Clinical Evaluation of a Fully-automatic Segmentation Method for Longitudinal Brain Tumor Volumetry

Meier, Raphael; Knecht, Urspeter; Loosli, Tina; Bauer, Stefan; Slotboom, Johannes; Wiest, Roland; Reyes, Mauricio

doi:10.1038/srep23376

Cited by 96 publications

(99 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Very few completely automatic segmentation algorithms have been adopted in the clinic. Recently, only one automated tool has been clinically evaluated [33]. …”

Section: Introductionmentioning

confidence: 99%

Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI

et al. 2016

View full text Add to dashboard Cite

PurposeWe propose a fully automated method for detection and segmentation of the abnormal tissue associated with brain tumour (tumour core and oedema) from Fluid- Attenuated Inversion Recovery (FLAIR) Magnetic Resonance Imaging (MRI).MethodsThe method is based on superpixel technique and classification of each superpixel. A number of novel image features including intensity-based, Gabor textons, fractal analysis and curvatures are calculated from each superpixel within the entire brain area in FLAIR MRI to ensure a robust classification. Extremely randomized trees (ERT) classifier is compared with support vector machine (SVM) to classify each superpixel into tumour and non-tumour.ResultsThe proposed method is evaluated on two datasets: (1) Our own clinical dataset: 19 MRI FLAIR images of patients with gliomas of grade II to IV, and (2) BRATS 2012 dataset: 30 FLAIR images with 10 low-grade and 20 high-grade gliomas. The experimental results demonstrate the high detection and segmentation performance of the proposed method using ERT classifier. For our own cohort, the average detection sensitivity, balanced error rate and the Dice overlap measure for the segmented tumour against the ground truth are 89.48 %, 6 % and 0.91, respectively, while, for the BRATS dataset, the corresponding evaluation results are 88.09 %, 6 % and 0.88, respectively.ConclusionsThis provides a close match to expert delineation across all grades of glioma, leading to a faster and more reproducible method of brain tumour detection and delineation to aid patient management.

show abstract

“…Very few completely automatic segmentation algorithms have been adopted in the clinic. Recently, only one automated tool has been clinically evaluated [33]. …”

Section: Introductionmentioning

confidence: 99%

Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Segmentation of brain volumes into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) tissue types relying on MRI data has a wide clinical relevance, including volumetric analysis in neurodegenerative disorders, as well as reliable quantification of lesion morphometry and classification in longitudinal examinations …”

Section: Introductionmentioning

confidence: 99%

Automated pixel‐wise brain tissue segmentation of diffusion‐weighted images via machine learning

2018

View full text Add to dashboard Cite

The diffusion-weighted (DW) MR signal sampled over a wide range of b-values potentially allows for tissue differentiation in terms of cellularity, microstructure, perfusion, and T relaxivity. This study aimed to implement a machine learning algorithm for automatic brain tissue segmentation from DW-MRI datasets, and to determine the optimal sub-set of features for accurate segmentation. DWI was performed at 3 T in eight healthy volunteers using 15 b-values and 20 diffusion-encoding directions. The pixel-wise signal attenuation, as well as the trace and fractional anisotropy (FA) of the diffusion tensor, were used as features to train a support vector machine classifier for gray matter, white matter, and cerebrospinal fluid classes. The datasets of two volunteers were used for validation. For each subject, tissue classification was also performed on 3D T -weighted data sets with a probabilistic framework. Confusion matrices were generated for quantitative assessment of image classification accuracy in comparison with the reference method. DWI-based tissue segmentation resulted in an accuracy of 82.1% on the validation dataset and of 82.2% on the training dataset, excluding relevant model over-fitting. A mean Dice coefficient (DSC) of 0.79 ± 0.08 was found. About 50% of the classification performance was attributable to five features (i.e. signal measured at b-values of 5/10/500/1200 s/mm and the FA). This reduced set of features led to almost identical performances for the validation (82.2%) and the training (81.4%) datasets (DSC = 0.79 ± 0.08). Machine learning techniques applied to DWI data allow for accurate brain tissue segmentation based on both morphological and functional information.

show abstract

“…The OsiriX4647 and 3D slicer4849 methods have been used to estimate the tumor volume. Other software also demonstrated high reliability5051525354.…”

Section: Discussionmentioning

confidence: 99%

Different Volumetric Measurement Methods for Pituitary Adenomas and Their Crucial Clinical Significance

Chuang

Lin

Pai

et al. 2017

Sci Rep

View full text Add to dashboard Cite

Confirming the status of residual tumors is crucial. In stationary or spontaneous regression cases, early treatments are inappropriate. The long-used geometric calculation formula is 1/2 (length × width × height). However, it yields only rough estimates and is particularly unreliable for irregularly shaped masses. In our study, we attempted to propose a more accurate method. Between 2004 and 2014, 94 patients with pituitary tumors were enrolled in this retrospective study. All patients underwent transsphenoidal surgery and received magnetic resonance imaging (MRI). The pre- and postoperative volumes calculated using the traditional formula were termed A1 and A2, and those calculated using the proposed method were termed O1 and O2, respectively. Wilcoxon signed rank test revealed no significant difference between the A1 and O1 groups (P = 0.1810) but a significant difference between the A2 and O2 groups (P < 0.0001). Significant differences were present in the extent of resection (P < 0.0001), high-grade cavernous sinus invasion (P = 0.0312), and irregular shape (P = 0.0116). Volume is crucial in evaluating tumor status and determining treatment. Therefore, a more scientific method is especially useful when lesions are irregularly shaped or when treatment is determined exclusively based on the tumor volume.

show abstract

Clinical Evaluation of a Fully-automatic Segmentation Method for Longitudinal Brain Tumor Volumetry

Cited by 96 publications

References 44 publications

Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI

Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI

Automated pixel‐wise brain tissue segmentation of diffusion‐weighted images via machine learning

Different Volumetric Measurement Methods for Pituitary Adenomas and Their Crucial Clinical Significance

Contact Info

Product

Resources

About