Showing papers by "Johannes Slotboom published in 2014"

Multi-modal glioblastoma segmentation: man versus machine.

Abstract: BACKGROUND AND PURPOSE Reproducible segmentation of brain tumors on magnetic resonance images is an important clinical need. This study was designed to evaluate the reliability of a novel fully automated segmentation tool for brain tumor image analysis in comparison to manually defined tumor segmentations. METHODS We prospectively evaluated preoperative MR Images from 25 glioblastoma patients. Two independent expert raters performed manual segmentations. Automatic segmentations were performed using the Brain Tumor Image Analysis software (BraTumIA). In order to study the different tumor compartments, the complete tumor volume TV (enhancing part plus non-enhancing part plus necrotic core of the tumor), the TV+ (TV plus edema) and the contrast enhancing tumor volume CETV were identified. We quantified the overlap between manual and automated segmentation by calculation of diameter measurements as well as the Dice coefficients, the positive predictive values, sensitivity, relative volume error and absolute volume error. RESULTS Comparison of automated versus manual extraction of 2-dimensional diameter measurements showed no significant difference (p = 0.29). Comparison of automated versus manual segmentation of volumetric segmentations showed significant differences for TV+ and TV (p 0.05) with regard to the Dice overlap coefficients. Spearman's rank correlation coefficients (ρ) of TV+, TV and CETV showed highly significant correlations between automatic and manual segmentations. Tumor localization did not influence the accuracy of segmentation. CONCLUSIONS In summary, we demonstrated that BraTumIA supports radiologists and clinicians by providing accurate measures of cross-sectional diameter-based tumor extensions. The automated volume measurements were comparable to manual tumor delineation for CETV tumor volumes, and outperformed inter-rater variability for overlap and sensitivity.

5-ALA complete resections go beyond MR contrast enhancement: shift corrected volumetric analysis of the extent of resection in surgery for glioblastoma.

Abstract: Background The technique of 5-aminolevulinic acid (5-ALA) tumor fluorescence is increasingly used to improve visualization of tumor tissue and thereby to increase the rate of patients with gross total resections. In this study, we measured the resection volumes in patients who underwent 5-ALA-guided surgery for non-eloquent glioblastoma and compared them with the preoperative tumor volume.

Patient-Specific Semi-Supervised Learning for Postoperative Brain Tumor Segmentation

Abstract: In contrast to preoperative brain tumor segmentation, the problem of postoperative brain tumor segmentation has been rarely approached so far. We present a fully-automatic segmentation method using multimodal magnetic resonance image data and patient-specific semi-supervised learning. The idea behind our semi-supervised approach is to effectively fuse information from both pre- and postoperative image data of the same patient to improve segmentation of the postoperative image. We pose image segmentation as a classification problem and solve it by adopting a semi-supervised decision forest. The method is evaluated on a cohort of 10 high-grade glioma patients, with segmentation performance and computation time comparable or superior to a state-of-the-art brain tumor segmentation method. Moreover, our results confirm that the inclusion of preoperative MR images lead to a better performance regarding postoperative brain tumor segmentation.

Atorvastatin Added to Interferon Beta for Relapsing Multiple Sclerosis: 12-Month Treatment Extension of the Randomized Multicenter SWABIMS Trial

Abstract: BACKGROUND Statins have anti-inflammatory and immunomodulatory properties in addition to lipid-lowering effects. OBJECTIVES To report the 12-month extension of a phase II trial evaluating the efficacy, safety and tolerability of atorvastatin 40 mg/d added to interferon beta-1b (IFNB-1b) in relapsing-remitting multiple sclerosis (RRMS). METHODS In the randomized, multicenter, parallel-group, rater-blinded core study, 77 RRMS patients started IFNB-1b. At month three they were randomized 1∶1 to receive atorvastatin 40 mg/d or not in addition to IFNB-1b until month 15. In the subsequent extension study, patients continued with unchanged medication for another 12 months. Data at study end were compared to data at month three of the core study. RESULTS 27 of 72 patients that finished the core study entered the extension study. 45 patients were lost mainly due to a safety analysis during the core study including a recruitment stop for the extension study. The primary end point, the proportion of patients with new lesions on T2-weighted images was equal in both groups (odds ratio 1.926; 95% CI 0.265-14.0007; p = 0.51). All secondary endpoints including number of new lesions and total lesion volume on T2-weighted images, total number of Gd-enhancing lesions on T1-weighted images, volume of grey and white matter, EDSS, MSFC, relapse rate, number of relapse-free patients and neutralizing antibodies did not show significant differences either. The combination therapy was well tolerated. CONCLUSIONS Atorvastatin 40 mg/day in addition to IFNB-1b did not have any beneficial effects on RRMS compared to IFNB-1b monotherapy over a period of 24 months.

Dose reduction in standard head CT: first results from a new scanner using iterative reconstruction and a new detector type in comparison with two previous generations of multi-slice CT.

