Search

Patients with Glioblastoma multiforme (GBM) have a very low overall survival (OS) time, due to the rapid growth an invasiveness of this brain tumor. As a contribution to the overall survival (OS) prediction task within the Brain Tumor Segmentation Challenge (BraTS), we classify the OS of GBM patients into overall survival classes based on information derived from pre-treatment Magnetic Resonance Imaging (MRI). The top-ranked methods from the past years almost exclusively used shape and position features. This is a remarkable contrast to the current advances in GBM radiomics showing a benefit of intensity-based features. This discrepancy may be caused by the inconsistent acquisition parameters in a multi-center setting. In this contribution, we test if normalizing the images based on the healthy tissue intensities enables the robust use of intensity features in this challenge. Based on these normalized images, we test the performance of 176 combinations of feature selection techniques and classifiers. Additionally, we test the incorporation of a sequence and robustness prior to limit the performance drop when models are applied to unseen data. The most robust performance on the training data (accuracy: \(0.52\pm 0.09\)) was achieved with random forest regression, but this accuracy could not be maintained on the test set.

Background Automated brain tumor segmentation methods are computational algorithms that yield tumor delineation from, in this case, multimodal magnetic resonance imaging (MRI). We present an automated segmentation method and its results for resection cavity (RC) in glioblastoma multiforme (GBM) patients using deep learning (DL) technologies. Methods Post-operative, T1w with and without contrast, T2w and fluid attenuated inversion recovery MRI studies of 30 GBM patients were included. Three radiation oncologists manually delineated the RC to obtain a reference segmentation. We developed a DL cavity segmentation method, which utilizes all four MRI sequences and the reference segmentation to learn to perform RC delineations. We evaluated the segmentation method in terms of Dice coefficient (DC) and estimated volume measurements. Results Median DC of the three radiation oncologist were 0.85 (interquartile range [IQR]: 0.08), 0.84 (IQR: 0.07), and 0.86 (IQR: 0.07). The results of the automatic segmentation compared to the three different raters were 0.83 (IQR: 0.14), 0.81 (IQR: 0.12), and 0.81 (IQR: 0.13) which was significantly lower compared to the DC among raters (chi-square = 11.63, p = 0.04). We did not detect a statistically significant difference of the measured RC volumes for the different raters and the automated method (Kruskal-Wallis test: chi-square = 1.46, p = 0.69). The main sources of error were due to signal inhomogeneity and similar intensity patterns between cavity and brain tissues. Conclusions The proposed DL approach yields promising results for automated RC segmentation in this proof of concept study. Compared to human experts, the DC are still subpar.

Additional file 1. The supplementary material contains further information on the data used, methods, software tools and versions, and additional experiment results to improve reproducibility.

Disease progression for Glioblastoma multiforme patients is currently assessed with manual bi-dimensional measurements of the active contrast-enhancing tumor on Magnetic Resonance Images (MRI). This method is known to be susceptible to error; in the lack of a data-driven approach, progression thresholds had been set rather arbitrarily considering measurement inaccuracies. We propose a data-driven methodology for disease progression assessment, building on tumor volumetry, classical radiomics, and deep learning based features. For each feature type, we infer progression thresholds by maximizing the correlation of the time-to-progression (TTP) and overall survival (OS). On a longitudinal study comprising over 500 data points, we observed considerable underestimation of the current volumetric disease progression threshold. We evaluate the data-driven disease progression thresholds based on expert ratings using the current clinical practice.

