Decision Forests for Tissue-Specific Segmentation of High-Grade Gliomas in Multi-channel MR

Zikic, Darko; Glocker, Ben; Konukoğlu, Ender; Criminisi, Antonio; Demiralp, Çağatay; Shotton, Jamie; Thomas, Owen; Das, Tilak; Jena, R.; Price, Stephen J.

doi:10.1007/978-3-642-33454-2_46

Cited by 206 publications

(165 citation statements)

References 18 publications

Supporting

Mentioning

157

Contrasting

Order By: Relevance

“…Criminisi et al proposed a random forest-based method for efficient detection and localization of anatomical structures within CT volumes (Criminisi et al, 2009). Zikic et al proposed a novel method based on the random forests for automatic segmentation of high-grade gliomas and their subregions from multi-channel MR images (Zikic et al, 2012). Similar work was also presented in (Zikic et al, 2013a;Zikic et al, 2013bZikic et al, , 2014, in which an atlas forest was introduced and iteratively employed in an auto-context scheme for efficient adult brain labeling.…”

Section: Introductionmentioning

confidence: 94%

“…To address these limitations, inspired by the pioneering work Zikic et al, 2013a;Zikic et al, 2014;Zikic et al, 2012), we propose a novel Learning-based multi-source IntegratioN frameworK for Segmentation of infant brain images (LINKS). The proposed framework is able to integrate information from multisource images together for efficient tissue segmentation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

LINKS: Learning-Based Multi-source IntegratioN FrameworK for Segmentation of Infant Brain Images

Wang

Gao

Shi

et al. 2014

Medical Computer Vision: Algorithms for Big Data

View full text Add to dashboard Cite

Segmentation of infant brain MR images is challenging due to insufficient image quality, severe partial volume effect, and ongoing maturation and myelination processes. In the first year of life, the image contrast between white and gray matters of the infant brain undergoes dramatic changes. In particular, the image contrast is inverted around 6-8 months of age, and the white and gray matter tissues are isointense in both T1-and T2-weighted MR images and thus exhibit the extremely low tissue contrast, which poses significant challenges for automated segmentation. Most previous studies used multi-atlas label fusion strategy, which has the limitation of equally treating the different available image modalities and is often computationally expensive. To cope with these limitations, in this paper, we propose a novel learning-based multi-source integration framework for segmentation of infant brain images. Specifically, we employ the random forest technique to effectively integrate features from multi-source images together for tissue segmentation. Here, the multi-source images include initially only the multi-modality (T1, T2 and FA) images and later also the iteratively estimated and refined tissue probability maps of gray matter, white matter, and cerebrospinal fluid. Experimental results on 119 infants show that the proposed method achieves better performance than other state-of-the-art automated segmentation methods. Further validation was performed on the MICCAI grand challenge and the proposed method was ranked top among all competing methods. Moreover, to alleviate the possible Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. anatomical errors, our method can also be combined with an anatomically-constrained multi-atlas labeling approach for further improving the segmentation accuracy. NIH Public Access

show abstract

Section: Introductionmentioning

confidence: 94%

Section: Introductionmentioning

confidence: 99%

LINKS: Learning-Based Multi-source IntegratioN FrameworK for Segmentation of Infant Brain Images

Wang

Gao

Shi

et al. 2014

Medical Computer Vision: Algorithms for Big Data

View full text Add to dashboard Cite

show abstract

“…A visual feature at scale k is a map defined by θ k : P k I → R. Arbitrary large amounts of task-specific features can be derived in a straightforward way from the multi-scale representation of the image [7,8,6].…”

Section: Visual Featuresmentioning

confidence: 99%

“…H(T 1 p ) = 0. Unlike previous work [6,7,8], we introduce spatial refinement in the random forest framework to capture fine structures. Indeed, when the supervoxels are too large to properly describe annotated image regions, the scale k is decremented.…”

Section: Trainingmentioning

confidence: 99%

“…As a result, the classifiers miss critical coarse-scale cues which penalizes the performance of the final segmentation. Other supervised approaches build on context-rich random forests for segmentation of brain lesions in MRIs [7,8]. Still, their excellent performance requires a careful tuning of class weights during training, and is highly sensitive to the spatial sub-sampling of the training data for the different classes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spatially Adaptive Random Forests

Geremia

Menze

Ayache

2013

2013 IEEE 10th International Symposium on Biomedical Imaging

View full text Add to dashboard Cite

Medical imaging protocols produce large amounts of multimodal volumetric images. The large size of the datasets contributes to the success of supervised discriminative methods for semantic image segmentation. Classifying relevant structures in medical images is challenging due to (a) the large size of data volumes, and (b) the severe class overlap in the feature space. Subsampling the training data addresses the first issue at the cost of discarding potentially useful image information. Increasing feature dimensionality addresses the second but requires dense sampling. We propose a general and efficient solution to these problems. "Spatially Adaptive Random Forests" (SARF) is a supervised learning algorithm. SARF aims at automatic semantic labelling of large medical volumes. During training, it learns the optimal image sampling associated to the classification task. During testing, the algorithm quickly handles the background and focuses challenging image regions to refine the classification. SARF demonstrated top performance in the context of multi-class gliomas segmentation in multi-modal MR images.

show abstract

Accurate and robust segmentation of neuroanatomy in T1‐weighted MRI by combining spatial priors with deep convolutional neural networks

Novosad

Fonov

Collins

et al. 2019

Human Brain Mapping

View full text Add to dashboard Cite

Neuroanatomical segmentation in magnetic resonance imaging (MRI) of the brain is a prerequisite for quantitative volume, thickness, and shape measurements, as well as an important intermediate step in many preprocessing pipelines. This work introduces a new highly accurate and versatile method based on 3D convolutional neural networks for the automatic segmentation of neuroanatomy in T1-weighted MRI. In combination with a deep 3D fully convolutional architecture, efficient linear registration-derived spatial priors are used to incorporate additional spatial context into the network. An aggressive data augmentation scheme using random elastic deformations is also used to regularize the networks, allowing for excellent performance even in cases where only limited labeled training data are available. Applied to hippocampus segmentation in an elderly population (mean Dice coefficient = 92.1%) and subcortical segmentation in a healthy adult population (mean Dice coefficient = 89.5%), we demonstrate new state-ofthe-art accuracies and a high robustness to outliers. Further validation on a multistructure segmentation task in a scan-rescan dataset demonstrates accuracy (mean Dice coefficient = 86.6%) similar to the scan-rescan reliability of expert manual segmentations (mean Dice coefficient = 86.9%), and improved reliability compared to both expert manual segmentations and automated segmentations using FIRST. Furthermore, our method maintains a highly competitive runtime performance (e.g., requiring only 10 s for left/ right hippocampal segmentation in 1 × 1 × 1 mm 3 MNI stereotaxic space), orders of magnitude faster than conventional multiatlas segmentation methods. K E Y W O R D Sdeep learning, magnetic resonance imaging, neural networks, neuroanatomy, segmentation, spatial priors † Data used in preparation of this article were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at

show abstract

Decision Forests for Tissue-Specific Segmentation of High-Grade Gliomas in Multi-channel MR

Cited by 206 publications

References 18 publications

LINKS: Learning-Based Multi-source IntegratioN FrameworK for Segmentation of Infant Brain Images

LINKS: Learning-Based Multi-source IntegratioN FrameworK for Segmentation of Infant Brain Images

Spatially Adaptive Random Forests

Accurate and robust segmentation of neuroanatomy in T1‐weighted MRI by combining spatial priors with deep convolutional neural networks

Contact Info

Product

Resources

About