TY - GEN
T1 - Segmentation of perivascular spaces using vascular features and structured random forest from 7T MR image
AU - Zhang, Jun
AU - Gao, Yaozong
AU - Park, Sang Hyun
AU - Zong, Xiaopeng
AU - Lin, Weili
AU - Shen, Dinggang
N1 - Publisher Copyright:
© Springer International Publishing AG 2016.
PY - 2016
Y1 - 2016
N2 - Quantitative analysis of perivascular spaces (PVSs) is important to reveal the correlations between cerebrovascular lesions and neurodegenerative diseases. In this study, we propose a learning-based segmentation framework to extract the PVSs from high-resolution 7T MR images. Specifically, we integrate three types of vascular filter responses into a structured random forest for classifying voxels into PVS and background. In addition, we also propose a novel entropy-based sampling strategy to extract informative samples in the background for training the classification model. Since various vascular features can be extracted by the three vascular filters, even thin and low-contrast structures can be effectively extracted from the noisy background. Moreover, continuous and smooth segmentation results can be obtained by utilizing the patch-based structured labels. The segmentation performance is evaluated on 19 subjects with 7T MR images, and the experimental results demonstrate that the joint use of entropy-based sampling strategy, vascular features and structured learning improves the segmentation accuracy, with the Dice similarity coefficient reaching 66%.
AB - Quantitative analysis of perivascular spaces (PVSs) is important to reveal the correlations between cerebrovascular lesions and neurodegenerative diseases. In this study, we propose a learning-based segmentation framework to extract the PVSs from high-resolution 7T MR images. Specifically, we integrate three types of vascular filter responses into a structured random forest for classifying voxels into PVS and background. In addition, we also propose a novel entropy-based sampling strategy to extract informative samples in the background for training the classification model. Since various vascular features can be extracted by the three vascular filters, even thin and low-contrast structures can be effectively extracted from the noisy background. Moreover, continuous and smooth segmentation results can be obtained by utilizing the patch-based structured labels. The segmentation performance is evaluated on 19 subjects with 7T MR images, and the experimental results demonstrate that the joint use of entropy-based sampling strategy, vascular features and structured learning improves the segmentation accuracy, with the Dice similarity coefficient reaching 66%.
UR - http://www.scopus.com/inward/record.url?scp=84992491329&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-47157-0_8
DO - 10.1007/978-3-319-47157-0_8
M3 - Conference contribution
AN - SCOPUS:84992491329
SN - 9783319471563
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 61
EP - 68
BT - Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings
A2 - Wang, Li
A2 - Suk, Heung-Il
A2 - Shi, Yinghuan
A2 - Adeli, Ehsan
A2 - Wang, Qian
PB - Springer Verlag
T2 - 7th International Workshop on Machine Learning in Medical Imaging, MLMI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016
Y2 - 17 October 2016 through 17 October 2016
ER -