PURPOSE To improve the detection of peritumoral changes in GBM patients by exploring the relation between MRSI information and the distance to the solid tumor volume (STV) defined using structural MRI (sMRI). METHODS Twenty-three MRSI studies (PRESS, TE 135 ms) acquired from different patients with untreated GBM were used in this study. For each MRSI examination, the STV was identified by segmenting the corresponding sMRI images using BraTumIA, an automatic segmentation method. The relation between different metabolite ratios and the distance to STV was analyzed. A regression forest was trained to predict the distance from each voxel to STV based on 14 metabolite ratios. Then, the trained model was used to determine the expected distance to tumor (EDT) for each voxel of the MRSI test data. EDT maps were compared against sMRI segmentation. RESULTS The features showing abnormal values at the longest distances to the tumor were: %NAA, Glx/NAA, Cho/NAA, and Cho/Cr. These four features were also the most important for the prediction of the distances to STV. Each EDT value was associated with a specific metabolic pattern, ranging from normal brain tissue to actively proliferating tumor and necrosis. Low EDT values were highly associated with malignant features such as elevated Cho/NAA and Cho/Cr. CONCLUSION The proposed method enables the automatic detection of metabolic patterns associated with different distances to the STV border and may assist tumor delineation of infiltrative brain tumors such as GBM.

OBJECTIVEIn the treatment of glioblastoma, residual tumor burden is the only prognostic factor that can be actively influenced by therapy. Therefore, an accurate, reproducible, and objective measurement of residual tumor burden is necessary. This study aimed to evaluate the use of a fully automatic segmentation method—brain tumor image analysis (BraTumIA)—for estimating the extent of resection (EOR) and residual tumor volume (RTV) of contrast-enhancing tumor after surgery.METHODSThe imaging data of 19 patients who underwent primary resection of histologically confirmed supratentorial glioblastoma were retrospectively reviewed. Contrast-enhancing tumors apparent on structural preoperative and immediate postoperative MR imaging in this patient cohort were segmented by 4 different raters and the automatic segmentation BraTumIA software. The manual and automatic results were quantitatively compared.RESULTSFirst, the interrater variabilities in the estimates of EOR and RTV were assessed for all human raters. Interrater agreement in terms of the coefficient of concordance (W) was higher for RTV (W = 0.812; p < 0.001) than for EOR (W = 0.775; p < 0.001). Second, the volumetric estimates of BraTumIA for all 19 patients were compared with the estimates of the human raters, which showed that for both EOR (W = 0.713; p < 0.001) and RTV (W = 0.693; p < 0.001) the estimates of BraTumIA were generally located close to or between the estimates of the human raters. No statistically significant differences were detected between the manual and automatic estimates. BraTumIA showed a tendency to overestimate contrast-enhancing tumors, leading to moderate agreement with expert raters with respect to the literature-based, survival-relevant threshold values for EOR.CONCLUSIONSBraTumIA can generate volumetric estimates of EOR and RTV, in a fully automatic fashion, which are comparable to the estimates of human experts. However, automated analysis showed a tendency to overestimate the volume of a contrast-enhancing tumor, whereas manual analysis is prone to subjectivity, thereby causing considerable interrater variability.

Deep learning for regression tasks on medical imaging data has shown promising results. However, compared to other approaches, their power is strongly linked to the dataset size. In this study, we evaluate 3D-convolutional neural networks (CNNs) and classical regression methods with hand-crafted features for survival time regression of patients with high grade brain tumors. The tested CNNs for regression showed promising but unstable results. The best performing deep learning approach reached an accuracy of 51.5% on held-out samples of the training set. All tested deep learning experiments were outperformed by a Support Vector Classifier (SVC) using 30 radiomic features. The investigated features included intensity, shape, location and deep features. The submitted method to the BraTS 2018 survival prediction challenge is an ensemble of SVCs, which reached a cross-validated accuracy of 72.2% on the BraTS 2018 training set, 57.1% on the validation set, and 42.9% on the testing set. The results suggest that more training data is necessary for a stable performance of a CNN model for direct regression from magnetic resonance images, and that non-imaging clinical patient information is crucial along with imaging information., Contribution to The International Multimodal Brain Tumor Segmentation (BraTS) Challenge 2018, survival prediction task

