Segmentation of MR images via discriminative dictionary learning and sparse coding: Application to hippocampus labeling

• Published in: NeuroImage (Orlando, Fla.) Vol. 76; pp. 11 - 23
• Main Authors: Tong, Tong, Wolz, Robin, Coupé, Pierrick, Hajnal, Joseph V., Rueckert, Daniel
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01.08.2013; Elsevier; Elsevier Limited
• Subjects: Adult; Aged; Automation; Bioengineering; Biological and medical sciences; Brain Mapping - methods; Computer Science; Dictionaries; Discrimination Learning - physiology; Discriminative dictionary learning; Distance learning; Engineering Sciences; Female; Fundamental and applied biological sciences. Psychology; Hippocampus - physiology; Humans; Image Interpretation, Computer-Assisted - methods; Image Processing; Labeling; Libraries; Life Sciences; Magnetic Resonance Imaging - methods; Male; Medical Imaging; Methods; NMR; Nuclear magnetic resonance; Patch-based segmentation; Registration; Signal and Image processing; Sparse representation; Structural MR images; Vertebrates: nervous system and sense organs; Structural MR images; Patch-based segmentation; Sparse representation; Discriminative dictionary learning; Learning; Acquisition process; Segmentation; Central nervous system; Hippocampus; Encephalon
• DOI: 10.1016/j.neuroimage.2013.02.069
• ISSN: 1053-8119; 1095-9572

We propose a novel method for the automatic segmentation of brain MRI images by using discriminative dictionary learning and sparse coding techniques. In the proposed method, dictionaries and classifiers are learned simultaneously from a set of brain atlases, which can then be used for the reconstruction and segmentation of an unseen target image. The proposed segmentation strategy is based on image reconstruction, which is in contrast to most existing atlas-based labeling approaches that rely on comparing image similarities between atlases and target images. In addition, we propose a Fixed Discriminative Dictionary Learning for Segmentation (F-DDLS) strategy, which can learn dictionaries offline and perform segmentations online, enabling a significant speed-up in the segmentation stage. The proposed method has been evaluated for the hippocampus segmentation of 80 healthy ICBM subjects and 202 ADNI images. The robustness of the proposed method, especially of our F-DDLS strategy, was validated by training and testing on different subject groups in the ADNI database. The influence of different parameters was studied and the performance of the proposed method was also compared with that of the nonlocal patch-based approach. The proposed method achieved a median Dice coefficient of 0.879 on 202 ADNI images and 0.890 on 80 ICBM subjects, which is competitive compared with state-of-the-art methods. •Sparse representation technique is applied to segmentations of brain MR images.•Discriminative dictionary learning is used to achieve a fast implementation.•Validation is carried out on hippocampus of 80 ICBM subjects and 202 ADNI images.•Segmentation results demonstrate the accuracy of the proposed method.•The proposed method may provide a potential direction for human brain labeling.