Refine
Year of publication
Document Type
- Article (60)
- Conference Proceeding (4)
Language
- English (64)
Has Fulltext
- yes (64)
Is part of the Bibliography
- no (64)
Keywords
- glioblastoma (5)
- Epilepsy (3)
- epilepsy (3)
- glioma (3)
- meningioma (3)
- quantitative MRI (3)
- Fasting (2)
- Glioblastoma (2)
- Glioma (2)
- Glucose (2)
Institute
- Medizin (63)
- MPI für Hirnforschung (2)
- Sportwissenschaften (2)
- Georg-Speyer-Haus (1)
- Geowissenschaften (1)
Amide proton transfer-chemical exchange saturation transfer (APT-CEST) imaging provides important information for the diagnosis and monitoring of tumors. For such analysis, complete coverage of the brain is advantageous, especially when registration is performed with other magnetic resonance (MR) modalities, such as MR spectroscopy (MRS). However, the acquisition of Z-spectra across several slices via multislice imaging may be time-consuming. Therefore, in this paper, we present a new approach for fast multislice imaging, allowing us to acquire 16 slices per frequency offset within 8 s. The proposed fast CEST-EPI sequence employs a presaturation module, which drives the magnetization into the steady-state equilibrium for the first frequency offset. A second module, consisting of a single CEST pulse (for maintaining the steady-state) followed by an EPI acquisition, passes through a loop to acquire multiple slices and adjacent frequency offsets. Thus, the whole Z-spectrum can be recorded much faster than the conventional saturation scheme, which employs a presaturation for each single frequency offset. The validation of the CEST sequence parameters was performed by using the conventional saturation scheme. Subsequently, the proposed and a modified version of the conventional CEST sequence were compared in vitro on a phantom with different T1 times and in vivo on a brain tumor patient. No significant differences between both sequences could be found in vitro. The in vivo data yielded almost identical MTRasym contrasts for the white and gray matter as well as for tumor tissue. Our results show that the proposed fast CEST-EPI sequence allows for rapid data acquisition and provides similar CEST contrasts as the modified conventional scheme while reducing the scanning time by approximately 50%.
Posterior fossa tumor surgery is challenging due to the proximity and exposure of cerebellar structures. A favorable operative approach is unknown. Following lesions to the dentato–rubro–olivary-pathway, a neurodegenerative disease called hypertrophic olivary degeneration (HOD) can occur. This study for the first time demonstrates that paravermal trans-cerebellar approaches are associated with a significantly higher likelihood of HOD on MRI when compared to other approaches. This finding can well be attributed to dentate nucleus (DN) injury. Furthermore, cerebellar mutism syndrome (CMS) was discussed in the literature to be correlated with HOD due to a functional overlap of pathways involved. We found no such correlation in this study, but HOD was shown to be a reliable indicator for surgical disruption of efferent cerebellar pathways involving the DN. Henceforth, neurosurgeons should consider more midline or lateral approaches in posterior fossa surgery to spare the DN whenever feasible, and focus on cerebellar functional anatomy in their preoperative planning.
Simple Summary: Pseudoprogression detection in glioblastoma patients remains a challenging task. Although pseudoprogression has only a moderate prevalence of 10–30% following first-line treatment of glioblastoma patients, it bears critical implications for affected patients. Non-invasive techniques, such as amino acid PET imaging using the tracer O-(2-[18F]-fluoroethyl)-L-tyrosine (FET), expose features that have been shown to provide useful information to distinguish tumor progression from pseudoprogression. The usefulness of FET-PET in IDH-wildtype glioblastoma exclusively, however, has not been investigated so far. Recently, machine learning (ML) algorithms have been shown to offer great potential particularly when multiparametric data is available. In this preliminary study, a Linear Discriminant Analysis-based ML algorithm was deployed in a cohort of newly diagnosed IDH-wildtype glioblastoma patients (n = 44) and demonstrated a significantly better diagnostic performance than conventional ROC analysis. This preliminary study is the first to assess the performance of ML in FET-PET for diagnosing pseudoprogression exclusively in IDH-wildtype glioblastoma and demonstrates its potential.
