On the compactness, efficiency, and representation of 3D convolutional networks: Brain parcellation as a pretext task

TY - CHAP

T1 - On the compactness, efficiency, and representation of 3D convolutional networks

T2 - 25th International Conference on Information Processing in Medical Imaging, IPMI 2017

AU - Li, Wenqi

AU - Wang, Guotai

AU - Fidon, Lucas

AU - Ourselin, Sebastien

AU - Cardoso, M. Jorge

AU - Vercauteren, Tom

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Deep convolutional neural networks are powerful tools for learning visual representations from images. However, designing efficient deep architectures to analyse volumetric medical images remains challenging. This work investigates efficient and flexible elements of modern convolutional networks such as dilated convolution and residual connection. With these essential building blocks, we propose a high-resolution, compact convolutional network for volumetric image segmentation. To illustrate its efficiency of learning 3D representation from large-scale image data, the proposed network is validated with the challenging task of parcellating 155 neuroanatomical structures from brain MR images. Our experiments show that the proposed network architecture compares favourably with state-of-the-art volumetric segmentation networks while being an order of magnitude more compact. We consider the brain parcellation task as a pretext task for volumetric image segmentation; our trained network potentially provides a good starting point for transfer learning. Additionally, we show the feasibility of voxel-level uncertainty estimation using a sampling approximation through dropout.

AB - Deep convolutional neural networks are powerful tools for learning visual representations from images. However, designing efficient deep architectures to analyse volumetric medical images remains challenging. This work investigates efficient and flexible elements of modern convolutional networks such as dilated convolution and residual connection. With these essential building blocks, we propose a high-resolution, compact convolutional network for volumetric image segmentation. To illustrate its efficiency of learning 3D representation from large-scale image data, the proposed network is validated with the challenging task of parcellating 155 neuroanatomical structures from brain MR images. Our experiments show that the proposed network architecture compares favourably with state-of-the-art volumetric segmentation networks while being an order of magnitude more compact. We consider the brain parcellation task as a pretext task for volumetric image segmentation; our trained network potentially provides a good starting point for transfer learning. Additionally, we show the feasibility of voxel-level uncertainty estimation using a sampling approximation through dropout.

UR - http://www.scopus.com/inward/record.url?scp=85020488981&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-59050-9_28

DO - 10.1007/978-3-319-59050-9_28

M3 - Conference paper

AN - SCOPUS:85020488981

SN - 9783319590493

VL - 10265 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 348

EP - 360

BT - Information Processing in Medical Imaging - 25th International Conference, IPMI 2017, Proceedings

PB - Springer Verlag

Y2 - 25 June 2017 through 30 June 2017

ER -