Introduction Anaplastic pilocytic astrocytoma (APA) is an exceptionally rare type of high-grade glioma in adults. Establishing histopathological diagnosis is challenging and its clinical and radiological appearance insidious. By this case series and first literature review we investigated the various clinical, neuroradiological, and histopathological features of APA in adults. Methods An in hospital screening of the database from the Institute of Pathology was conducted to identify cases of APA. Further, we performed a literature review in PubMed using the keywords “anaplastic/malignant/atypical AND pilocytic astrocytoma” and “anaplastic astrocytoma/glioblastoma AND Rosenthal fibers” and summarized the current knowledge about APA in adults. Results Over the last decade we were able to identify 3 adult patients with APA in our hospital. According to the pertinent literature, the prognosis of APA in adults (documented survival of up to 10 years) appears to be better than in other high-grade gliomas. Few cases were associated with neurofibromatosis type 1, which seems to predispose for development of APA. Although molecular genetics is still of limited value for differentiation of APA from other high-grade glioma, advanced neuroimaging techniques such as magnetic resonance perfusion imaging and spectroscopy allow improved differential work-up. In particular, APA in adults has the ability to mimic various neurological diseases such as tumefactive demyelinating lesions, low-, or high-grade gliomas. Conclusions Although currently not explicitly recognized as a distinct clinico-pathologic entity it seems that adult APA behaves differently from conventional high-grade glioma and should be included in differential diagnostics to enable adequate patient care. However, further studies are needed to better understand this extremely rare disease.

MRSI grids frequently show spectra with poor quality, mainly because of the high sensitivity of MRS to field inhomogeneities. These poor quality spectra are prone to quantification and/or interpretation errors that can have a significant impact on the clinical use of spectroscopic data. Therefore, quality control of the spectra should always precede their clinical use. When performed manually, quality assessment of MRSI spectra is not only a tedious and time-consuming task, but is also affected by human subjectivity. Consequently, automatic, fast and reliable methods for spectral quality assessment are of utmost interest. In this article, we present a new random forest-based method for automatic quality assessment of (1)H MRSI brain spectra, which uses a new set of MRS signal features. The random forest classifier was trained on spectra from 40 MRSI grids that were classified as acceptable or non-acceptable by two expert spectroscopists. To account for the effects of intra-rater reliability, each spectrum was rated for quality three times by each rater. The automatic method classified these spectra with an area under the curve (AUC) of 0.976. Furthermore, in the subset of spectra containing only the cases that were classified every time in the same way by the spectroscopists, an AUC of 0.998 was obtained. Feature importance for the classification was also evaluated. Frequency domain skewness and kurtosis, as well as time domain signal-to-noise ratios (SNRs) in the ranges 50-75 ms and 75-100 ms, were the most important features. Given that the method is able to assess a whole MRSI grid faster than a spectroscopist (approximately 3 s versus approximately 3 min), and without loss of accuracy (agreement between classifier trained with just one session and any of the other labelling sessions, 89.88%; agreement between any two labelling sessions, 89.03%), the authors suggest its implementation in the clinical routine. The method presented in this article was implemented in jMRUI's SpectrIm plugin.

To improve the detection of peritumoral changes in GBM patients by exploring the relation between MRSI information and the distance to the solid tumor volume (STV) defined using structural MRI (sMRI).Twenty-three MRSI studies (PRESS, TE 135 ms) acquired from different patients with untreated GBM were used in this study. For each MRSI examination, the STV was identified by segmenting the corresponding sMRI images using BraTumIA, an automatic segmentation method. The relation between different metabolite ratios and the distance to STV was analyzed. A regression forest was trained to predict the distance from each voxel to STV based on 14 metabolite ratios. Then, the trained model was used to determine the expected distance to tumor (EDT) for each voxel of the MRSI test data. EDT maps were compared against sMRI segmentation.The features showing abnormal values at the longest distances to the tumor were: %NAA, Glx/NAA, Cho/NAA, and Cho/Cr. These four features were also the most important for the prediction of the distances to STV. Each EDT value was associated with a specific metabolic pattern, ranging from normal brain tissue to actively proliferating tumor and necrosis. Low EDT values were highly associated with malignant features such as elevated Cho/NAA and Cho/Cr.The proposed method enables the automatic detection of metabolic patterns associated with different distances to the STV border and may assist tumor delineation of infiltrative brain tumors such as GBM.

