Uncertainty-Guided Efficient Interactive Refinement of Fetal Brain Segmentation from Stacks of MRI Slices

TY - CHAP

T1 - Uncertainty-Guided Efficient Interactive Refinement of Fetal Brain Segmentation from Stacks of MRI Slices

AU - Wang, Guotai

AU - Aertsen, Michael

AU - Deprest, Jan

AU - Ourselin, Sébastien

AU - Vercauteren, Tom

AU - Zhang, Shaoting

PY - 2020

Y1 - 2020

N2 - Segmentation of the fetal brain from stacks of motion-corrupted fetal MRI slices is important for motion correction and high-resolution volume reconstruction. Although Convolutional Neural Networks (CNNs) have been widely used for automatic segmentation of the fetal brain, their results may still benefit from interactive refinement for challenging slices. To improve the efficiency of interactive refinement process, we propose an Uncertainty-Guided Interactive Refinement (UGIR) framework. We first propose a grouped convolution-based CNN to obtain multiple automatic segmentation predictions with uncertainty estimation in a single forward pass, then guide the user to provide interactions only in a subset of slices with the highest uncertainty. A novel interactive level set method is also proposed to obtain a refined result given the initial segmentation and user interactions. Experimental results show that: (1) our proposed CNN obtains uncertainty estimation in real time which correlates well with mis-segmentations, (2) the proposed interactive level set is effective and efficient for refinement, (3) UGIR obtains accurate refinement results with around 30% improvement of efficiency by using uncertainty to guide user interactions. Our code is available online (https://github.com/HiLab-git/UGIR).

AB - Segmentation of the fetal brain from stacks of motion-corrupted fetal MRI slices is important for motion correction and high-resolution volume reconstruction. Although Convolutional Neural Networks (CNNs) have been widely used for automatic segmentation of the fetal brain, their results may still benefit from interactive refinement for challenging slices. To improve the efficiency of interactive refinement process, we propose an Uncertainty-Guided Interactive Refinement (UGIR) framework. We first propose a grouped convolution-based CNN to obtain multiple automatic segmentation predictions with uncertainty estimation in a single forward pass, then guide the user to provide interactions only in a subset of slices with the highest uncertainty. A novel interactive level set method is also proposed to obtain a refined result given the initial segmentation and user interactions. Experimental results show that: (1) our proposed CNN obtains uncertainty estimation in real time which correlates well with mis-segmentations, (2) the proposed interactive level set is effective and efficient for refinement, (3) UGIR obtains accurate refinement results with around 30% improvement of efficiency by using uncertainty to guide user interactions. Our code is available online (https://github.com/HiLab-git/UGIR).

KW - Fetal brain

KW - Interactive segmentation

KW - Uncertainty

UR - http://www.scopus.com/inward/record.url?scp=85092792261&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-59719-1_28

DO - 10.1007/978-3-030-59719-1_28

M3 - Conference paper

AN - SCOPUS:85092792261

SN - 9783030597184

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

SP - 279

EP - 288

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2020 - 23rd International Conference, Proceedings

A2 - Martel, Anne L.

A2 - Abolmaesumi, Purang

A2 - Stoyanov, Danail

A2 - Mateus, Diana

A2 - Zuluaga, Maria A.

A2 - Zhou, S. Kevin

A2 - Racoceanu, Daniel

A2 - Joskowicz, Leo

PB - Springer Science and Business Media Deutschland GmbH

T2 - 23rd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2020

Y2 - 4 October 2020 through 8 October 2020

ER -