantly increased GMV values in the left CPL, right CPL, right superior temporal gyrus, right thalamus and right insula (P<0.01). Moreover, in the PACG group, the left mean RNFL showed a positive correlation with the mean GMV values of the left CPL (r=0.719; P<0.001) and the right mean RNFL showed a positive correlation with the mean GMV values of the left CPL (r=0.721; P<0.001). The Hamilton depression score showed a positive correlation with the mean GMV values of right insula (r=0.897; P<0.001). Our results demonstrated that PACG patients showed altered brain structure in various regions related to visuomotor function, thalamocortical pathway, and emotion function, which might provide a useful informations to understanding the anatomy neural mechanisms of deficit in vision loss and depression in PACG.