BraTS Challenge Research Articles

Texture based localization of a brain tumor from MR-images by using a machine learning approach

In this paper, a machine learning approach was used for brain tumour localization on FLAIR scans of magnetic resonance images (MRI). The multi-modal brain images dataset (BraTs 2012) was used, that is a skull stripped and co-registered. In order to remove the noise, bilateral filtering is applied and then texton-map images are created by using the Gabor filter bank. From the texton-map, the image is segmented out into superpixel and then the low-level features are extracted: the first order intensity statistical features and also calculates the histogram level of texton-map at each superpixel level. There is a significant contribution here that the low features are not too much significant for the localization of brain tumour from MR images, but we have to make them meaningful by integrating features with the texton-map images at the region level approach. Then these features which are provided later to classifier for the prediction of three classes: background, tumour and non-tumour region, and used the labels for computation of two different areas (i.e. complete tumour and non-tumour). A Leave-one-out (LOOCV) cross validation technique is applied and achieves the dice overlap score of 88% for the whole tumour area localization, which is similar to the declared score in MICCAI BraTS challenge. This brain tumour localization approach using the textonmap image based on superpixel features illustrates the equivalent performance with other contemporary techniques. Recently, medical hypothesis generation by using autonomous computer based systems in disease diagnosis have given the great contribution in medical diagnosis.

An Efficient Hybrid Fuzzy-Clustering Driven 3D-Modeling of Magnetic Resonance Imagery for Enhanced Brain Tumor Diagnosis

Brain tumor detection and its analysis are essential in medical diagnosis. The proposed work focuses on segmenting abnormality of axial brain MR DICOM slices, as this format holds the advantage of conserving extensive metadata. The axial slices presume the left and right part of the brain is symmetric by a Line of Symmetry (LOS). A semi-automated system is designed to mine normal and abnormal structures from each brain MR slice in a DICOM study. In this work, Fuzzy clustering (FC) is applied to the DICOM slices to extract various clusters for different k. Then, the best-segmented image that has high inter-class rigidity is obtained using the silhouette fitness function. The clustered boundaries of the tissue classes further enhanced by morphological operations. The FC technique is hybridized with the standard image post-processing techniques such as marker controlled watershed segmentation (MCW), region growing (RG), and distance regularized level sets (DRLS). This procedure is implemented on renowned BRATS challenge dataset of different modalities and a clinical dataset containing axial T2 weighted MR images of a patient. The sequential analysis of the slices is performed using the metadata information present in the DICOM header. The validation of the segmentation procedures against the ground truth images authorizes that the segmented objects of DRLS through FC enhanced brain images attain maximum scores of Jaccard and Dice similarity coefficients. The average Jaccard and dice scores for segmenting tumor part for ten patient studies of the BRATS dataset are 0.79 and 0.88, also for the clinical study 0.78 and 0.86, respectively. Finally, 3D visualization and tumor volume estimation are done using accessible DICOM information.

Robustness of Radiomics for Survival Prediction of Brain Tumor Patients Depending on Resection Status.

Prediction of overall survival based on multimodal MRI of brain tumor patients is a difficult problem. Although survival also depends on factors that cannot be assessed via preoperative MRI such as surgical outcome, encouraging results for MRI-based survival analysis have been published for different datasets. We assess if and how established radiomic approaches as well as novel methods can predict overall survival of brain tumor patients on the BraTS challenge dataset. This dataset consists of multimodal preoperative images of 211 glioblastoma patients from several institutions with reported resection status and known survival. In the official challenge setting, only patients with a reported gross total resection (GTR) are taken into account. We therefore evaluated previously published methods as well as different machine learning approaches on the BraTS dataset. For different types of resection status, these approaches are compared to a baseline, a linear regression on patient age only. This naive approach won the 3rd place out of 26 participants in the BraTS survival prediction challenge 2018. Previously published radiomic signatures show significant correlations and predictiveness to patient survival for patients with a reported subtotal resection. However, for patients with reported GTR, none of the evaluated approaches was able to outperform the age-only baseline in a cross-validation setting, explaining the poor performance of approaches based on radiomics in the BraTS challenge 2018.

