Xuzhu Chen

Background: Intracranial solitary fibrous tumors (ISFTs) are frequently misdiagnosed as meningiomas on MRI. Existing MRI signs have been evaluated individually, without a standardized and externally validated diagnostic framework. Purpose: To develop and validate an MRI‐based diagnostic score for intracranial solitary fibrous tumors (ISFT‐DS) for the preoperative identification of ISFTs. Study Type: Retrospective multicenter diagnostic accuracy study. Population: Five hundred with ISFTs (mean age, 46.4 ± 11.7 years; 269 men), 250 with low‐grade meningiomas (mean age, 57.0 ± 11.1 years; 51 men), and 250 with high‐grade meningiomas (mean age, 49.6 ± 14.3 years; 108 men). The development and external validation cohorts included 840 (84%) and 160 (16%) patients, respectively. Fieldstrength/Sequence: Spin‐echo T1‐weighted imaging (T1WI), fast or turbo spin‐echo T2‐weighted imaging, and contrast‐enhanced spin‐echo T1‐weighted imaging. Assessment: Three potential MRI features—T1WI hyperintensity, honeycomb‐like cystic change, and trans‐compartmental growth pattern—were assessed by three neuroradiologists and combined into the 0–3‐point ISFT‐DS. Diagnostic performance and the effects of standardized training in six readers were evaluated. Statistical Tests: Sensitivity, specificity, accuracy, kappa statistics, receiver operating characteristic analysis, DeLong tests, χ ² and Fisher exact tests, analysis of variance, McNemar test, Kruskal–Wallis tests. p < 0.05 indicated statistical significance. Results: At an ISFT‐DS cutoff of ≥ 1, sensitivity/specificity were 81.4%/83.3% in the development cohort and 76.3%/86.3% in the external validation cohort, respectively. Specificity reached 98.3%/98.8% at ≥ 2 and 100%/100% at 3. Interreader agreement was good to excellent for the three MRI features and the ISFT‐DS (κ = 0.80–0.87). Standardized reader training on the ISFT‐DS increased the overall AUC from 0.67 to 0.80 (ΔAUC = 0.13), with larger gains for Readers 1 and 2 (ΔAUC = 0.21 and 0.15, respectively). Data Conclusion: The ISFT‐DS provides a reproducible and teachable MRI‐based scoring system for standardized preoperative identification of ISFTs. Evidence Level: 3. Technical Efficacy: Stage 2.

Purpose The blood-supplying state of skull base chordomas has been scarcely investigated. We aimed to elucidate the role and superiority of CT perfusion (CTP) in predicting the tumor blood-supplying state to guide surgical strategy. Methods Between 2023 and 2024, 79 consecutive patients with skull base chordoma treated at Beijing Tiantan Hospital were retrospectively identified, and their clinical, imaging, and pathological data were analyzed. We further attempted to establish a grading system for the tumor blood-supplying state. Differences among blood-supplying state groups were assessed, and associations between imaging parameters and the blood-supplying state were examined. Results CTP can distinguish different blood-supplying states of skull base chordoma (rCBF: P = 0.006; rCBV: P < 0.001). CTP (rCBF: r = 0.361, P = 0.001; rCBV: r = 0.425, P < 0.001) has an advantage over MRI (T2WI: r=-0.353, P = 0.001; contrast-enhanced T1WI: r = 0.383, P < 0.001; nADC: r = 0.084, P = 0.462) in reflecting the skull base chordoma blood-supplying state. Between the two CTP parameters, rCBV demonstrated higher diagnostic value than rCBF. Although both could distinguish rich from poor blood-supply (rCBF: 1.90 vs.0.98, r = 0.359, P < 0.01, 95%CI[0.198, 1.253]; rCBV: 3.09 vs.1.55, r = 0.420, P < 0.001, 95%CI[0.728, 2.335]), only rCBV was sensitive enough to differentiate the moderate from the poor blood-supply group (1.55 vs.2.30, r = 0.277, P < 0.05, 95%CI[0.122, 1.263]). Conclusion CTP can be used to judge the blood-supplying states of skull base chordomas, which helps plan the surgical strategy.

