Random forest regression for magnetic resonance image synthesis

•We describe an MRI image synthesis algorithm capable of synthesizing full-head T2w images and FLAIR images.•Our algorithm, REPLICA, is a supervised method and learns the nonlinear intensity mappings for synthesis using innovative features and a multi-resolution design.•We show significant improveme...

Full description

Saved in:

Bibliographic Details
Published in:	Medical image analysis 2017-01, Vol.35, p.475-488
Main Authors:	Jog, Amod, Carass, Aaron, Roy, Snehashis, Pham, Dzung L., Prince, Jerry L.
Format:	Article
Language:	eng
Subjects:	Algorithms Brain Datasets Datasets as Topic Experiments Humans Image analysis Image enhancement Image Enhancement - methods Image processing Image synthesis Magnetic resonance imaging Magnetic Resonance Imaging - methods MRI Neuroimaging NMR Nuclear magnetic resonance Random forests Regression analysis Resonance Standardization Synthesis
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	•We describe an MRI image synthesis algorithm capable of synthesizing full-head T2w images and FLAIR images.•Our algorithm, REPLICA, is a supervised method and learns the nonlinear intensity mappings for synthesis using innovative features and a multi-resolution design.•We show significant improvement in synthetic image quality over state-of-the-art image synthesis algorithms.•We also demonstrate that image analysis tasks like segmentation perform similarly for real and REPLICA-generated synthetic images.•REPLICA is computationally very fast and can be easily used as a preprocessing tool before further image analysis. [Display omitted] By choosing different pulse sequences and their parameters, magnetic resonance imaging (MRI) can generate a large variety of tissue contrasts. This very flexibility, however, can yield inconsistencies with MRI acquisitions across datasets or scanning sessions that can in turn cause inconsistent automated image analysis. Although image synthesis of MR images has been shown to be helpful in addressing this problem, an inability to synthesize both T2-weighted brain images that include the skull and FLuid Attenuated Inversion Recovery (FLAIR) images has been reported. The method described herein, called REPLICA, addresses these limitations. REPLICA is a supervised random forest image synthesis approach that learns a nonlinear regression to predict intensities of alternate tissue contrasts given specific input tissue contrasts. Experimental results include direct image comparisons between synthetic and real images, results from image analysis tasks on both synthetic and real images, and comparison against other state-of-the-art image synthesis methods. REPLICA is computationally fast, and is shown to be comparable to other methods on tasks they are able to perform. Additionally REPLICA has the capability to synthesize both T2-weighted images of the full head and FLAIR images, and perform intensity standardization between different imaging datasets.
ISSN:	1361-8415 1361-8423