Joint super-resolution and synthesis of 1 mm isotropic MP-RAGE volumes from clinical MRI exams with scans of different orientation, resolution and contrast

•SynthSR turns clinical scans of different resolution and contrast into 1 mm MPRAGEs.•It relies on a CNN trained on fake images synthesized on the fly at every minibatch.•It can be retrained for any combination of resolutions / contrasts without new data.•It enables segmentation, registration, etc w...

Full description

Saved in:

Bibliographic Details
Published in:	NeuroImage (Orlando, Fla.) Fla.), 2021-08, Vol.237, p.118206-118206, Article 118206
Main Authors:	Iglesias, Juan Eugenio, Billot, Benjamin, Balbastre, Yaël, Tabari, Azadeh, Conklin, John, Gilberto González, R., Alexander, Daniel C., Golland, Polina, Edlow, Brian L., Fischl, Bruce
Format:	Article
Language:	eng
Subjects:	Accuracy Algorithms Brain - diagnostic imaging Brain research Clinical medicine Clinical scans Computer Simulation Convolutional neural network Deep Learning Humans Image processing Magnetic resonance imaging Magnetic Resonance Imaging - methods Models, Theoretical Morphometry Neural networks Neuroimaging Neuroimaging - methods Public software Registration Segmentation Super-resolution
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	•SynthSR turns clinical scans of different resolution and contrast into 1 mm MPRAGEs.•It relies on a CNN trained on fake images synthesized on the fly at every minibatch.•It can be retrained for any combination of resolutions / contrasts without new data.•It enables segmentation, registration, etc with existing software (e.g. FreeSurfer) Code is open source. [Display omitted] Most existing algorithms for automatic 3D morphometry of human brain MRI scans are designed for data with near-isotropic voxels at approximately 1 mm resolution, and frequently have contrast constraints as well-typically requiring T1-weighted images (e.g., MP-RAGE scans). This limitation prevents the analysis of millions of MRI scans acquired with large inter-slice spacing in clinical settings every year. In turn, the inability to quantitatively analyze these scans hinders the adoption of quantitative neuro imaging in healthcare, and also precludes research studies that could attain huge sample sizes and hence greatly improve our understanding of the human brain. Recent advances in convolutional neural networks (CNNs) are producing outstanding results in super-resolution and contrast synthesis of MRI. However, these approaches are very sensitive to the specific combination of contrast, resolution and orientation of the input images, and thus do not generalize to diverse clinical acquisition protocols – even within sites. In this article, we present SynthSR, a method to train a CNN that receives one or more scans with spaced slices, acquired with different contrast, resolution and orientation, and produces an isotropic scan of canonical contrast (typically a 1 mm MP-RAGE). The presented method does not require any preprocessing, beyond rigid coregistration of the input scans. Crucially, SynthSR trains on synthetic input images generated from 3D segmentations, and can thus be used to train CNNs for any combination of contrasts, resolutions and orientations without high-resolution real images of the input contrasts. We test the images generated with SynthSR in an array of common downstream analyses, and show that they can be reliably used for subcortical segmentation and volumetry, image registration (e.g., for tensor-based morphometry), and, if some image quality requirements are met, even cortical thickness morphometry. The source code is publicly available at https://github.com/BBillot/SynthSR.
ISSN:	1053-8119 1095-9572