Super-resolution reconstruction of neonatal brain magnetic resonance images via residual structured sparse representation

Yongqin Zhang, Pew Thian Yap, Geng Chen, Weili Lin, Li Wang, Dinggang Shen

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Magnetic resonance images of neonates, compared with toddlers, exhibit lower signal-to-noise ratio and spatial resolution. In this paper, we propose a novel method for super-resolution reconstruction of neonate images with the help of toddler images, using residual-structured sparse representation with convex regularization. Specifically, we introduce a two-layer image representation, consisting of a base layer and a detail layer, to cater to signal variation across scanners and sites. The base layer consists of the smoothed version of the image obtained via Gaussian filtering. The detail layer is the difference between the original image and the base layer. High-frequency details in the detail layer are borrowed across subjects for super-resolution reconstruction. Experimental results on T1 and T2 images demonstrate that the proposed algorithm can recover fine anatomical structures, and generally outperform the state-of-the-art methods both qualitatively and quantitatively.

Original languageEnglish
Pages (from-to)76-87
Number of pages12
JournalMedical Image Analysis
Volume55
DOIs
Publication statusPublished - 2019 Jul 1

Fingerprint

Magnetic resonance
Brain
Signal to noise ratio
Magnetic Resonance Spectroscopy
Computer-Assisted Image Processing
Signal-To-Noise Ratio

Keywords

  • Convex optimization
  • Dictionary learning
  • Magnetic resonance imaging
  • Sparse representation

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Graphics and Computer-Aided Design

Cite this

Super-resolution reconstruction of neonatal brain magnetic resonance images via residual structured sparse representation. / Zhang, Yongqin; Yap, Pew Thian; Chen, Geng; Lin, Weili; Wang, Li; Shen, Dinggang.

In: Medical Image Analysis, Vol. 55, 01.07.2019, p. 76-87.

Research output: Contribution to journalArticle

@article{cc6c407fe3ce4f3ca4e9dfed49af11eb,
title = "Super-resolution reconstruction of neonatal brain magnetic resonance images via residual structured sparse representation",
abstract = "Magnetic resonance images of neonates, compared with toddlers, exhibit lower signal-to-noise ratio and spatial resolution. In this paper, we propose a novel method for super-resolution reconstruction of neonate images with the help of toddler images, using residual-structured sparse representation with convex regularization. Specifically, we introduce a two-layer image representation, consisting of a base layer and a detail layer, to cater to signal variation across scanners and sites. The base layer consists of the smoothed version of the image obtained via Gaussian filtering. The detail layer is the difference between the original image and the base layer. High-frequency details in the detail layer are borrowed across subjects for super-resolution reconstruction. Experimental results on T1 and T2 images demonstrate that the proposed algorithm can recover fine anatomical structures, and generally outperform the state-of-the-art methods both qualitatively and quantitatively.",
keywords = "Convex optimization, Dictionary learning, Magnetic resonance imaging, Sparse representation",
author = "Yongqin Zhang and Yap, {Pew Thian} and Geng Chen and Weili Lin and Li Wang and Dinggang Shen",
year = "2019",
month = "7",
day = "1",
doi = "10.1016/j.media.2019.04.010",
language = "English",
volume = "55",
pages = "76--87",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",

}

TY - JOUR

T1 - Super-resolution reconstruction of neonatal brain magnetic resonance images via residual structured sparse representation

AU - Zhang, Yongqin

AU - Yap, Pew Thian

AU - Chen, Geng

AU - Lin, Weili

AU - Wang, Li

AU - Shen, Dinggang

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Magnetic resonance images of neonates, compared with toddlers, exhibit lower signal-to-noise ratio and spatial resolution. In this paper, we propose a novel method for super-resolution reconstruction of neonate images with the help of toddler images, using residual-structured sparse representation with convex regularization. Specifically, we introduce a two-layer image representation, consisting of a base layer and a detail layer, to cater to signal variation across scanners and sites. The base layer consists of the smoothed version of the image obtained via Gaussian filtering. The detail layer is the difference between the original image and the base layer. High-frequency details in the detail layer are borrowed across subjects for super-resolution reconstruction. Experimental results on T1 and T2 images demonstrate that the proposed algorithm can recover fine anatomical structures, and generally outperform the state-of-the-art methods both qualitatively and quantitatively.

AB - Magnetic resonance images of neonates, compared with toddlers, exhibit lower signal-to-noise ratio and spatial resolution. In this paper, we propose a novel method for super-resolution reconstruction of neonate images with the help of toddler images, using residual-structured sparse representation with convex regularization. Specifically, we introduce a two-layer image representation, consisting of a base layer and a detail layer, to cater to signal variation across scanners and sites. The base layer consists of the smoothed version of the image obtained via Gaussian filtering. The detail layer is the difference between the original image and the base layer. High-frequency details in the detail layer are borrowed across subjects for super-resolution reconstruction. Experimental results on T1 and T2 images demonstrate that the proposed algorithm can recover fine anatomical structures, and generally outperform the state-of-the-art methods both qualitatively and quantitatively.

KW - Convex optimization

KW - Dictionary learning

KW - Magnetic resonance imaging

KW - Sparse representation

UR - http://www.scopus.com/inward/record.url?scp=85066918559&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066918559&partnerID=8YFLogxK

U2 - 10.1016/j.media.2019.04.010

DO - 10.1016/j.media.2019.04.010

M3 - Article

VL - 55

SP - 76

EP - 87

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

ER -