Supplemental_Material for Computer-aided radiological diagnostics improves the preoperative diagnoses of medulloblastoma, pilocytic astrocytoma, and ependymoma: A reproducibility study by Nicole Porz, Urspeter Knecht, Beate Sick, Elvis Murina, Nuno Barros, Philippe Schucht, Evelyn Herrmann, Jan Gralla, Roland Wiest, Marwan El-Koussy, and Johannes Slotboom in Clinical and Translational Neuroscience

Uncertainty measures of medical image analysis technologies, such as deep learning, are expected to facilitate their clinical acceptance and synergies with human expertise. Therefore, we propose a full-resolution residual convolutional neural network (FRRN) for brain tumor segmentation and examine the principle of Monte Carlo (MC) Dropout for uncertainty quantification by focusing on the Dropout position and rate. We further feed the resulting brain tumor segmentation into a survival prediction model, which is built on age and a subset of 26 image-derived geometrical features such as volume, volume ratios, surface, surface irregularity and statistics of the enhancing tumor rim width. The results show comparable segmentation performance between MC Dropout models and a standard weight scaling Dropout model. A qualitative evaluation further suggests that informative uncertainty can be obtained by applying MC Dropout after each convolution layer. For survival prediction, results suggest only using few features besides age. In the BraTS17 challenge, our method achieved the 2nd place in the survival task and completed the segmentation task in the 3rd best-performing cluster of statistically different approaches.

OBJECTIVE Comparison of a fully-automated segmentation method that uses compartmental volume information to a semi-automatic user-guided and FDA-approved segmentation technique. METHODS Nineteen patients with a recently diagnosed and histologically confirmed glioblastoma (GBM) were included and MR images were acquired with a 1.5 T MR scanner. Manual segmentation for volumetric analyses was performed using the open source software 3D Slicer version 4.2.2.3 (www.slicer.org). Semi-automatic segmentation was done by four independent neurosurgeons and neuroradiologists using the computer-assisted segmentation tool SmartBrush® (referred to as SB), a semi-automatic user-guided and FDA-approved tumor-outlining program that uses contour expansion. Fully automatic segmentations were performed with the Brain Tumor Image Analysis (BraTumIA, referred to as BT) software. We compared manual (ground truth, referred to as GT), computer-assisted (SB) and fully-automated (BT) segmentations with regard to: (1) products of two maximum diameters for 2D measurements, (2) the Dice coefficient, (3) the positive predictive value, (4) the sensitivity and (5) the volume error. RESULTS Segmentations by the four expert raters resulted in a mean Dice coefficient between 0.72 and 0.77 using SB. BT achieved a mean Dice coefficient of 0.68. Significant differences were found for intermodal (BT vs. SB) and for intramodal (four SB expert raters) performances. The BT and SB segmentations of the contrast-enhancing volumes achieved a high correlation with the GT. Pearson correlation was 0.8 for BT; however, there were a few discrepancies between raters (BT and SB 1 only). Additional non-enhancing tumor tissue extending the SB volumes was found with BT in 16/19 cases. The clinically motivated sum of products of diameters measure (SPD) revealed neither significant intermodal nor intramodal variations. The analysis time for the four expert raters was faster (1 minute and 47 seconds to 3 minutes and 39 seconds) than with BT (5 minutes). CONCLUSION BT and SB provide comparable segmentation results in a clinical setting. SB provided similar SPD measures to BT and GT, but differed in the volume analysis in one of the four clinical raters. A major strength of BT may its independence from human interactions, it can thus be employed to handle large datasets and to associate tumor volumes with clinical and/or molecular datasets ("-omics") as well as for clinical analyses of brain tumor compartment volumes as baseline outcome parameters. Due to its multi-compartment segmentation it may provide information about GBM subcompartment compositions that may be subjected to clinical studies to investigate the delineation of the target volumes for adjuvant therapies in the future.