Automated Brain Tumor Segmentation Using Multimodal Brain Scans: A Survey Based on Models Submitted to the BraTS 2012-2018 Challenges.

Reliable brain tumor segmentation is essential for accurate diagnosis and treatment planning. Since manual segmentation of brain tumors is a highly time-consuming, expensive and subjective task, practical automated methods for this purpose are greatly appreciated. But since brain tumors are highly heterogeneous in terms of location, shape, and size, developing automatic segmentation methods has remained a challenging task over decades. This paper aims to review the evolution of automated models for brain tumor segmentation using multimodal MR images. In order to be able to make a just comparison between different methods, the proposed models are studied for the most famous benchmark for brain tumor segmentation, namely the BraTS challenge [1]. The BraTS 2012-2018 challenges and the state-of-the-art automated models employed each year are analysed. The changing trend of these automated methods since 2012 are studied and the main parameters that affect the performance of different models are analysed.

Automatic Brain Tumor Segmentation Based on Cascaded Convolutional Neural Networks With Uncertainty Estimation.

Automatic segmentation of brain tumors from medical images is important for clinical assessment and treatment planning of brain tumors. Recent years have seen an increasing use of convolutional neural networks (CNNs) for this task, but most of them use either 2D networks with relatively low memory requirement while ignoring 3D context, or 3D networks exploiting 3D features while with large memory consumption. In addition, existing methods rarely provide uncertainty information associated with the segmentation result. We propose a cascade of CNNs to segment brain tumors with hierarchical subregions from multi-modal Magnetic Resonance images (MRI), and introduce a 2.5D network that is a trade-off between memory consumption, model complexity and receptive field. In addition, we employ test-time augmentation to achieve improved segmentation accuracy, which also provides voxel-wise and structure-wise uncertainty information of the segmentation result. Experiments with BraTS 2017 dataset showed that our cascaded framework with 2.5D CNNs was one of the top performing methods (second-rank) for the BraTS challenge. We also validated our method with BraTS 2018 dataset and found that test-time augmentation improves brain tumor segmentation accuracy and that the resulting uncertainty information can indicate potential mis-segmentations and help to improve segmentation accuracy.

Learning Cross-Modality Representations From Multi-Modal Images.

Machine learning algorithms can have difficulties adapting to data from different sources, for example from different imaging modalities. We present and analyze three techniques for unsupervised cross-modality feature learning, using a shared autoencoder-like convolutional network that learns a common representation from multi-modal data. We investigate a form of feature normalization, a learning objective that minimizes cross-modality differences, and modality dropout, in which the network is trained with varying subsets of modalities. We measure the same-modality and cross-modality classification accuracies and explore whether the models learn modality-specific or shared features. This paper presents experiments on two public data sets, with knee images from two MRI modalities, provided by the Osteoarthritis Initiative, and brain tumor segmentation on four MRI modalities from the BRATS challenge. All three approaches improved the cross-modality classification accuracy, with modality dropout and per-feature normalization giving the largest improvement. We observed that the networks tend to learn a combination of cross-modality and modality-specific features. Overall, a combination of all three methods produced the most cross-modality features and the highest cross-modality classification accuracy, while maintaining most of the same-modality accuracy.

User-Guided Segmentation of Multi-modality Medical Imaging Datasets with ITK-SNAP.

ITK-SNAP is an interactive software tool for manual and semi-automatic segmentation of 3D medical images. This paper summarizes major new features added to ITK-SNAP over the last decade. The main focus of the paper is on new features that support semi-automatic segmentation of multi-modality imaging datasets, such as MRI scans acquired using different contrast mechanisms (e.g., T1, T2, FLAIR). The new functionality uses decision forest classifiers trained interactively by the user to transform multiple input image volumes into a foreground/background probability map; this map is then input as the data term to the active contour evolution algorithm, which yields regularized surface representations of the segmented objects of interest. The new functionality is evaluated in the context of high-grade and low-grade glioma segmentation by three expert neuroradiogists and a non-expert on a reference dataset from the MICCAI 2013 Multi-Modal Brain Tumor Segmentation Challenge (BRATS). The accuracy of semi-automatic segmentation is competitive with the top specialized brain tumor segmentation methods evaluated in the BRATS challenge, with most results obtained in ITK-SNAP being more accurate, relative to the BRATS reference manual segmentation, than the second-best performer in the BRATS challenge; and all results being more accurate than the fourth-best performer. Segmentation time is reduced over manual segmentation by 2.5 and 5 times, depending on the rater. Additional experiments in interactive placenta segmentation in 3D fetal ultrasound illustrate the generalizability of the new functionality to a different problem domain.

