Search

Background Three-dimensional T1 magnetization prepared rapid acquisition gradient echo (3D-T1-MPRAGE) is preferred in detecting brain metastases (BM) among MRI. We developed an automatic deep learning–based detection and segmentation method for BM (named BMDS net) on 3D-T1-MPRAGE images and evaluated its performance. Methods The BMDS net is a cascaded 3D fully convolution network (FCN) to automatically detect and segment BM. In total, 1652 patients with 3D-T1-MPRAGE images from 3 hospitals (n = 1201, 231, and 220, respectively) were retrospectively included. Manual segmentations were obtained by a neuroradiologist and a radiation oncologist in a consensus reading in 3D-T1-MPRAGE images. Sensitivity, specificity, and dice ratio of the segmentation were evaluated. Specificity and sensitivity measure the fractions of relevant segmented voxels. Dice ratio was used to quantitatively measure the overlap between automatic and manual segmentation results. Paired samples t-tests and analysis of variance were employed for statistical analysis. Results The BMDS net can detect all BM, providing a detection result with an accuracy of 100%. Automatic segmentations correlated strongly with manual segmentations through 4-fold cross-validation of the dataset with 1201 patients: the sensitivity was 0.96 ± 0.03 (range, 0.84–0.99), the specificity was 0.99 ± 0.0002 (range, 0.99–1.00), and the dice ratio was 0.85 ± 0.08 (range, 0.62–0.95) for total tumor volume. Similar performances on the other 2 datasets also demonstrate the robustness of BMDS net in correctly detecting and segmenting BM in various settings. Conclusions The BMDS net yields accurate detection and segmentation of BM automatically and could assist stereotactic radiotherapy management for diagnosis, therapy planning, and follow-up.

BACKGROUND AND PURPOSE: The differential diagnosis of radionecrosis and tumor recurrence in brain metastases is challenging. We investigated the diagnostic efficiency of postcontrast T1 mapping in solving this problem. MATERIALS AND METHODS: Between March 2016 and June 2017, fifty-six patients with brain metastases who underwent contrast-enhanced cerebral T1 mapping were recruited for this prospective study. The findings revealed new enhancement after gamma knife radiosurgery. The subjects were assigned to radionecrosis and recurrence groups based on follow-up (median, 11.5 months) and histopathologic results. T1 values of lesions 5 (T1(5min)) and 60 (T1(60min)) minutes after administration of contrast agent and their difference (T1(differ)) were compared between the 2 groups with the 2-tailed Mann-Whitney U test. Receiver operating characteristic curves were used to determine the optimum cutoff values for differential diagnosis. RESULTS: There were significant differences between the 2 groups in T1(5min), T1(60min), and T1(differ) values (P = .012, P = .004, and P < .001, respectively). Relative to T1(5min) and T1(60min), T1(differ) exhibited greater sensitivity and specificity (P < .001, respectively) in identifying radionecrosis. The optimum T1(differ) value for differential diagnosis was 71.1 ms (area under the curve = 0.97; 95% CI, 0.93–1.00), with sensitivity and specificity of 81.5% and 96.5%, respectively. CONCLUSIONS: Postcontrast T1 mapping is optimal for the differential diagnosis of radionecrosis and tumor recurrence. Among T1 parameters, T1(differ) is the most powerful parameter for differential diagnosis. Advantages in terms of quantitative analysis and high resolution portend the wide use of postcontrast T1 mapping in the future.

Objective To investigate the association between polymorphisms in the gene encoding activin receptorlike kinase 1 (ACVRL1) with brain arteriovenous malformations (BAVMs) using a case-control study in a Chinese Han population, followed by a meta-analysis of the published literature. Methods This study focused on the genotypic analysis of 4 single nucleotide polymorphisms (SNPs; rs2071219, rs706819, rs2293094, and rs11169953) in 50 patients with BAVM and 120 healthy volunteers attending Provincial Hospital in China. A meta-analysis was subsequently conducted involving an extensive literature search for relevant studies. Results Our cohort study showed a significant association between ACVRL1 rs706819 and increased risk for BAVM. Reduced BAVM risk was correlated with the G allele of rs2293094 and the C allele of rs11169953. However, neither the genotype nor allele frequencies of rs2071219 were found to be significantly different between the BAVM and control groups. Meta-analysis further confirmed that no significant evidence of association was found between rs2071219 and BAVM risk. Haplotype analysis of rs706819, rs2293094, and rs11169953 showed that the GGT haplotype could reduce the risk of BAVM, whereas the GAC haplotype may increase the risk of BAVM. Conclusions The present study indicates an association between 3 susceptibility SNPs, rs706819, rs2293094, and rs11169953, in the ACVRL1 gene and BAVM. Follow-up functional studies on the ACVRL1 gene are required to better understand its roles in BAVM development.