Information about the size of a tumor and its temporal evolution is needed for diagnosis as well as treatment of brain tumor patients. The aim of the study was to investigate the potential of a fully-automatic segmentation method, called BraTumIA, for longitudinal brain tumor volumetry by comparing the automatically estimated volumes with ground truth data acquired via manual segmentation. Longitudinal Magnetic Resonance (MR) Imaging data of 14 patients with newly diagnosed glioblastoma encompassing 64 MR acquisitions, ranging from preoperative up to 12 month follow-up images, was analysed. Manual segmentation was performed by two human raters. Strong correlations (R = 0.83–0.96, p T2-hyperintense tumor compartments (NCE-T2). A quantitative analysis of the inter-rater disagreement showed that the disagreement between BraTumIA and each of the human raters was comparable to the disagreement between the human raters. In summary, BraTumIA generated volumetric trend curves of contrast-enhancing and non-enhancing T2-hyperintense tumor compartments comparable to estimates of human raters. These findings suggest the potential of automated longitudinal tumor segmentation to substitute manual volumetric follow-up of contrast-enhancing and non-enhancing T2-hyperintense tumor compartments.

Introduction:Imaging-based diagnosis of intra-axial contrast-enhancing brain tumors is frequently challenging. We show that the diagnosis of medulloblastoma (MDB) versus pilocytic astrocytoma (PA) ...

Die radiologischen Schlagzeilen gehorten zuletzt der Computertomografie. Sie wird immer schneller und genauer. Trotzdem ist der „gute alte” Rontgenthorax wegen seiner Verbreitung insgesamt gesehen die bedeutendere Untersuchung geblieben. Radiologe Dr. Urspeter Knecht aus Bern erklart Ihnen, wie Sie dieses „Urgestein” der Radiologie richtig beurteilen.

MRSI grids frequently show spectra with poor quality, mainly because of the high sensitivity of MRS to field inhomogeneities. These poor quality spectra are prone to quantification and/or interpretation errors that can have a significant impact on the clinical use of spectroscopic data. Therefore, quality control of the spectra should always precede their clinical use. When performed manually, quality assessment of MRSI spectra is not only a tedious and time-consuming task, but is also affected by human subjectivity. Consequently, automatic, fast and reliable methods for spectral quality assessment are of utmost interest. In this article, we present a new random forest-based method for automatic quality assessment of (1)H MRSI brain spectra, which uses a new set of MRS signal features. The random forest classifier was trained on spectra from 40 MRSI grids that were classified as acceptable or non-acceptable by two expert spectroscopists. To account for the effects of intra-rater reliability, each spectrum was rated for quality three times by each rater. The automatic method classified these spectra with an area under the curve (AUC) of 0.976. Furthermore, in the subset of spectra containing only the cases that were classified every time in the same way by the spectroscopists, an AUC of 0.998 was obtained. Feature importance for the classification was also evaluated. Frequency domain skewness and kurtosis, as well as time domain signal-to-noise ratios (SNRs) in the ranges 50-75 ms and 75-100 ms, were the most important features. Given that the method is able to assess a whole MRSI grid faster than a spectroscopist (approximately 3 s versus approximately 3 min), and without loss of accuracy (agreement between classifier trained with just one session and any of the other labelling sessions, 89.88%; agreement between any two labelling sessions, 89.03%), the authors suggest its implementation in the clinical routine. The method presented in this article was implemented in jMRUI's SpectrIm plugin.