Abstract: Purpose Computed tomography (CT) accounts for more than half of the total radiation exposure from medical procedures, which makes dose reduction in CT an effective means of reducing radiation exposure. We analysed the dose reduction that can be achieved with a new CT scanner [Somatom Edge (E)] that incorporates new developments in hardware (detector) and software (iterative reconstruction).

Interactive segmentation of MR images from brain tumor patients

Abstract: Medical doctors often do not trust the result of fully automatic segmentations because they have no possibility to make corrections if necessary. On the other hand, manual corrections can introduce a user bias. In this work, we propose to integrate the possibility for quick manual corrections into a fully automatic segmentation method for brain tumor images. This allows for necessary corrections while maintaining a high objectiveness. The underlying idea is similar to the well-known Grab-Cut algorithm, but here we combine decision forest classification with conditional random field regularization for interactive segmentation of 3D medical images. The approach has been evaluated by two different users on the BraTS2012 dataset. Accuracy and robustness improved compared to a fully automatic method and our interactive approach was ranked among the top performing methods. Time for computation including manual interaction was less than 10 minutes per patient, which makes it attractive for clinical use.

P16.03unifying clinical routine brain tumor mr-spectroscopy and mr-image analysis: novel jmrui plug-ins for brain tumor analysis

Abstract: 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.

O2.03target delineation in glioblastoma: is preoperative automatic comparable to expert based segmentation?

Abstract: INTRODUCTION: To evaluate preoperative automatic structure segmentation of glioblastoma patients undergoing radiation therapy in clinical practice. METHOD: Automatic tumor segmentation was performed on preoperative MR images using the Brain Tumor Software Analysis (BraTumIA) software (www.istb.unibe.ch). We compared manual tumor volumetry based on postoperative MR images of 14 GBM patients (7 patients received 60 Gy/30 fractions; 7 other patients received 39.9 Gy/15 fractions) to the automatic approach. We employed four standard MRI sequences (T1, T1contrast, T2 and FLAIR) in order to segment the tumor sub-compartments (contrast enhancing tumor; CET, necrosis, non enhancing tumor; NET and edema). Dice Similarity Coefficient (DSC) and Mean Surface Distance (MSD) were calculated to evaluate the quality of segmentations. For the evaluation of the dose deposition in GTV, CTV, CET, necrosis, NET and edema, an analysis of the following dose-volume histogram (DVH) parameters has been performed: D2%, D90%, D95%, D98% (dose exceeded in 2%, 90%, 95%, 98% of the volume) and Dmean (mean dose). Furthermore, the same DVH analysis of the composite structure “complete tumor volume +” (TV +) encompassing the CET, the necrosis, the NET and the edema has also been carried out. Based on these histograms a homogeneity index (HI) has been defined as HI = (D2% - D98%) / Dp * 100 where Dp is the prescribed dose. RESULT: DSC and MSD mean/SD overlap between automatic segmentation and manual contours was 0.52 ± 0.20; 5.1 ± 4.0 mm for necrotic tissue combined with CET vs GTV;0.62 ± 0.18 and 8.2 ± 5.1 mm for TV+ vs CTV, respectively. No significant differences were observed between means of D2% D90% D95%, D98% and Dmean in GTV, CET and necrosis. Differences were found between the means of D98% in CTV60Gy and TV + 60Gy (p = 0.006) and CTV39.9Gy and TV + 39.9Gy (p = 0.033). Despite a good mean HI in PTV (8.1 ± 3.0) %, we found a higher mean value for TV+ (49.7 ± 41.1)%. These findings were due to false positives in the segmentation of the edema compartment of the TV+ in 4 out of 14 patients. This could indicate a weakness of BraTumIA in the segmentation of the edema compartment. CONCLUSION: Automated volumetry on preoperative MR revealed only subtle volume differences for the necrotic and enhancing tumor segments and postoperatively delineated GTV, whereas non-enhancing tumor plus edema volumes vs CTV were discordant. The DVH evaluation highlighted an underdosage of the edema in the D98% regions causing high HI values. Our data suggest that preoperative automatic volumetry adds converging information to adapt CTV of patients with GBM (including CET, necrosis and NET). The incorporation of the complete edema into the target volume has still to be addressed.

Atlas-Based Segmentation of Tumor-Bearing Brain Images

Abstract: In diagnostic neuroradiology as well as in radiation oncology and neurosurgery, there is an increasing demand for accurate segmentation of tumor-bearing brain images. Atlas-based segmentation is an appealing automatic technique thanks to its robustness and versatility. However, atlas-based segmentation of tumor-bearing brain images is challenging due to the confounding effects of the tumor in the patient image. In this article, we provide a brief background on brain tumor imaging and introduce the clinical perspective, before we categorize and review the state of the art in the current literature on atlas-based segmentation for tumor-bearing brain images. We also present selected methods and results from our own research in more detail. Finally, we conclude with a short summary and look at new developments in the field, including requirements for future routine clinical use.