Robust Super-resolution Volume Reconstruction from Slice Acquisitions: Application to Fetal Brain MRI

By A. Gholipour, Judy A. Estroff, Simon Keith Warfield. IEEE Transactions on Medical Imaging, 2010

366 citations recorded by the source corpus.

Fast magnetic resonance imaging slice acquisition techniques such as single shot fast spin echo are routinely used in the presence of uncontrollable motion. These techniques are widely used for fetal magnetic resonance imaging (MRI) and MRI of moving subjects and organs. Although high-quality slices are frequently acquired by these techniques, inter-slice motion leads to severe motion artifacts that are apparent in out-of-plane views. Slice sequential acquisitions do not enable 3-D volume representation. In this study, we have developed a novel technique based on a slice acquisition model, which enables the reconstruction of a volumetric image from multiple-scan slice acquisitions. The super-resolution volume reconstruction is formulated as an inverse problem of finding the underlying structure generating the acquired slices. We have developed a robust M-estimation solution which minimizes a robust error norm function between the model-generated slices and the acquired slices. The accuracy and robustness of this novel technique has been quantitatively assessed through simulations with digital brain phantom images as well as high-resolution newborn images. We also report here successful application of our new technique for the reconstruction of volumetric fetal brain MRI from clinically acquired data.

Subfields: Pediatrics, Perinatology and Child Health, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/da8f3d5a6e689caa2f7ac6afc18ccbaea35d4683

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Pediatrics, Perinatology and Child Health

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Reconstruction of fetal brain MRI with intensity matching and complete outlier removal (2012, methodology): “It therefore became a standard to simulate acquisition of 2D slices corrupted by motion from neonatal 3D MRI (Rousseau et al., 2006; Jiang et al., 2007; Kim et al., 2010; Gholipour et al., 2010).”
A normative spatiotemporal MRI atlas of the fetal brain for automatic segmentation and analysis of early brain growth (2017, methodology): “The reconstruction processing time was between 1 and 20 hours depending on the size and number of the images and the amount of motion, and involved between 3 to 10 iterations of motion correction and super-resolution volume reconstruction using 38 which was done automatically.”
Multiscale brain MRI super-resolution using deep 3D convolutional networks (2019, background): “Deeper architectures may even lower the reconstruction performance with non-289 residual learning. can be unstable when adding layers (Glorot and Bengio 2010).”
A review on automatic fetal and neonatal brain MRI segmentation (2017, neutral): “(2011)(1) T1,T2,FA,MD 122 regions 1 term 1 Gousias et al. (2012)(2) T1,T2 50 regions 20 term 20 de Macedo Rodrigues et al.”
Real-time Deep Pose Estimation with Geodesic Loss for Image-to-Template Rigid Registration (2018, methodology): “structed fetal brain MRI images [23] to a standard (atlas) space.”
Cardiac Image Super-Resolution with Global Correspondence Using Multi-Atlas PatchMatch (2013, methodology): “By combining patch redundancy [10] and the formulation proposed in [7], we propose a novel model with two terms ΦSR = Φ 1 SR +Φ 2 SR to reconstruct the SR image I where Ω is the image domain.”
Slice-to-volume medical image registration: a survey (2017, methodology): “…as proposed in a number of papers (Chandler et al., 2008; Elen et al., 2010; Fei et al., 2002, 2003a; Frühwald et al., 2009; Fuerst et al., 2014; Gholipour et al., 2010; Huang et al., 2009; Jiang et al., 2007a; Kainz et al., 2015; Kim et al., 1999, 2010a; Nir and Tannenbaum, 2011; Penney et…”