INTRODUCTION: This prospective study is derived from a framework investigating effects of tumor progression on MRI/MRS-parameters in high grade gliomas (WHO III-IV) in patients undergoing different types of neuro-oncologic therapies. Preliminary data are presented here. We aim to identify MR-related parameters that segregate a. real tumor progression from pseudo tumor progression and b. real tumor response from pseudo-response, two effects which complicate appropriate tumor diagnostics. Forty-nine patients are currently enrolled into the study. The MRI/MRS findings of a subgroup of 9 patients who received a combined irradiation (39.9 Gy in 15 fractions) with the angiogenesis inhibitor Bevacizumab are discussed here. METHODS: All patients received multi parametric MRI encompassing DWI (ADC & DTI), Perfusion Imaging (PI), unenhanced and Gd-enhanced 3D-T1w, 3D-T2w, T2w-FLAIR, and MRS (Single Voxel Spectroscopy (SVS) and Chemical Shift Imaging (CSI), both at TE 30ms and TE135ms) before and after administration of Bevacizumab. The effects of Bevacizumab on PI, Gd-enhanced T1w and MRS was examined. For the evaluation of the MRS we used the Cho/Cr and the Cho/NAA peak-area ratio of qualitative MRS, for the Gd-enhanced T1w and PI we used a semi-quantitative classification. For statistical analysis T-Tests was performed. The study was approved by our local ethics committee. RESULTS: Pre-treatment with Bevacizumab, three and six month follow up MRIs revealed significant reductions in Gd-enhanced T1w MRI (p < 0.03). No significant differences were detected for MRS and PI (regional cerebral blood volume, rCBV). Statistical significant differences were also detected between preoperative and postoperative MRI for Gd-enhanced T1w and rCBV for all five time points during follow up (p < 0.011). However, MRS findings showed no improvement in all MRS parameters during follow up. DISCUSSION AND CONCLUSION: No significant improvements in spectroscopic parameters could be observed in response to Bevacizumab treatment. In contrast, the improvement in Gd-enhanced T1w MRI as well as MRI Perfusion does not reflect the true tumor response state in Bevacizumab treated patients. MRS findings seem to reflect tumor progression state better than MRI under Bevacizumab treatment. Our findings are in keeping with the current literature.

Central nervous system aspergillosis is an often fatal complication of invasive Aspergillus infection. Relevant disease models are needed to study the pathophysiology of cerebral aspergillosis and to develop novel therapeutic approaches. This study presents a model of central nervous system aspergillosis that mimics important aspects of human disease. Eleven-day-old non-immunosuppressed male Wistar rats were infected by an intracisternal injection of 10 mul of a conidial suspension of Aspergillus fumigatus. An inoculum of 7.18 log(10) colony-forming units (CFU) consistently produced cerebral infection and resulted in death of all animals (n = 25) within 3-10 days. Median survival time was 3 days. Histomorphologically, all animals developed intracerebral abscesses (2-26 per brain) containing abundant fungal hyphae and neutrophils. Fungal culture of cortical homogenates yielded maximal growth on day 3 after infection (5.4 log(10) CFU/g, n = 15) that declined over time. Galactomannan concentrations in cortical homogenates, assessed as an index for hyphal burden, peaked on days 3-5. Fungal infection spread to peripheral organs in 83% of animals. Fungal burden in lung, liver, spleen and kidney was two orders of magnitude lower than in the brain. The successful establishment of a model of cerebral aspergillosis in a non-immunosuppressed host provides the opportunity to investigate mechanisms of disease and to develop novel treatment regimens for this commonly fatal infection.