Eleated expression of NIMA-related kinase 2 (NEK2) was frequently observed in a variety of malignant cancers, and it appears to be involved in the initiation, maintenance, progression, metastasis of cancer and is positively associated with poor prognosis. We sought to investigate NEK2 expression and its predictive roles in malignant gliomas, and study the correlation of NEK2 protein expression with proliferation, clinical parameters, overall survival and some other parameters. We investigate NEK2 protein expression in 99 samples of malignant gliomas, including 35 WHO grade II, 22 grade III, and 42 grade IV gliomas, by immunohistochemistry and western blot (n = 50). We then made correlative analysis of protein overexpression using the Kaplan-Meier method, Log rank test, and Cox proportional-hazards model analysis. NEK2 protein was overexpressed in malignant gliomas, but not in normal brain tissues. Overexpression of NEK2 correlated with malignancy, proliferation and adverse overall survival in gliomas. Moreover, chemotherapy, resection extent and WHO grade also correlate with overall survival in gliomas. However, within WHO grade II glioma subgroup, NEK2 overexpression showed no impact on overall survival. The present study firstly reveals that NEK2 protein is widely overexpressed in gliomas. NEK2 overexpression correlates significantly with malignancy (WHO grades), proliferation (Ki-67) and prognosis in malignant gliomas. NEK2 is a potential gene therapy target and prognostic indicator.

This study aimed to investigate the feasibility of segmentation-independent volume rendering (SI-VR) in visualising the root entry zone (REZ), and to explore the influence on the management of vascular compression syndromes (VCSs). Two hundred and twenty patients with VCSs were recruited in this prospective study from July 2015 to May 2019. SI-VR was reconstructed based on inverted 3D fast spin echo T2WI. They were assigned to the experimental group and control group randomly. Patients in the experimental group would accept extra evaluation based on SI-VR before microvascular decompression. Image quality and diagnostic accuracy between SI-VR and 3D fast spin echo T2WI in the experimental group were compared by Mann–Whitney U test and chi-square test, separately. Interobserver agreement was performed with intraclass correlation coefficient. Postsurgical outcomes and complications between two groups were compared by chi-square test. SI-VR had a better interobserver agreement (0.82 vs 0.68) and diagnostic accuracy (95.5% vs 83.6%, p = 0.004) than that of 3D fast spin echo T2WI. Especially, significantly improved diagnostic accuracy was reached in detecting the multi-vascular branches compression (100% vs 15.4%, p

Accurate pathological diagnosis of gliomas recurrence is crucial for the optimal management and prognosis prediction. The study here unravels that our newly developed γ-glutamyltranspeptidase (GGT) fluorescence probe (Figure 1A) imaging in twenty recurrent glioma tissues selectively recognizes the most malignant portion from treatment responsive tissues induced by radio/chemo-therapy (Figure 1B). The overexpression of GGT in recurrent gliomas and low level in radiation necrosis were validated by western blot analysis and immunohistochemistry. Furthermore, the ki-67 index evaluation demonstrated the significant increase of malignancy, aided by the GGT-responsive fluorescent probe to screen out the right specimen through fast enhanced imaging of enzyme activity. Importantly, our GGT-targeting probe can be used for accurate determination of pathologic evaluation of tumor malignancy, and eventually for guiding the following management in patients with recurrent gliomas. Open in a separate window FIGURE 1 (A) Schematic illustrations of GGT-catalyzed transformation of CN-BOD-GSH to CN-BOD-N and (B) the cartoon graph showed the process of the visualization of recurrent glioma for precise pathology diagnosis.