Abstract: Pseudoprogression (PSP) detection in glioblastoma remains challenging and has important clinical implications. We investigated the potential of machine learning (ML) in improving the performance of PET using O-(2-[18F]-fluoroethyl)-L-tyrosine (FET) for differentiation of tumor progression from PSP in IDH-wildtype glioblastoma. We retrospectively evaluated the PET data of patients with newly diagnosed IDH-wildtype glioblastoma following chemoradiation. Contrast-enhanced MRI suspected PSP/TP and all patients underwent subsequently an additional dynamic FET-PET scan. The modified Response Assessment in Neuro-Oncology (RANO) criteria served to diagnose PSP. We trained a Linear Discriminant Analysis (LDA)-based classifier using FET-PET derived features on a hold-out validation set. The results of the ML model were compared with a conventional FET-PET analysis using the receiver-operating-characteristic (ROC) curve. Of the 44 patients included in this preliminary study, 14 patients were diagnosed with PSP. The mean (TBRmean) and maximum tumor-to-brain ratios (TBRmax) were significantly higher in the TP group as compared to the PSP group (p = 0.014 and p = 0.033, respectively). The area under the ROC curve (AUC) for TBRmax and TBRmean was 0.68 and 0.74, respectively. Using the LDA-based algorithm, the AUC (0.93) was significantly higher than the AUC for TBRmax. This preliminary study shows that in IDH-wildtype glioblastoma, ML-based PSP detection leads to better diagnostic performance.
Purpose: The diagnosis of abusive head trauma (AHT) is complex and neuroimaging plays a crucial role. Our goal was to determine whether non-neuroradiologists with standard neuroradiology knowledge perform as well as neuroradiologists with experience in pediatric neuroimaging in interpreting MRI in cases of presumptive AHT (pAHT).
Methods: Twenty children were retrospectively evaluated. Patients had been diagnosed with pAHT (6 patients), non-abusive head trauma-NAHT (5 patients), metabolic diseases (3 patients), and benign enlargement of the subarachnoid spaces (BESS) (6 patients). The MRI was assessed blindly, i.e., no clinical history was given to the 3 non-neuroradiologists and 3 neuroradiologists from 2 different institutions.
Results: Blindly, neuroradiologists demonstrated higher levels of sensitivity and positive predictive value in the diagnosis of pAHT (89%) than non-neuroradiologists (50%). Neuroradiologists chose correctly pAHT as the most probable diagnosis 16 out of 18 times; in contrast, non-neuroradiologists only chose 9 out of 18 times. In our series, the foremost important misdiagnosis for pAHT was NAHT (neuroradiologists twice and non-neuroradiologists 5 times). Only victims of motor vehicle accidents were blindly misdiagnosed as pAHT. No usual household NAHT was not misdiagnosed as pAHT. Neuroradiologists correctly ruled out pAHT in all cases of metabolic diseases and BESS.
Conclusion: MRI in cases of suspected AHT should be evaluated by neuroradiologists with experience in pediatric neuroimaging. Neuroradiologists looked beyond the subdural hemorrhage (SDH) and were more precise in the assessment of pAHT and its differential diagnosis than non-neuroradiologists were. It seems that non-neuroradiologists mainly assess whether or not a pAHT is present depending on the presence or absence of SDH.
Background and Objectives: Proteins of the coagulation system contribute to autoimmune inflammation in patients with multiple sclerosis (MS). On blood-brain barrier (BBB) disruption, fibrinogen enters the CNS and is rapidly converted to fibrin, unfolding pleiotropic autoimmune mechanisms. Fibrin accumulation leads to subsequent proteolytic degradation that results in D-dimer generation. The primary objective of this study was to determine intrathecal levels of D-dimer in CSF as a measure of intrathecal coagulation cascade activation and to evaluate its diagnostic utility in patients with MS in contrast to healthy subjects. Key secondary objectives included analysis of CSF D-dimer in differential diagnoses of MS and its relation to routine clinical markers of disease activity.
Methods: Patients admitted for the assessment of suspected MS were prospectively recruited from October 2017 to December 2020. Blood plasma and citrated CSF samples were analyzed using a highly sensitive luminescent oxygen channeling immunoassay. Intrathecal generation of D-dimer was analyzed by adjusting for CSF/serum albumin (Qalb) and CSF/plasma D-dimer quotients (QD-dimer), and corresponding CSF fibrinogen levels were determined. Final diagnoses after full evaluation and clinical data were recorded.
Results: Of 187 patients, 113 patients received a diagnosis of MS or clinically/radiologically isolated syndrome. We found increased intrathecal CSF D-dimer generation levels (QD-dimer/Qalb-index) for patients with relapsing-remitting MS (RRMS; n = 71, median 4.7, interquartile range [IQR] 2.5–8.0) when compared with those for disease controls (n = 22, median 2.6, IQR 2.1–4.8, p = 0.031). Absolute CSF D-dimer values correlated with CSF fibrinogen levels (r = 0.463; p < 0 .001) and CSF leukocytes (r = 0.273; p = 0.003) and were elevated in MS patients with contrast enhancement (CE) compared with MS patients without CE on MRI (n = 48, median 6 ng/mL, and IQR 3–15.25 vs n = 41, median 4 ng/mL, and IQR 2–7; p = 0.026). Exploratory subgroup analyses indicated a correlation of intrathecal inflammatory activity and CSF D-dimer levels.