INTRODUCTION: Magnetic resonance imaging and spectroscopy are the neuroradiological methods of first choice for the diagnostics of de novo brain tumors, the evaluation tumor response to therapy and the tumor progression. Whereas (i.) T2 and post contrast T1-weighted MRI gives information on the brain tumors' anatomy and integrity of the blood brain barrier, (ii.) perfusion weighted MRI on the perfusion state (important to study processes like neo-angiogenis, apoptosis and necrosis), (iii.) diffusion weighted MRI (evaluation for cellular density, and white matter tract integrity), information on the tumors' metabolism is obtained by MR-spectroscopy. Lactate gives information on the ischemic state, choline/ mobile lipids on the membrane turnover and necrosis/apoptosis. The lactate, as present in many high grade glioma, is an indicator an ischemic condition which can ge regarded as an indicator for tumors resistance to radiation therapy. However, despite the fact that MRS gives valuable additional information to MRI, the available software for clinical routine analysis of MRS data together with MRI data is far from ideal. This abstracts reports on the software, which is currently being developed within an EU-funded Marie Curie Initial Training Network (ITN) (http://www.transact-itn.eu/) named TRANSACT which stands for “Transforming Magnetic Resonance Spectroscopy into a Clinical Tool”. METHODS: Basis for the developed software within the TRANSACT-project is the software package jMRUI, which was developed during former EU-funded projects, and targeted mainly on scientific users of MR-spectroscopy; the current TRANSACT project however focuses on the develop software for clinicians. The novel plug-ins for jMRUI were entirely developed in JAVA. RESULTS: The following important clinical work flow related requirements were incorporated: (a.) full support of DICOM image and spectroscopy data format; (b.) a fully featured integrated patient/study/series examination browser; (c.) automatic projection of spectral voxels within automatically loaded reference image stacks; (d.) DICOM transfer using the standard DICOM network data transfer protocol; (e.) DICOM reporting of spectroscopy results, and possibility to transfer of these results into PACS systems; (f.) absolute quantification of single voxel spectra; (g.) advanced display of spectroscopic SVS and MRSI data; (h.) automated fast metabolite image generation (i.) easy correlation of numeric image data and parametric spectral data. CONCLUSION: An intuitive software has been developed for a clinical setting that allows the simutaneous study of MRI/MRS brain tumor data.

BACKGROUND: We present results of an automated segmentation method for resection cavity in Glioblastoma (GBM) patients using deep learning technologies. The approach fully segments the resection cavity and yields estimates of uncertainty. We compare the approach to a previously reported machine learning method and show the superiority in performance of the deep learning approach while yielding estimates of segmentation uncertainty, which can be used by an operator to monitor computer’s results. MATERIAL AND METHODS: We included post-OP, pre- and post-contrast T1, T2 and FLAIR MRI datasets of 30 patients with newly diagnosed, surgically resected and histologically confirmed GBM. Radiation Cavity (RC) were defined as surgical cavity plus blood products. Three radiation oncologists manually delineated the RC using all four MRI sequences. For fully automatic segmentation, we developed a deep learning cavity segmentation method, which was trained on a subset of 25 cases, and utilizes all four MRI sequences to learn RC delineations. We evaluated the segmentation method in terms of Dice coefficient (DC) (volumetric overlap) and estimated volume measurements using the Kruskal-Wallis test followed by a 6-fold cross-validation using the imaging data. RESULTS: Median DC and interquartile range (IQR) of the different pairings of expert raters (reported as tuples) were (0.85, 0.08), (0.84, 0.07), and (0.86, 0.07). The results of the automatic segmentation compared to the three different raters were (0.83, 0.14), (0.81, 0.12), and (0.81, 0.13). We did not detect a statistically significant difference regarding the distribution of the measured volumes for the different raters and methods (Kruskal-Wallis test: chi-square=1.46, p=0.69). The main sources of error consisted in over- or under-segmentation of the RC due to signal inhomogeneity (especially in T2 and FLAIR sequences) and over-segmentation due to similar intensity patterns (edema, subarachnoid space, or ventricles). Uncertainty maps generated by the deep learning approach were visually compared against the segmentation errors, and a good spatial overlap was found between uncertainty and RC delineation errors. This allow us to conclude on the reliability and good calibration properties of the deep learning approach. CONCLUSION: The proposed deep learning approach yields good results for the segmentation of the RC. However, blood products and air can affect its performance. Compared to human experts, the computer-generated results are still subpar but outperform previously presented approaches, while providing estimates of segmentation uncertainty, which can leverage the interactions between these systems and human experts.