Gliomas are the most common primary malignant brain tumors in adults. Accurate grading is crucial as therapeutic strategies are often disparate for different grades and may influence patient prognosis. This study aims to provide an automated glioma grading platform on the basis of machine learning models. In this paper, we investigate contributions of multi-parameters from multimodal data including imaging parameters or features from the Whole Slide images (WSI) and the proliferation marker Ki-67 for automated brain tumor grading. For each WSI, we extract both visual parameters such as morphology parameters and sub-visual parameters including first-order and second-order features. On the basis of machine learning models, our platform classifies gliomas into grades II, III, and IV. Furthermore, we quantitatively interpret and reveal the important parameters contributing to grading with the Local Interpretable Model-Agnostic Explanations (LIME) algorithm. The quantitative analysis and explanation may assist clinicians to better understand the disease and accordingly to choose optimal treatments for improving clinical outcomes. The performance of our grading model was evaluated with cross-validation, which randomly divided the patients into non-overlapping training and testing sets and repeatedly validated the model on the different testing sets. The primary results indicated that this modular platform approach achieved the highest grading accuracy of 0.90 ± 0.04 with support vector machine (SVM) algorithm, with grading accuracies of 0.91 ± 0.08, 0.90 ± 0.08, and 0.90 ± 0.07 for grade II, III, and IV gliomas, respectively.

Background Intracranial primary collision tumors of different histologic types are rare, and their occurrence is still unclear. Case Description We describe a 66-year-old female who presented with headache, nausea, and vomiting. Magnetic resonance imaging scan showed that there were 2 primary intracranial tumors occurring simultaneously at adjacent sites of the right cerebral hemisphere. Tumor pathology showed 2 distinct tumors: meningioma (World Health Organization I) and glioblastoma. This is a rare case in which 2 different intracranial primary tumors occurred at adjacent sites, but the patient had no history of head trauma, neurologic surgery, or radiation therapy. Conclusions According to previous and present reports, the most common type of intracranial primary collision tumor is composed of a benign meningioma and a glioblastoma. During the occurrence of collision tumors, 1 tumor can play a role in the formation and growth of the other.

Treatment of recurrent high-grade gliomas (rHGG) has always been challenging. This study aimed to explore the treatment effect of quantitative dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI)-guided gamma knife radiosurgery (GKRS) on rHGG. Between April 2014 and July 2016, 26 consecutive patients were treated by quantitative DSC-PWI-guided GKRS as salvage treatment for rHGG. The gross tumor volume (GTV) was defined as the high perfusion area on absolute cerebral blood volume maps, with a cutoff value of 22 ml/100 g. The clinical target volume (CTV) encompassed the GTV by 3 mm. Overall survival (OS) and progression-free survival (PFS) were calculated by the Kaplan–Meier method. Prognostic factors were tested by the log-rank (Mantel–Cox) test. With a median follow-up of 32 months, the median PFS after GKRS was 8 months (95% CI [6, 12]); the 1- and 2-year survival rates were 30.8 and 11.5%, respectively. The median OS was 25.5 months (95% CI [18, 40]); the 1- and 2-year survival rates were 96.2 and 57.7%, respectively. Pathology grade and CTV were identified as prognostic factors for PFS. However, none of the parameters tested were independent prognostic factors for OS among these selected patients. No severe radiotoxicity was observed among all patients. Quantitative DSC-PWI-guided GKRS is feasible for the treatment of rHGG and that these outcomes remain to be validated. Despite this, we think that carefully selected patients can benefit from this treatment method.

Background To investigate the asymmetry of quantitative cerebral blood flow (CBF) values in interictal temporal lobe epilepsy (TLE) patients with (TLE-MTS) and without (TLE-no) mesial temporal sclerosis. Methods Twenty-six TLE patients (14 TLE-MTS/12 TLE-no) and 22 controls were studied with a three-dimensional whole-brain pseudocontinuous arterial spin-labeling technique at 3.0 Tesla. Regions of interest were placed at the bilateral hippocampal heads, amygdalas and temporal lobes to get the average regional cerebral blood flow (rCBF) and calculate asymmetry indices (AIs). Statistical parametric mapping detected the whole-brain CBF changes of the two subgroups (right-sided). Results AIs of the hippocampus and amygdala in TLE-MTS and AIs of hippocampus in TLE-no were significantly different versus controls (P

