Abstract
In many neuroscience and clinical studies, accurate measurement of hippocampus is very important to reveal the inter-subject anatomical differences or the subtle intra-subject longitudinal changes due to aging or dementia. Although many automatic segmentation methods have been developed, their performances are still challenged by the poor image contrast of hippocampus in the MR images acquired especially from 1.5 or 3.0Tesla (T) scanners. With the recent advance of imaging technology, 7.0T scanner provides much higher image contrast and resolution for hippocampus study. However, the previous methods developed for segmentation of hippocampus from 1.5T or 3.0T images do not work for the 7.0T images, due to different levels of imaging contrast and texture information. In this paper, we present a learning-based algorithm for automatic segmentation of hippocampi from 7.0T images, by taking advantages of the state-of-the-art multi-atlas framework and also the auto-context model (ACM). Specifically, ACM is performed in each atlas domain to iteratively construct sequences of location-adaptive classifiers by integrating both image appearance and local context features. Due to the plenty texture information in 7.0T images, more advanced texture features are also extracted and incorporated into the ACM during the training stage. Then, under the multi-atlas segmentation framework, multiple sequences of ACM-based classifiers are trained for all atlases to incorporate the anatomical variability. In the application stage, for a new image, its hippocampus segmentation can be achieved by fusing the labeling results from all atlases, each of which is obtained by applying the atlas-specific ACM-based classifiers. Experimental results on twenty 7.0T images with the voxel size of 0.35×0.35×0.35mm3 show very promising hippocampus segmentations (in terms of Dice overlap ratio 89.1±0.020), indicating high applicability for the future clinical and neuroscience studies.
| Original language | English |
|---|---|
| Pages (from-to) | 335-345 |
| Number of pages | 11 |
| Journal | NeuroImage |
| Volume | 83 |
| DOIs | |
| Publication status | Published - 2013 Dec 1 |
Fingerprint
Keywords
- 7.0T MRI
- Auto-context model
- Automatic hippocampus segmentation
- Label fusion
- Multiple atlases based segmentation
ASJC Scopus subject areas
- Cognitive Neuroscience
- Neurology
Cite this
Automatic hippocampus segmentation of 7.0Tesla MR images by combining multiple atlases and auto-context models. / Kim, Minjeong; Wu, Guorong; Li, Wei; Wang, Li; Son, Young Don; Cho, Zang Hee; Shen, Dinggang.
In: NeuroImage, Vol. 83, 01.12.2013, p. 335-345.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Automatic hippocampus segmentation of 7.0Tesla MR images by combining multiple atlases and auto-context models
AU - Kim, Minjeong
AU - Wu, Guorong
AU - Li, Wei
AU - Wang, Li
AU - Son, Young Don
AU - Cho, Zang Hee
AU - Shen, Dinggang
PY - 2013/12/1
Y1 - 2013/12/1
N2 - In many neuroscience and clinical studies, accurate measurement of hippocampus is very important to reveal the inter-subject anatomical differences or the subtle intra-subject longitudinal changes due to aging or dementia. Although many automatic segmentation methods have been developed, their performances are still challenged by the poor image contrast of hippocampus in the MR images acquired especially from 1.5 or 3.0Tesla (T) scanners. With the recent advance of imaging technology, 7.0T scanner provides much higher image contrast and resolution for hippocampus study. However, the previous methods developed for segmentation of hippocampus from 1.5T or 3.0T images do not work for the 7.0T images, due to different levels of imaging contrast and texture information. In this paper, we present a learning-based algorithm for automatic segmentation of hippocampi from 7.0T images, by taking advantages of the state-of-the-art multi-atlas framework and also the auto-context model (ACM). Specifically, ACM is performed in each atlas domain to iteratively construct sequences of location-adaptive classifiers by integrating both image appearance and local context features. Due to the plenty texture information in 7.0T images, more advanced texture features are also extracted and incorporated into the ACM during the training stage. Then, under the multi-atlas segmentation framework, multiple sequences of ACM-based classifiers are trained for all atlases to incorporate the anatomical variability. In the application stage, for a new image, its hippocampus segmentation can be achieved by fusing the labeling results from all atlases, each of which is obtained by applying the atlas-specific ACM-based classifiers. Experimental results on twenty 7.0T images with the voxel size of 0.35×0.35×0.35mm3 show very promising hippocampus segmentations (in terms of Dice overlap ratio 89.1±0.020), indicating high applicability for the future clinical and neuroscience studies.
AB - In many neuroscience and clinical studies, accurate measurement of hippocampus is very important to reveal the inter-subject anatomical differences or the subtle intra-subject longitudinal changes due to aging or dementia. Although many automatic segmentation methods have been developed, their performances are still challenged by the poor image contrast of hippocampus in the MR images acquired especially from 1.5 or 3.0Tesla (T) scanners. With the recent advance of imaging technology, 7.0T scanner provides much higher image contrast and resolution for hippocampus study. However, the previous methods developed for segmentation of hippocampus from 1.5T or 3.0T images do not work for the 7.0T images, due to different levels of imaging contrast and texture information. In this paper, we present a learning-based algorithm for automatic segmentation of hippocampi from 7.0T images, by taking advantages of the state-of-the-art multi-atlas framework and also the auto-context model (ACM). Specifically, ACM is performed in each atlas domain to iteratively construct sequences of location-adaptive classifiers by integrating both image appearance and local context features. Due to the plenty texture information in 7.0T images, more advanced texture features are also extracted and incorporated into the ACM during the training stage. Then, under the multi-atlas segmentation framework, multiple sequences of ACM-based classifiers are trained for all atlases to incorporate the anatomical variability. In the application stage, for a new image, its hippocampus segmentation can be achieved by fusing the labeling results from all atlases, each of which is obtained by applying the atlas-specific ACM-based classifiers. Experimental results on twenty 7.0T images with the voxel size of 0.35×0.35×0.35mm3 show very promising hippocampus segmentations (in terms of Dice overlap ratio 89.1±0.020), indicating high applicability for the future clinical and neuroscience studies.
KW - 7.0T MRI
KW - Auto-context model
KW - Automatic hippocampus segmentation
KW - Label fusion
KW - Multiple atlases based segmentation
UR - http://www.scopus.com/inward/record.url?scp=84880956855&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84880956855&partnerID=8YFLogxK
U2 - 10.1016/j.neuroimage.2013.06.006
DO - 10.1016/j.neuroimage.2013.06.006
M3 - Article
C2 - 23769921
AN - SCOPUS:84880956855
VL - 83
SP - 335
EP - 345
JO - NeuroImage
JF - NeuroImage
SN - 1053-8119
ER -