The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

Menze, Bjoern; Jakab, András; Bauer, Stefan; Kalpathy–Cramer, Jayashree; Farahani, Keyvan; Kirby, Justin; Burren, Yuliya; Porz, Nicole; Slotboom, Johannes; Wiest, Roland; Lánczi, Levente; Gerstner, Elizabeth R.; Weber, Marc‐André; Arbel, Tal; Avants, Brian B.; Ayache, Nicholas; Buendia, Patricia; Collins, D. Louis; Cordier, Nicolas; Corso, Jason J.; Criminisi, Antonio; Das, Tilak; Delingette, Hervé; Demiralp, Çağatay; Durst, Christopher R.; Dojat, Michel; Doyle, Senan; Festa, Joana; Forbes, Florence; Geremia, Ezequiel; Glocker, Ben; Golland, Polina; Guo, X. Edward; Hamamcı, Andaç; Iftekharuddin, Khan M.; Jena, R.; John, Nigel; Konukoğlu, Ender; Lashkari, Danial; Mariz, José; Meier, Raphael; Pereira, Sérgio; Precup, Doina; Price, Stephen J.; Raviv, Tammy Riklin; Reza, Syed M. S.; Ryan, Michael; Sarıkaya, Duygu; Schwartz, Lawrence H.; Shin, Hoo-Chang; Shotton, Jamie; Silva, Carlos A.; Sousa, Nuno; Subbanna, Nagesh K.; Székely, Gábor; Taylor, Thomas J.; Thomas, Owain; Tustison, Nicholas J.; Ünal, Gözde; Vasseur, Flor; Wintermark, Max; Ye, Dong Hye; Zhao, Liang; Zhao, Binsheng; Zikic, Darko; Prastawa, Marcel; Reyes, Mauricio; Leemput, Koen Van

doi:10.1109/tmi.2014.2377694

Cited by 4,064 publications

(2,672 citation statements)

References 116 publications

Supporting

Mentioning

2,630

Contrasting

Unclassified

Order By: Relevance

“…Specifically, these data are a combination of the training set (10 LGGs and 20 HGGs) used in the BRATS 2013 challenge [17], as well as 44 LGG and 112 HGG scans provided from the National Institutes of Health (NIH) Cancer Imaging Archive (TCIA). The data of each patient consists of native and contrast-enhanced (CE) T1-weighted, as well as T2-weighted and T2 Fluid-attenuated inversion recovery (FLAIR) MRI volumes.…”

Section: Methodsmentioning

confidence: 99%

GLISTRboost: Combining Multimodal MRI Segmentation, Registration, and Biophysical Tumor Growth Modeling with Gradient Boosting Machines for Glioma Segmentation

Bakas

Zeng

Sotiras

et al. 2016

Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries

View full text Add to dashboard Cite

We present an approach for segmenting low-and high-grade gliomas in multimodal magnetic resonance imaging volumes. The proposed approach is based on a hybrid generativediscriminative model. Firstly, a generative approach based on an Expectation-Maximization framework that incorporates a glioma growth model is used to segment the brain scans into tumor, as well as healthy tissue labels. Secondly, a gradient boosting multi-class classification scheme is used to refine tumor labels based on information from multiple patients. Lastly, a probabilistic Bayesian strategy is employed to further refine and finalize the tumor segmentation based on patient-specific intensity statistics from the multiple modalities. We evaluated our approach in 186 cases during the training phase of the BRAin Tumor Segmentation (BRATS) 2015 challenge and report promising results. During the testing phase, the algorithm was additionally evaluated in 53 unseen cases, achieving the best performance among the competing methods.

show abstract

Section: Methodsmentioning

confidence: 99%

GLISTRboost: Combining Multimodal MRI Segmentation, Registration, and Biophysical Tumor Growth Modeling with Gradient Boosting Machines for Glioma Segmentation