Manual processing of spinal CTA images for detecting dural arteriovenous fistulas (SDAVF) is laborious and operator-dependent. We developed an automated AI system (SDAVFdoc) that integrates a 3D convolutional neural network with anatomical prior knowledge to both identify SDAVF and localize the fistula site in a multicenter study of 718 patients. The system sequentially segments spinal structures, localizes the fistula region using anatomical priors, and finally classifies SDAVF likelihood via DenseNet within the foramina. The draining vein cluster segmentation model, using a threshold of 42.5, achieved high accuracy in distinguishing SDAVF cases, with F1-scores ranging from 0.932 to 0.960 across test sets. The DenseNet-based fistula detection model showed a high AUC of 0.928-0.954 across test sets. Compared to technicians, the system reduced processing time from 40.75 ± 11.57 min to 1.05 ± 0.29 min (P < 0.001) and clicks from 761.80 ± 202.05 to 9.68 ± 2.12 (P < 0.001), greatly streamlining clinical workflows. This AI-driven approach enables fast, accurate screening and localization of SDAVF.

Background: This retrospective study compared two types of gliomas and two subtypes of multiple gliomas.Methods: The clinical manifestations, magnetic resonance imaging (MRI) findings, pathological characteristics, and clinical outcomes of 188 patients with unifocal and 94 patients with multiple gliomas (59 with multifocal and 35 with multicentric gliomas) were analyzed.Results: Compared with patients with unifocal glioma, those with multiple gliomas were older (P=0.001) and more likely to be male (χ² = 4.857, P=0.028). Patients with multiple gliomas had smaller extent of surgical resection (χ² = 161.016, P2 = 43.733, P2 = 51.758, P2 = 9.688, P=0.008), intense enhancement (χ² = 24.547, PP=0.001), a lower ratio of isocitrate dehydrogenase (IDH) mutation (χ² = 51.770, P2 = 8.637, P=0.003). Tumor location and IDH status were identified as independent risk factors for multiple gliomas (PP=0.003, respectively). Deep tumor location was found to be the only factor related to unfavorable overall survival (OS) in multiple gliomas. Patients with multifocal gliomas were more likely to be male than patients with multicentric gliomas (χ² = 6.521, P=0.011). The locations of multifocal and multicentric gliomas were significantly different (P=0.048). WHO grade was identified as an independent prognostic factor (P=0.034) in patients with multicentric gliomas but not in those with multifocal gliomas.Conclusions: The demographic characteristics, extent of resection, radiological features, pathological features and prognostic factors differ between patients with multiple gliomas and those with unifocal gliomas. The clinical and radiological features differ between patients with different subtypes of multiple gliomas. Multiple gliomas located only in superficial regions are more likely to be multicentric gliomas and the prognosis is solely related to the WHO grades, providing valuable guidance for clinical treatment.

The study aimed to predict expression of pituitary transcription factor 1 (PIT1) in pituitary adenomas using habitat, intra-tumoral and peri-tumoral radiomics models. A total of 129 patients with pituitary adenoma (training set, n = 103; test set, n = 26) were retrospectively enrolled. A total of 12, 18, 14, 13, and 14 radiomics features were selected from the ROI_intra, ROI_intra+peri (ROI_intra+2mm, ROI_intra+4mm, ROI_intra+6mm), and ROI_habitat, respectively. Then, three machine learning algorithms were employed to develop radiomic models, including logistic regression (LR), support vector machines (SVM), and multilayer perceptron (MLP). The performances of the intra-tumoral, combined intra-tumoral and peri-tumoral, and habitat models were evaluated. The peritumoral region (ROI_2mm, ROI_4mm, ROI_6mm) of the combined model with the highest performance was individually selected for further peritumoral analysis. Moreover, a deep learning radiomics nomogram (DLRN) was constructed incorporating clinical characteristics and the peri-tumoral and habitat models for individual prediction. The combined model_intra+2mm based on ROI_intra+2mm achieved a better performance (AUC, 0.800) than that of the intra-tumoral model alone (AUC, 0.731). And the habitat model showed a higher performance (AUC, 0.806) than that of the intra-tumoral model. In addition, the performance of the peri-tumoral model based on ROI_2mm was 0.694 in the testing set. Furthermore, the DLRN achieved the highest performance of 0.900 in the test set. The DLRN showed the best performance for PIT1 expression in pituitary adenomas, followed by the habitat, combined model_intra+2mm, intra-tumoral model, and peri-tumoral model based on ROI_2mm, respectively. These different models are helpful for the model choice in clinical work.