Discussion: D-dimer in CSF can be reliably determined and correlates with markers of CNS inflammation and CSF fibrinogen levels. Adjusted for BBB dysfunction, CSF D-dimer may allow the identification of intrathecal coagulation cascade activation in patients with MS.
Classification of Evidence: This study provides Class I evidence that CSF D-dimer levels are elevated in patients with RRMS.
Highlights
• Constrictional structures range from dome-and-basin folds to coeval folds and boudins.
• Under bulk constriction, the competent layer rotates slower than a passive plane.
• Extension-parallel and –perpendicular folds grow simultaneously.
• Extension-perpendicular folds affect previous boudins.
Abstract
We conducted scaled analogue modelling to show the influence of varying single layer initial orientation on the geometry of folds and boudins in a bulk constrictional strain field. The initial angle between the plane of shortening and the competent layer (θZ(i)) was incrementally increased from 0° to 90° by multiples of 11.25°. While the amount of layer thickening decreased with increasing θZ(i), the deformation structures produced range from pure dome-and-basin folds to coeval folds and boudins. Based on the attitude of fold axes, there are extension-parallel (FEPR) and extension-perpendicular (FEPP) folds, with axes subparallel and subperpendicular to the principal stretching axis (X), respectively. Coeval growth of FEPR folds and boudins occurred when θZ(i) > ca. 25°. The FEPP folds can be subdivided into a first type which affect the entire layer (if θZ(i) ranges between 11.25 and 78.75°) and a second type, referred to as FBEPP folds, which are affecting pre-existing boudins if θZ(i) > 45°. The interlimb angle of all types of folds increases with increasing θZ(i). Folds and boudins similar to the ones produced in this study can be found in salt domes and in tectonites of subduction zones.
Transfusion of red blood cells (RBC) in patients undergoing major elective cranial surgery is associated with increased morbidity, mortality and prolonged hospital length of stay (LOS). This retrospective single center study aims to identify the clinical outcome of RBC transfusions on skull base and non-skull base meningioma patients including the identification of risk factors for RBC transfusion. Between October 2009 and October 2016, 423 patients underwent primary meningioma resection. Of these, 68 (16.1%) received RBC transfusion and 355 (83.9%) did not receive RBC units. Preoperative anaemia rate was significantly higher in transfused patients (17.7%) compared to patients without RBC transfusion (6.2%; p = 0.0015). In transfused patients, postoperative complications as well as hospital LOS was significantly higher (p < 0.0001) compared to non-transfused patients. After multivariate analyses, risk factors for RBC transfusion were preoperative American Society of Anaesthesiologists (ASA) physical status score (p = 0.0247), tumor size (p = 0.0006), surgical time (p = 0.0018) and intraoperative blood loss (p < 0.0001). Kaplan-Meier curves revealed significant influence on overall survival by preoperative anaemia, RBC transfusion, smoking, cardiovascular disease, preoperative KPS ≤ 60% and age (elderly ≥ 75 years). We concluded that blood loss due to large tumors or localization near large vessels are the main triggers for RBC transfusion in meningioma patients paired with a potential preselection that masks the effect of preoperative anaemia in multivariate analysis. Further studies evaluating the impact of preoperative anaemia management for reduction of RBC transfusion are needed to improve the clinical outcome of meningioma patients.
Oral e-Poster Presentations - Booth 1: Vascular 3, September 27, 2023, 10:00 AM - 10:40 AM
Background: Despite current clinical guidelines recommending suboccipital decompressive craniectomy (SDC) in patients with space-occupying cerebellar infarction when neurological deterioration occurs, the precise definition of such deterioration remains unclear. The current study aimed at characterizing whether clinical outcomes can be predicted by the GCS score immediately prior to SDC, and whether higher GCS scores are associated with better clinical outcomes. We aimed to characterize whether clinical outcomes can be predicted by the GCS score immediately prior to SDC, and if higher GCS scores are associated with better clinical outcomes.
Methods: In a single-center, retrospective analysis of 51 patients treated with SDC for space-occupying cerebellar infarction clinical and imaging data were evaluated at the timepoints of symptom onset, hospital admission and preoperatively. Clinical outcome was measured by mRS at the last available follow-up. Preoperative GCS scores were stratified into three groups (GCS 3-8, 9-11 and 12-15). Univariate and multivariate Cox regression analyses were performed using clinical and radiological parameters as predictors of clinical outcome.
