Real-time automatic fetal brain extraction in fetal MRI by deep learning

Salehi, Seyed Sadegh Mohseni; Hashemi, Seyed Raein; Velasco‐Annis, Clemente; Ouaalam, Abdelhakim; Estroff, Judy A.; Erdoğmuş, Deni̇z; Warfield, Simon K.; Gholipour, Ali

doi:10.1109/isbi.2018.8363675

Cited by 81 publications

(81 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both P-Net (S) and P-Net (S) + ML were applied to the output of P-Net (L). The method of Salehi et al [10] has a lower performance than our coarse-to-fine segmentation methods. P-Net (S) + ML achieves a better spatial consistency with reduced noises in the segmentation than P-Net (S).…”

Section: Experiments and Resultsmentioning

confidence: 66%

“…Fig. 2 presents a visual comparison of three methods for fetal brain segmentation applied to Group B1 and Group B2 respectively: 1) Salehi et al 2 [10], applying the U-Net to the whole input image for segmentation without a localization stage, 2) P-Net (S) trained with the basic Dice loss function (at a single scale), and 3) P-Net (S) + ML where P-Net (S) was trained with our proposed multi-scale loss function. Both P-Net (S) and P-Net (S) + ML were applied to the output of P-Net (L).…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…Compared with Keraudren et al [7], our localization method does not need prior information such as gestational age and achieved superior performance in less time. Unlike [10] which takes a whole image input to a CNN, Table 4. Summary of clinical evaluation.…”

Section: Discussionmentioning

confidence: 99%

“…The template-based automatic method proposed in [11] takes hours for localization. Convolutional Neural Networks (CNN) are used in in [10] for fast slice-based fetal brain segmentation but can easily obtain false positives and show poor performance for challenging cases.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An Automated Localization, Segmentation and Reconstruction Framework for Fetal Brain MRI

Ebner

Wang

et al. 2018

Medical Image Computing and Computer Assisted Intervention – MICCAI 2018

View full text Add to dashboard Cite

Reconstructing a high-resolution (HR) volume from motioncorrupted and sparsely acquired stacks plays an increasing role in fetal brain Magnetic Resonance Imaging (MRI) studies. Existing reconstruction methods are time-consuming and often require user interaction to localize and extract the brain from several stacks of 2D slices. In this paper, we propose a fully automatic framework for fetal brain reconstruction that consists of three stages: 1) brain localization based on a coarse segmentation of a down-sampled input image by a Convolutional Neural Network (CNN), 2) fine segmentation by a second CNN trained with a multi-scale loss function, and 3) novel, single-parameter outlier-robust super-resolution reconstruction (SRR) for HR visualization in the standard anatomical space. We validate our framework with images from fetuses with variable degrees of ventriculomegaly associated with spina bifida. Experiments show that each step of our proposed pipeline outperforms state-of-the-art methods in both segmentation and reconstruction comparisons. Overall, we report automatic SRR reconstructions that compare favorably with those obtained by manual, labor-intensive brain segmentations. This potentially unlocks the use of automatic fetal brain reconstruction studies in clinical practice.

show abstract

Section: Experiments and Resultsmentioning

confidence: 66%

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Automated Localization, Segmentation and Reconstruction Framework for Fetal Brain MRI

Ebner

Wang

et al. 2018

Medical Image Computing and Computer Assisted Intervention – MICCAI 2018

View full text Add to dashboard Cite

show abstract

“…Another category of automatic segmentation methods that do not rely on atlases are convolutional neural networks (CNNs). One popular type of CNN for the automatic segmentation of medical images is the U-Net [18], which has been used for the localization and segmentation of the fetal brain within the overall fetal image [14] and has been tested on the segmentation of a normal fetal brain [19]. We propose to use the same network to segment brain tissues within a high-resolution 3D reconstructed volume of a fetal SB brain, both pre-operatively and post-operatively.…”

Section: Introductionmentioning

confidence: 99%

Longitudinal Analysis of Fetal MRI in Patients with Prenatal Spina Bifida Repair

Payette

Moehrlen

Mazzone

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Open spina bifida (SB) is one of the most common congenital defects and can lead to impaired brain development. Emerging fetal surgery methods have shown considerable success in the treatment of patients with this severe anomaly. Afterwards, alterations in the brain development of these fetuses have been observed. Currently no longitudinal studies exist to show the effect of fetal surgery on brain development. In this work, we present a fetal MRI neuroimaging analysis pipeline for fetuses with SB, including automated fetal ventricle segmentation and deformation-based morphometry, and demonstrate its applicability with an analysis of ventricle enlargement in fetuses with SB. Using a robust super-resolution algorithm, we reconstructed fetal brains at both preoperative and post-operative time points and trained a U-Net CNN in order to automatically segment the ventricles. We investigated the change of ventricle shape post-operatively, and the impacts of lesion size, type, and GA at operation on the change in ventricle shape. No impact was found, except for moderately larger ventriculomegaly progression in myeloschisis patients. Prenatal ventricle volume growth was also investigated. Our method allows for the quantification of longitudinal morphological changes to fully quantify the impact of prenatal SB repair and could be applied to predict postnatal outcomes.

show abstract

Classification of Fetal Brain Abnormalities with MRI Images

Shinde

Thakare

2022

Object Detection by Stereo Vision Images

View full text Add to dashboard Cite

Real-time automatic fetal brain extraction in fetal MRI by deep learning

Cited by 81 publications

References 14 publications

An Automated Localization, Segmentation and Reconstruction Framework for Fetal Brain MRI

An Automated Localization, Segmentation and Reconstruction Framework for Fetal Brain MRI

Longitudinal Analysis of Fetal MRI in Patients with Prenatal Spina Bifida Repair

Classification of Fetal Brain Abnormalities with MRI Images

Contact Info

Product

Resources

About