Objective: To develop a nomogram based on tumor and peritumoral edema (PE) radiomics features extracted from preoperative multiparameter MRI for predicting brain invasion (BI) in atypical meningioma (AM). Methods: In this retrospective study, according to the 2021 WHO classification criteria, a total of 469 patients with pathologically confirmed AM from three medical centres were enrolled and divided into training (n = 273), internal validation (n = 117) and external validation (n = 79) cohorts. BI was diagnosed based on the histopathological examination. Preoperative contrast-enhanced T1-weighted MR images (T1C) and T2-weighted MR images (T2) for extracting meningioma features and T2-fluid attenuated inversion recovery (FLAIR) sequences for extracting meningioma and PE features were obtained. The multiple logistic regression was applied to develop separate multiparameter radiomics models for comparison. A nomogram was developed by combining radiomics features and clinical risk factors, and the clinical usefulness of the nomogram was verified using decision curve analysis. Results: Among the clinical factors, PE volume and PE/tumor volume ratio are the risk of BI in AM. The combined nomogram based on multiparameter MRI radiomics features of meningioma and PE and clinical indicators achieved the best performance in predicting BI in AM, with area under the curve values of 0.862 (95% CI, 0.819–0.905) in the training cohort, 0.834 (95% CI, 0.780–0.908) in the internal validation cohort and 0.867 (95% CI, 0.785–0.950) in the external validation cohort, respectively. Conclusions: The nomogram based on tumor and PE radiomics features extracted from preoperative multiparameter MRI and clinical factors can predict the risk of BI in patients with AM.

Objectives To preoperatively predict the high expression of Ki67 and positive pituitary transcription factor 1 (PIT-1) simultaneously in pituitary adenoma (PA) using three different radiomics models. Methods A total of 247 patients with PA (training set: n = 198; test set: n = 49) were included in this retrospective study. The imaging features were extracted from preoperative contrast-enhanced T1WI (T1CE), T1-weighted imaging (T1WI), and T2-weighted imaging (T2WI). Feature selection was performed using Spearman’s rank correlation coefficient and least absolute shrinkage and selection operator (LASSO). The classic machine learning (CML), deep learning (DL), and deep learning radiomics (DLR) models were constructed using logistic regression (LR), support vector machine (SVM), and multi-layer perceptron (MLP) algorithms. The area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, specificity, accuracy, negative predictive value (NPV) and positive predictive value (PPV) were calculated for the training and test sets. In addition, combined with clinical characteristics, the best CML and the best DL models (SVM classifier), the DL radiomics nomogram (DLRN) was constructed to aid clinical decision-making. Results Seven CML features, 96 DL features, and 107 DLR features were selected to construct CML, DL and DLR models. Compared to CML and DL model, the DLR model had the best performance. The AUC, sensitivity, specificity, accuracy, NPV and PPV were 0.827, 0.792, 0.800, 0.796, 0.800 and 0.792 in the test set, respectively. Conclusions Compared with CML and DL models, the DLR model shows the best performance in predicting the Ki67 and PIT-1 expression in PAs simultaneously.

To analyse the imaging findings of papillary glioneuronal tumors (PGNTs), in order to improve the accuracy of preoperative diagnosis of this tumor. The clinical and imaging manifestations of 36 cases of PGNT confirmed by pathology were analyzed retrospectively. A total of 17 males and 19 females, averaging 22.47 (± 11.23) years. Initial symptoms included epilepsy in ten, headache in seven, and others in 19 cases. 97.2% (35/36) of the lesions were located in the supratentorial area, and 80.5% (29/36) in the intraventricular or deep white matter adjacent to the lateral ventricles. Twenty-four of the lesions (66.7%) were mixed cystic and solid, four (11.1%) were cystic with mural nodules, four (11.1%) were cystic, and four (11.1%) were solid. Four cases of PGNT of cystic imaging showed a “T2-FLAIR mismatch” sign. 69.4% (25/36) had septations. Nine lesions (25%) were accompanied by edema, and 9 (25%) of the mixed cystic and solid lesions were accompanied by hemorrhage. Among the 18 patients who underwent computed tomography (CT) or susceptibility-weighted imaging (SWI), nine had lesions with calcification. PGNTs mostly manifest as cystic mass with mural nodules or mixed cystic and solid mass in the white matter around the supratentorial ventricle, and the cystic part of the lesion is mostly accompanied by septations. Pure cystic lesions may exhibit the sign of “T2-FLAIR mismatch”. PGNT is rarely accompanied by edema but sometimes by calcification and hemorrhage. Patients often present with seizures, headaches, and mass effect symptoms.