Results: In Cox-regression analysis using mRS of 1-2 as a positive clinical outcome we found a significant increase in the proportional hazard ratio (HR) of 6.581 [CI 1.839-36.414]; p=0.031 for GCS scores of 12-15 prior to SDC. Clinical outcomes (mRS 3-6) were associated with infarct volume above 6.0 cm3 (HR 2.473 [CI 1.209-5.057]; p=0.013), tonsillar herniation (HR: 0.279 [CI 0.083-0.933]; p=0.038), brainstem compression (HR 0.304 [CI 0.123-0.749]; p=0.010) and a preoperative GCS score of 3-8 (HR 2.386 [CI 1.160-4.906]; p=0.018).
Conclusions: SDC should be considered in patients with infarct volumes above 6.0 cm3 with GCS scores higher than previously described in the literature, as these patients may show better long-term outcome than those in which surgery is delayed until a GCS score of 11 or lower.
Importance: The entry of artificial intelligence into medicine is pending. Several methods have been used for the predictions of structured neuroimaging data, yet nobody compared them in this context.
Objective: Multi-class prediction is key for building computational aid systems for differential diagnosis. We compared support vector machine, random forest, gradient boosting, and deep feed-forward neural networks for the classification of different neurodegenerative syndromes based on structural magnetic resonance imaging.
Design, setting, and participants: Atlas-based volumetry was performed on multi-centric T1-weighted MRI data from 940 subjects, i.e., 124 healthy controls and 816 patients with ten different neurodegenerative diseases, leading to a multi-diagnostic multi-class classification task with eleven different classes.
Interventions: N.A.
Main outcomes and measures: Cohen’s kappa, accuracy, and F1-score to assess model performance.
Results: Overall, the neural network produced both the best performance measures and the most robust results. The smaller classes however were better classified by either the ensemble learning methods or the support vector machine, while performance measures for small classes were comparatively low, as expected. Diseases with regionally specific and pronounced atrophy patterns were generally better classified than diseases with widespread and rather weak atrophy.
Conclusions and relevance: Our study furthermore underlines the necessity of larger data sets but also calls for a careful consideration of different machine learning methods that can handle the type of data and the classification task best.
Cerebral lesions may cause degeneration and neuroplastic reorganization in both the ipsi- and the contralesional hemisphere, presumably creating an imbalance of primarily inhibitory interhemispheric influences produced via transcallosal pathways. The two hemispheres are thought to mutually hamper neuroplastic reorganization of the other hemisphere. The results of preceding degeneration and neuroplastic reorganization of white matter may be reflected by Diffusion Tensor Imaging-derived diffusivity parameters such as fractional anisotropy (FA). In this study, we applied Diffusion Tensor Imaging (DTI) to contrast the white matter status of the contralesional hemisphere of young lesioned brains with and without contralateral influences by comparing patients after hemispherotomy to those who had not undergone neurosurgery. DTI was applied to 43 healthy controls (26 females, mean age ± SD: 25.07 ± 11.33 years) and two groups of in total 51 epilepsy patients with comparable juvenile brain lesions (32 females, mean age ± SD: 25.69 ± 12.77 years) either after hemispherotomy (30 of 51 patients) or without neurosurgery (21 of 51 patients), respectively. FA values were compared between these groups using the unbiased tract-based spatial statistics approach. A voxel-wise ANCOVA controlling for age at scan yielded significant group differences in FA. A post hoc t-test between hemispherotomy patients and healthy controls revealed widespread supra-threshold voxels in the contralesional hemisphere of hemispherotomy patients indicating comparatively higher FA values (p < 0.05, FWE-corrected). The non-surgery group, in contrast, showed extensive supra-threshold voxels indicating lower FA values in the contralesional hemisphere as compared to healthy controls (p < 0.05, FWE-corrected). Whereas lower FA values are suggestive of pronounced contralesional degeneration in the non-surgery group, higher FA values in the hemispherotomy group may be interpreted as a result of preceding plastic remodeling. We conclude that, whether juvenile brain lesions are associated with contralesional degeneration or reorganization partly depends on the ipsilesional hemisphere. Contralesional reorganization as observed in hemispherotomy patients was most likely enabled by the complete neurosurgical deafferentation of the ipsilesional hemisphere and, thereby, the disinhibition of the neuroplastic potential of the contralesional hemisphere. The main argument of this study is that hemispherotomy may be seen as a major plastic stimulus and as a prerequisite for contralesional neuroplastic remodeling in patients with juvenile brain lesions.