Bakas

Zeng

Sotiras

et al. 2016

Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries

View full text Add to dashboard Cite

show abstract

“…We present the preliminary results of our evaluation of this concept on MR datasets. Brain tumor image data used in this work were obtained from [21]. The challenge database [21] contains fully anonymized images with manually labeled tumor groundtruth.…”

Section: Brain Tumor Quantification In Mrmentioning

confidence: 99%

“…Brain tumor image data used in this work were obtained from [21]. The challenge database [21] contains fully anonymized images with manually labeled tumor groundtruth. This dataset consists volumes of size 256 × 256 × 176 .…”

Section: Brain Tumor Quantification In Mrmentioning

confidence: 99%

Fast 3D Salient Region Detection in Medical Images Using GPUs

Thota

Vaswani

Kale

et al. 2015

Machine Intelligence and Signal Processing

View full text Add to dashboard Cite

Automated detection of visually salient regions is an active area of research in computer vision. Salient regions can serve as inputs for object detectors as well as inputs for region-based registration algorithms. In this paper we consider the problem of speeding up computationally intensive bottom-up salient region detection in 3D medical volumes. The method uses the Kadir-Brady formulation of saliency. We show that in the vicinity of a salient region, entropy is a monotonically increasing function of the degree of overlap of a candidate window with the salient region. This allows us to initialize a sparse seed-point grid as the set of tentative salient region centers and iteratively converge to the local entropy maxima, thereby reducing the computation complexity compared to the Kadir Brady approach of performing this computation at every point in the image. We propose two different approaches for achieving this. The first approach involves evaluating entropy in the four quadrants around the seed point and iteratively moving in the direction that increases entropy. The second approach we propose makes use of mean shift tracking framework to affect entropy maximizing moves. Specifically, we propose the use of uniform pmf as the target distribution to seek high entropy regions. We demonstrate the use of our algorithm on medical volumes for left ventricle detection in PET images and tumor localization in brain MR sequences.

show abstract

“…Magnetic Resonance Imaging (MRI) is a preferred type of imaging modality as it maps the tumor changes and clearly indicates the tumor region [1,2]. The process of brain tumor segmentation is a challenging step as it has weak boundaries and inhomogeneous intensities [3].…”

Section: Introductionmentioning

confidence: 99%

Evaluating the Efficiency of different Feature Sets on Brain Tumor Classification in MR Images

Eltayeb¹,

Salem²,

Al-Atabany³

2018

IJCA

View full text Add to dashboard Cite

In this paper, a study for evaluating the efficacy of different feature sets that used brain tumor classification is presented. Different features sets are extracted as shape, 1 st order texture features (FOS), 2 nd order (GLCM, GLRLM), boundary features, and wavelet-based features. The brain tumors are extracted using the k-means clustering algorithm. Then different classifiers such as Artificial Neural Network (ANN), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) were used in the classification process.A set of 65 real and simulated (Flair modality) MRI images from multimodal brain tumor image segmentation benchmark (BRATS) organized by MICCAI 2012 challenge is used for performance evaluation. The overall segmentation results for the 65 volumes are 90.15±0.12. For the Feature sets efficacy step, the highest accuracy of 94.74% is achieved by the SVM when using the wavelet-based features. The lowest accuracy achieved by the three classifiers obtained when using the second order texture features..

show abstract

The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

Cited by 4,064 publications

References 116 publications

GLISTRboost: Combining Multimodal MRI Segmentation, Registration, and Biophysical Tumor Growth Modeling with Gradient Boosting Machines for Glioma Segmentation

GLISTRboost: Combining Multimodal MRI Segmentation, Registration, and Biophysical Tumor Growth Modeling with Gradient Boosting Machines for Glioma Segmentation

Fast 3D Salient Region Detection in Medical Images Using GPUs

Evaluating the Efficiency of different Feature Sets on Brain Tumor Classification in MR Images

Contact Info

Product

Resources

About