A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery.

Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients' data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.

An automatic glioma grading method based on multi-feature extraction and fusion.

An accurate assessment of tumor malignancy grade in the preoperative situation is important for clinical management. However, the manual grading of gliomas from MRIs is both a tiresome and time consuming task for radiologists. Thus, it is a priority to design an automatic and effective computer-aided diagnosis (CAD) tool to assist radiologists in grading gliomas. To design an automatic computer-aided diagnosis for grading gliomas using multi-sequence magnetic resonance imaging. The proposed method consists of two steps: (1) the features of high and low grade gliomas are extracted from multi-sequence magnetic resonance images, and (2) then, a KNN classifier is trained to grade the gliomas. In the feature extraction step, the intensity, volume, and local binary patterns (LBP) of the gliomas are extracted, and PCA is used to reduce the data dimension. The proposed "Intensity-Volume-LBP-PCA-KNN" method is validated on the MICCAI 2015 BraTS challenge dataset, and an average grade accuracy of 87.59% is obtained. The proposed method is an effective method for automatically grading gliomas and can be applied to real situations.

Brain tumor classification from multi-modality MRI using wavelets and machine learning

In this paper, we propose a brain tumor segmentation and classification method for multi-modality magnetic resonance imaging scans. The data from multi-modal brain tumor segmentation challenge (MICCAI BraTS 2013) are utilized which are co-registered and skull-stripped, and the histogram matching is performed with a reference volume of high contrast. From the preprocessed images, the following features are then extracted: intensity, intensity differences, local neighborhood and wavelet texture. The integrated features are subsequently provided to the random forest classifier to predict five classes: background, necrosis, edema, enhancing tumor and non-enhancing tumor, and then these class labels are used to hierarchically compute three different regions (complete tumor, active tumor and enhancing tumor). We performed a leave-one-out cross-validation and achieved 88% Dice overlap for the complete tumor region, 75% for the core tumor region and 95% for enhancing tumor region, which is higher than the Dice overlap reported from MICCAI BraTS challenge.

Brain tumor detection and segmentation in a CRF (conditional random fields) framework with pixel-pairwise affinity and superpixel-level features

Detection and segmentation of a brain tumor such as glioblastoma multiforme (GBM) in magnetic resonance (MR) images are often challenging due to its intrinsically heterogeneous signal characteristics. A robust segmentation method for brain tumor MRI scans was developed and tested. Simple thresholds and statistical methods are unable to adequately segment the various elements of the GBM, such as local contrast enhancement, necrosis, and edema. Most voxel-based methods cannot achieve satisfactory results in larger data sets, and the methods based on generative or discriminative models have intrinsic limitations during application, such as small sample set learning and transfer. A new method was developed to overcome these challenges. Multimodal MR images are segmented into superpixels using algorithms to alleviate the sampling issue and to improve the sample representativeness. Next, features were extracted from the superpixels using multi-level Gabor wavelet filters. Based on the features, a support vector machine (SVM) model and an affinity metric model for tumors were trained to overcome the limitations of previous generative models. Based on the output of the SVM and spatial affinity models, conditional random fields theory was applied to segment the tumor in a maximum a posteriori fashion given the smoothness prior defined by our affinity model. Finally, labeling noise was removed using "structural knowledge" such as the symmetrical and continuous characteristics of the tumor in spatial domain. The system was evaluated with 20GBM cases and the BraTS challenge data set. Dice coefficients were computed, and the results were highly consistent with those reported by Zikic et al. (MICCAI 2012, Lecture notes in computer science. vol 7512, pp 369-376, 2012). A brain tumor segmentation method using model-aware affinity demonstrates comparable performance with other state-of-the art algorithms.