Robust Multimodal Dictionary Learning

Cao, Tian; Jojic, Vladimir; Powell, Debbie; Czymmek, Kirk J.; Niethammer, Marc

doi:10.1007/978-3-642-40811-3_33

Cited by 13 publications

(14 citation statements)

References 10 publications

(16 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some other approaches have used atlas templates for single-label [43] or for multi-label brain tumor segmentation [10,18,19]. All these above methods need a multi-channel input (multi-modality MRI data).…”

Section: Figurementioning

confidence: 99%

“…All these above methods need a multi-channel input (multi-modality MRI data). In recent years, some approaches used image patch dictionary learning for single-label tumor segmentation [39,44] or multi-modal coupled dictionary learning for microscopical image registration [10]. These methods used one dictionary for each class and the residual error to discriminate the tumor/non-tumor classes.…”

Section: Figurementioning

confidence: 99%

See 1 more Smart Citation

Brain tumor classification and segmentation using sparse coding and dictionary learning

Al-Shaikhli¹,

Yang²,

Rosenhahn³

2016

Biomedical Engineering / Biomedizinische Technik

View full text Add to dashboard Cite

This paper presents a novel fully automatic framework for multi-class brain tumor classification and segmentation using a sparse coding and dictionary learning method. The proposed framework consists of two steps: classification and segmentation. The classification of the brain tumors is based on brain topology and texture. The segmentation is based on voxel values of the image data. Using K-SVD, two types of dictionaries are learned from the training data and their associated ground truth segmentation: feature dictionary and voxel-wise coupled dictionaries. The feature dictionary consists of global image features (topological and texture features). The coupled dictionaries consist of coupled information: gray scale voxel values of the training image data and their associated label voxel values of the ground truth segmentation of the training data. For quantitative evaluation, the proposed framework is evaluated using different metrics. The segmentation results of the brain tumor segmentation (MICCAI-BraTS-2013) database are evaluated using five different metric scores, which are computed using the online evaluation tool provided by the BraTS-2013 challenge organizers. Experimental results demonstrate that the proposed approach achieves an accurate brain tumor classification and segmentation and outperforms the state-of-the-art methods.

show abstract

Section: Figurementioning

confidence: 99%

Section: Figurementioning

confidence: 99%

Brain tumor classification and segmentation using sparse coding and dictionary learning

Al-Shaikhli¹,

Yang²,

Rosenhahn³

2016

Biomedical Engineering / Biomedizinische Technik

View full text Add to dashboard Cite

show abstract

“…Generally, most of multi-view learning methods belong to the category of the graph-based method. Among them, one representative group of multi-view methods [15,31,32] aim to fuse multiple features into single representation, by exploiting the common latent space shared by all views. For example, multi-view sparse coding [31,32] combines the shared latent representation for the multi-view information by a series of linear maps as dictionaries.…”

Section: A Multi-view Learningmentioning

confidence: 99%

A unified framework based on graph consensus term for multi-view learning

Meng¹,

Feng²,

Guo³

2021

Preprint

View full text Add to dashboard Cite

In recent years, multi-view learning technologies for various applications have attracted a surge of interest. Due to more compatible and complementary information from multiple views, existing multi-view methods could achieve more promising performance than conventional single-view methods in most situations. However, there are still no sufficient researches on the unified framework in existing multi-view works. Meanwhile, how to efficiently integrate multi-view information is still full of challenges. In this paper, we propose a novel multi-view learning framework, which aims to leverage most existing graph embedding works into a unified formula via introducing the graph consensus term. In particular, our method explores the graph structure in each view independently to preserve the diversity property of graph embedding methods. Meanwhile, we choose heterogeneous graphs to construct the graph consensus term to explore the correlations among multiple views jointly. To this end, the diversity and complementary information among different views could be simultaneously considered. Furthermore, the proposed framework is utilized to implement the multiview extension of Locality Linear Embedding, named Multi-view Locality Linear Embedding (MvLLE), which could be efficiently solved by applying the alternating optimization strategy. Empirical validations conducted on six benchmark datasets can show the effectiveness of our proposed method.

show abstract

“…For multi-label brain tumor segmentation, several researchers used either random forest based on feature extraction [8,9,11] or Markov Random Field (MRF) [10,13]. Some other approaches used atlas template for single label [14] or multi-label brain tumor segmentation [4,6,7]. All the above methods require multi-channel input (multi-modality MRI data).…”

Section: Introductionmentioning

confidence: 99%

“…learning for microscopical image registration [7]. These methods used the residual error to discriminate the tumor/non-tumor classes and required one dictionary for each class.…”

Section: Introductionmentioning

confidence: 99%

Coupled Dictionary Learning for Automatic Multi-Label Brain Tumor Segmentation in Flair MRI images

Al-Shaikhli

Yang

Rosenhahn

2014

Advances in Visual Computing

View full text Add to dashboard Cite

Brain tumor segmentation and labeling is a challenging task in medical imaging. In this paper, a novel patch based dictionary learning algorithm for automatic multi-label brain tumor segmentation is proposed. Based on image reconstruction, we present coupled dictionaries, one dictionary of grayscale brain tumor image patches and one dictionary of tumor labels, which can then be used for automatic multi-label brain tumor segmentation of a test image data. The dictionaries are learned from training images of BraTS-MICCAI and the SPL/NSG brain tumor databases. The label dictionary is proposed to select foreground and background labels for automatic graph-cut segmentation. For quantitative evaluation, five different metric scores are computed using the online evaluation tool provided by the BraTS organizers. Experimental results demonstrate that the proposed approach achieves accurate results and outperforms most of the state-of-the-art methods cited in BraTS-MICCAI challenge.

show abstract

Robust Multimodal Dictionary Learning

Cited by 13 publications

References 10 publications

Brain tumor classification and segmentation using sparse coding and dictionary learning

Brain tumor classification and segmentation using sparse coding and dictionary learning

A unified framework based on graph consensus term for multi-view learning

Coupled Dictionary Learning for Automatic Multi-Label Brain Tumor Segmentation in Flair MRI images

Contact Info

Product

Resources

About