Object The COL1A2 gene at 7q22.1 has been shown to be associated with familial intracranial aneurysms (IAs) in the Japanese population. In the present study, the authors investigated the correlation between the presence of the rs42524 polymorphism in COL1A2 and the occurrence of sporadic IAs in Chinese patients. Methods The polymorphism rs42524 of the COL1A2 gene was identified by polymerase chain reaction–based restriction analysis in genomic DNA from 226 patients with sporadic IAs (mean age 51.49 ± 11.47 years) and 326 control participants (mean age 52.33 ± 10.50 years). Neurological assessments were performed using the Hunt and Hess grading system, and differences in allelic and genotypic frequencies between the patient and control groups were evaluated with the chi-square test. Results There was a significant difference in either the genotype distribution (χ2 = 11.99, p = 0.002) or allelic frequencies (χ2 = 11.96, p = 0.001, odds ratio 2.579, 95% confidence interval 1.486–4.476) between patients with IAs and patients in the control group. Conclusions The rs42524 polymorphism of COL1A2 could be a genetic risk factor for sporadic IAs among individuals of Chinese Han ethnicity. This study is the first to confirm the association between COL1A2 and IAs.

This retrospective study aimed to identify the pattern of visual impairment in pituitary adenoma patients. Patients' eyes were categorized into four groups according to their visual acuity score. Trans-sphenoidal hypophysectomy was performed on all patients. Significant differences between groups were evaluated for visual field defects, visual symptoms duration, degree of suprasellar extension and tumour pathology. In the 201 patients (402 eyes) evaluated, 166 eyes had impaired visual acuity (45 eyes, < 4.0; 43 eyes, 4.0–4.4; and 78 eyes, 4.5–4.8), 236 eyes scored > 4.8 and were considered to be free of this disorder. A linear correlation was found between visual field defects and visual acuity score. Duration of visual symptoms between visual acuity groups was not significant. Suprasellar extension (measured by Hardy grades) was significant between visual acuity groups and may be the main cause of visual acuity loss in pituitary adenoma patients. Non-functional pituitary adenomas tended to affect visual acuity more than adenomas with other pathological diagnoses which are adrenocorticotropic hormone and prolactin. Further investigation is required to clarify visual loss cause in pituitary adenoma patients.

To further confirm at the molecular level that neuronal apoptosis occurs in mesial temporal sclerosis (MTS), the main substrate of mesial temporal lobe epilepsy (MTLE), 24 resected sclerotic hippocampi from 24 patients with drug-resistant MTLE associated with MTS were studied microscopically, electronmicroscopically and immunohistochemically, with detection of expression of apoptosis-associated genes including bcl-2, p53, bax, fas and caspase-3. Early apoptosis changes were found morphologically in hippocampi from three patients with MTLE using transmission electron microscopy. Positive immunostained neurons for bcl-2, p53, fas and caspase-3 were found in the sclerotic hippocampi of 19/24, 14/24, 22/24 and 20/24 patients respectively, which was statistically different from controls. Correlative analysis showed the expression of p53, fas and caspase-3 were positively correlated with seizure frequency. Apoptosis may contribute to MTS, and seizures may induce apoptosis, and thus contribute to neuronal loss in MTS.

Angiocentric glioma is a rare subtype of neuroepithelial tumor that is associated with a history of epilepsy. We report a case of cystoid angiocentric glioma associated with an area of calcification. This 25 year old male patient presented with tonic clonic spasm. He underwent craniotomy with complete resection of the lesion. Pathologic specimen showed monomorphous bipolar cells with angiocentric growth pattern.

The aim of this study was to explore the difference in c-Met expression between primary and recurrent glioblastoma multiforme (GBM), and to determine whether the dysregulation of c-Met expression has a role in the malignant progression of GBM. Paired primary and recurrent GBM specimens from the same patient were evaluated using immunohistochemical analysis. The association between c-Met expression and progression-free survival time (PFST) was analyzed. There was a significant difference in c-Met expression between primary and recurrent tumors (p=0.020), and patients with tumors expressing c-Met at a higher level had a significantly shorter PFST (6.1 months vs. 11.5 months; p=0.026). Our study indicates that recurrent GBM express c-Met at a higher level and that c-Met overexpression is associated with shorter PFST in patients with GBM. These findings suggest that c-Met potentially has an important role in the treatment of GBM.

