Journal
NEUROIMAGE
Volume 206, Issue -, Pages -Publisher
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.neuroimage.2019.116324
Keywords
Fetal MRI; Deep learning; Super resolution; Convolutional neural network; Brain localization; Segmentation
Funding
- Wellcome Trust [WT101957, WT97914, 203145Z/16/Z, 203148/Z/16/Z, 210182/Z/18/Z]
- Engineering and Physical Sciences Research Council (EPSRC) [NS/A000027/1, NS/A000049/1, NS/A000050/1, EP/L016478/1]
- National Institute for Health Research University College London Hospitals Biomedical Research Centre (NIHR BRC UCLH/UCL High Impact Initiative)
- Great Ormond Street Hospital Charity fund
- Klinische Onderzoeks en Opleidings-Raad UZ Leuven
- Medtronic/Royal Academy of Engineering Research Chair [RCSRF1819\7\34]
- Engineering and Physical Sciences Research Council [1676123] Funding Source: researchfish
- Wellcome Trust [101957/Z/13/Z] Funding Source: researchfish
Ask authors/readers for more resources
High-resolution volume reconstruction from multiple motion-corrupted stacks of 2D slices plays an increasing role for fetal brain Magnetic Resonance Imaging (MRI) studies. Currently existing reconstruction methods are time-consuming and often require user interactions to localize and extract the brain from several stacks of 2D slices. We propose a fully automatic framework for fetal brain reconstruction that consists of four stages: 1) fetal brain localization based on a coarse segmentation by a Convolutional Neural Network (CNN), 2) fine segmentation by another CNN trained with a multi-scale loss function, 3) novel, single-parameter outlier-robust super-resolution reconstruction, and 4) fast and automatic high-resolution visualization in standard anatomical space suitable for pathological brains. We validated our framework with images from fetuses with normal brains and with variable degrees of ventriculomegaly associated with open spina bifida, a congenital malformation affecting also the brain. Experiments show that each step of our proposed pipeline outperforms state-of-the-art methods in both segmentation and reconstruction comparisons including expert-reader quality assessments. The reconstruction results of our proposed method compare favorably with those obtained by manual, labor-intensive brain segmentation, which unlocks the potential use of automatic fetal brain reconstruction studies in clinical practice.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available