TY - GEN
T1 - Frnet
T2 - 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019
AU - Zhang, Qian
AU - Wang, Li
AU - Zong, Xiaopeng
AU - Lin, Weili
AU - Li, Gang
AU - Shen, Dinggang
N1 - Funding Information:
This work is supported in part by NIH grants (MH100217, MH107815, MH108914, MH109773, MH116225, MH117943, HD053000, MH104324 and U01MH110274) and the efforts of the UNC/UMN Baby Connectome Project Consortium.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/4
Y1 - 2019/4
N2 - Skull stripping for brain MR images is a basic segmentation task. Although many methods have been proposed, most of them focused mainly on the adult MR images. Skull strip-ping for infant MR images is more challenging due to the small size and dynamic intensity changes of brain tissues during the early ages. In this paper, we propose a novel CNN based framework to robustly extract brain region from infant MR image without any human assistance. Specifically, we propose a simplified but more robust flattened residual net-work architecture (FRnet). We also introduce a new boundary loss function to highlight ambiguous and low contrast regions between brain and non-brain regions. To make the whole framework more robust to MR images with different imaging quality, we further introduce an artifact simulator for data augmentation. We have trained and tested our proposed framework on a large dataset (N=343), covering newborns to 48-month-olds, and obtained performance better than the state-of-the-art methods in all age groups.
AB - Skull stripping for brain MR images is a basic segmentation task. Although many methods have been proposed, most of them focused mainly on the adult MR images. Skull strip-ping for infant MR images is more challenging due to the small size and dynamic intensity changes of brain tissues during the early ages. In this paper, we propose a novel CNN based framework to robustly extract brain region from infant MR image without any human assistance. Specifically, we propose a simplified but more robust flattened residual net-work architecture (FRnet). We also introduce a new boundary loss function to highlight ambiguous and low contrast regions between brain and non-brain regions. To make the whole framework more robust to MR images with different imaging quality, we further introduce an artifact simulator for data augmentation. We have trained and tested our proposed framework on a large dataset (N=343), covering newborns to 48-month-olds, and obtained performance better than the state-of-the-art methods in all age groups.
KW - Deep learning
KW - Infant brain
KW - Skull stripping
UR - http://www.scopus.com/inward/record.url?scp=85073914697&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2019.8759167
DO - 10.1109/ISBI.2019.8759167
M3 - Conference contribution
AN - SCOPUS:85073914697
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 999
EP - 1002
BT - ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
Y2 - 8 April 2019 through 11 April 2019
ER -