Glioma, the most common type of primary central nervous system cancers, was progressive with poor survival. MicroRNA, as a novel biomarker, was suspected to be novel biomarkers for glioma diagnosis and prognosis. The study aimed at investigating the diagnostic and predictive value of miR-221/222 family for glioma. In the first phase, we compared plasma miR-221/222 family levels between 50 glioma patients and 51 healthy controls by real-time qRT-PCR amplification. Meanwhile, a meta-analysis based on published studies and presents study was performed to explore the diagnostic performance of miR-221/222 family in human cancers. In the second phase, we correlated the miR-221/222 family expression level with prognosis of glioma using Kaplan-Meier survival curves. The plasma miR-221/222 family levels were found to be significantly upregulated in glioma patients (P = 0.001). The ROC curve analysis yielded an AUC values of 0.84 (95 % confidence interval (CI): 0.74–0.93) for miR-221 and 0.92 (95 % CI 0.87–0.94) for miR-222. In the meta-analysis, the summary receiver operating characteristic (sROC) was plotted with an AUC of 0.82 (95 % CI 0.78–0.85) for miR-221/222 family. It was also demonstrated that high positive plasma miR-221 and miR-222 were both correlated with poor survival rate (miR-221: HR = 2.13; 95 % CI, 1.05–4.31; miR-222: HR = 2.09; 95 % CI, 1.00–4.37). This study demonstrated that the detection of the miRNA-221/222 family should be considered as a new additional tool to better characterize glioma.

Objective: Pre-operative predictive factors of progression-free survival (PFS) and tumor recurrence after initial surgery are important in counseling patients and decision making. Though PFS after initial surgery in patients with low grade astrocytomas has been described, little is described about PFS in patients with different tumor grades. Our objective was to investigate potential predictive factors of PFS, and devise a scale to predict PFS and tumor recurrence after initial surgery in patients with primary and recurrent astrocytomas of low and high tumor grades. Methods: Clinical, radiographic, pathological and treatment data of 62 patients whose initial treatments of primary and recurrent astrocytomas were both surgeries were analyzed, and factors that had significant correlation with PFS was used to devise a scale. Results: Factors significantly related with PFS were: the time from onset of symptoms to clinical and radiological diagnosis of astrocytomas (Spearman correlation coefficient r=0.298, significance level P=0.019) and with the symptoms of seizures (r=0.292, P=0.021). Patients with age between 30 and 40 years had significant longer PFS than the rest age group (P=0.018, oneway ANOVA). A simple scale (from 0 to 3 points) comprised of the three factors distinguished four groups of patients with significant different post-operative PFS (0 point, 8.0 months; 1 point, 13.7 months; 2 points, 18.0 months; 3 points, 34.5 months) (P=0.004, oneway ANOVA). Conclusion: The simple scale we devised comprised of the three pre-operative prognostic factors can significantly distinguish patients with different post-operative survival after initial treatment of astrocytomas with surgery.

To investigate the hemodynamic response of the cerebral bridging veins to increased intracranial pressure (ICP) during normo- and hyperventilation.Flow velocity (FV(m)), diameter (D), and pulsatility index (PI) were measured and calculated in the cerebral bridging veins using Color Doppler Ultrasound in seven pigs, and cerebral blood flow (CBF) was measured using (133)Xe clearance in nine pigs, both during normo- and hyperventilation. ICP was increased stepwise from baseline (about 10 mmHg) to 20 and 30 mmHg by infusion of mock CSF into the cisterna magna.Moderate elevations of ICP caused venous relative stasis as evidenced by a decrease in FV(m) and increase in diameter and PI, but no change of volume flow in the veins. CBF was stable indicating autoregulation at ICP of 20 and 30 mmHg. Parallel observations were made during normo- and hyperventilation, but at two different levels of CBF.The cerebral bridging veins dilation and blood flow velocity decrease indicate the venous relative stasis in response to the elevated ICP. This response is proposed to be caused by an ICP-dependent increase in resistance to the outflow from the cerebral bridging veins.

A 58-year-old female and an 18-year-old male patient had progressive spastic quadriparesis of 10 years and 6 months duration, respectively. Proatlas segmentation malformation (PSM) was confirmed using three-dimensional (3D) reconstructive CT scans and MRI. Surgical procedures in one patient involved anterior decompression via a transoropharyngeal approach, cranial traction, and posterior occipital-cervical fixation and fusion. His postoperative neurological status had improved remarkably, with imaging showing good realignment of the occipito-atlanto-axial complex with comfortable decompression of the cervico-medulla junction and relief of syringomyelia. MRI and 3D-CT scans are the definitive diagnostic tools for PSM, and appropriate aggressive surgical intervention should be undertaken.