A Supervised Patch-Based Approach for Human Brain Labeling

Rousseau, François; Habas, Piotr A.; Studholme, Colin

doi:10.1109/tmi.2011.2156806

Cited by 251 publications

(327 citation statements)

References 37 publications

Supporting

Mentioning

312

Contrasting

Unclassified

Order By: Relevance

“…[Rohlfing et al, 2004;Warfield et al, 2004;Heckemann et al, 2006]. The second group of patch-based methods has recently enjoyed increased attention [Coupé et al, 2011;Rousseau et al, 2011;Wu et al, 2012]. Here, the label proposal is estimated for each point in the target image by a local similarity-based search in the atlas.…”

Section: Introductionmentioning

confidence: 99%

Encoding atlases by randomized classification forests for efficient multi-atlas label propagation

Zikic

Glocker

Criminisi

2014

Medical Image Analysis

105

View full text Add to dashboard Cite

We propose a method for multi-atlas label propagation (MALP) based on encoding the individual atlases by randomized classification forests. Most current approaches perform a non-linear registration between all atlases and the target image, followed by a sophisticated fusion scheme. While these approaches can achieve high accuracy, in general they do so at high computational cost. This might negatively affect the scalability to large databases and experimentation. To tackle this issue, we propose to use a small and deep classification forest to encode each atlas individually in reference to an aligned probabilistic atlas, resulting in an Atlas Forest (AF). Our classifier-based encoding differs from current MALP approaches, which represent each point in the atlas either directly as a single image/label value pair, or by a set of corresponding patches. At test time, each AF produces one probabilistic label estimate, and their fusion is done by averaging. Our scheme performs only one registration per target image, achieves good results with a simple fusion scheme, and allows for efficient experimentation. In contrast to standard forest schemes, in which each tree would be trained on all atlases, our approach retains the advantages of the standard MALP framework. The target-specific selection of atlases remains possible, and incorporation of new scans is straightforward without retraining. The evaluation on four different databases shows accuracy within the range of the state of the art at a significantly lower running time.

show abstract

Section: Introductionmentioning

confidence: 99%

Encoding atlases by randomized classification forests for efficient multi-atlas label propagation

Zikic

Glocker

Criminisi

2014

Medical Image Analysis

105

View full text Add to dashboard Cite

show abstract

“…Recently, Rousseau et al [26] applied the block-wise strategy proposed by Coupe et al [33] to nonlocal means label fusion resulting in slightly better segmentations than those obtained by a voxel-based implementation. In fact, the block-based approach not only provides a reduction of the computational load by processing only every other in all three dimensions as done by Coupe et al [33] but it also introduces an implicit regularization in the labels due to the overlap between blocks which increases the number of patches involved in the voting process for each voxel.…”

Section: Block-wise Nonlocal Means Estimator Multi-label Segmentationmentioning

confidence: 99%

“…As originally proposed by Coupé et al [23], our method is based on a nonlocal means label fusion where labels from multiple templates are weighted according to the Euclidean distance between patch intensities. Using this approach we avoid the one-to-one matching assumption of nonlinear registration label fusion methods by enabling a one-to-many matching, therefore reducing the segmentation errors [23,25,26] by better managing the inherent inter-subject variability of human brain anatomy.…”

Section: Introductionmentioning

confidence: 99%

NABS: non-local automatic brain hemisphere segmentation

Romero

Manjón

Tohka

et al. 2015

Magnetic Resonance Imaging

View full text Add to dashboard Cite

In this paper, we propose an automatic method to segment the five main brain sub-regions (i.e. left/right hemispheres, left/right cerebellum and brainstem) from magnetic resonance images.The proposed method uses a library of pre-labeled brain images in a stereotactic space in combination with a non local label fusion scheme for segmentation. The main novelty of the proposed method is the use of a multi-label block-wise label fusion strategy specifically designed to deal with the classification of main brain sub-volumes that process only specific parts of the brain images significantly reducing the computational burden. The proposed method has been quantitatively evaluated against manual segmentations. The evaluation showed that the proposed method was faster while producing more accurate segmentations than a current state-of-the-art method. We also present evidences suggesting that the proposed method was more robust against brain pathologies than the compared method. Finally, we demonstrate the clinical value of our method compared to the state-of-the-art approach in terms of the asymmetry quantification in Alzheimer's disease.

show abstract

“…In a recent work, Buades et al [5] shows that non-local means filtering gives state-of-the-art performance in structurepreserving image denoising. The strategy has also been applied to brain image labeling [6], image registration [8], and MR image super-resolution [9]. We employ nonlocal averaging for combining all matching patches that have been determined based on the distance measure as described in Section 2.2.…”

Section: Non-local Approachmentioning

confidence: 99%

“…The core of our proposed method is fuzzy patch matching, which is inspired by recent advances in non-local image processing, e.g., for denoising [5], labeling [6], segmentation [7]. First, a new patch distance measure is proposed for determining matching patches.…”

Section: Introductionmentioning

confidence: 99%

Non-local Means Resolution Enhancement of Lung 4D-CT Data

Zhang

Yap

et al. 2012

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2012

View full text Add to dashboard Cite

Abstract. Image resolution in 4D-CT is a crucial bottleneck that needs to be overcome for improved dose planning in radiotherapy for lung cancer. In this paper, we propose a novel patch-based algorithm to enhance the image quality of 4D-CT data. Our premise is that anatomical information missing in one phase can be recovered from complementary information embedded in other phases. We employ a patch-based mechanism to propagate information across phases for reconstruction of intermediate slices in the axial direction, where resolution is normally the lowest. Specifically, structurally-matching and spatially-nearby patches are combined for reconstruction of each patch. For greater sensitivity to anatomical nuances, we further employ a quad-tree technique to adaptively partition each slice of the image in each phase for more fine-grained refinement. Our evaluation based on a public 4D-CT lung data indicates that our algorithm gives very promising results with significantly enhanced image structures. Introduction4D-CT is becoming increasingly popular in lung cancer treatment for providing respiratory-related information that is essential for guiding radiation therapy effectively. However, due to the risk of radiation [1], only a limited number of CT segments are usually acquired, which often results in very low resolution along the inferior-superior direction. This low-resolution (LR) data are usually plagued with visible imaging artifacts such as vessel discontinuity and partial volume effect. More importantly, insufficient resolution further distorts the shape of a tumor. This distortion might finally interfere with optimal dose planning. Super-resolution (SR) reconstruction is an effective approach for improving image resolution. Classical SR methods can be divided into two major categories: interpolation-based and model-based. The main advantage of interpolation-based methods is their simplicity. However, blurred edges and undesirable artifacts are inevitable. Currently, more attention has been directed to model-based SR approach. The general assumption of model-based SR approaches is that the LR image is a degraded version

show abstract

A Supervised Patch-Based Approach for Human Brain Labeling

Cited by 251 publications

References 37 publications

Encoding atlases by randomized classification forests for efficient multi-atlas label propagation

Encoding atlases by randomized classification forests for efficient multi-atlas label propagation

NABS: non-local automatic brain hemisphere segmentation

Non-local Means Resolution Enhancement of Lung 4D-CT Data

Contact Info

Product

Resources

About