Implicit Regression in Subspace for High-Sensitivity CEST Imaging

Chu Chen; Yang Liu; Se Weon Park; Jizhou Li; Kannie W. Y. Chan; Raymond H. F. Chan

doi:10.1109/ISBI56570.2024.10635440

Implicit Regression in Subspace for High-Sensitivity CEST Imaging

Chu Chen
, Yang Liu
, Se Weon Park
, Jizhou Li
, Kannie W. Y. Chan
, Raymond H. F. Chan^*

^*Corresponding author for this work

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

2 Citations (Scopus)

Abstract

Chemical Exchange Saturation Transfer (CEST) MRI demonstrates its capability in significantly enhancing the detection of proteins and metabolites with low concentrations through exchangeable protons. The clinical application of CEST, however, is constrained by its low contrast and low signal-to-noise ratio (SNR) in the acquired data. Denoising, as one of the post-processing stages for CEST data, can effectively improve the accuracy of CEST quantification. In this work, by modeling spatial variant z-spectrums into low-dimensional subspace, we introduce Implicit Regression in Subspace (IRIS), which is an unsupervised denoising algorithm utilizing the excellent property of implicit neural representation for continuous mapping. Experiments conducted on both synthetic and in-vivo data demonstrate that our proposed method surpasses other CEST denoising methods regarding both qualitative and quantitative performance.

Original language	English
Title of host publication	IEEE International Symposium on Biomedical Imaging
Subtitle of host publication	ISBI 2024
Publisher	IEEE
Number of pages	5
ISBN (Electronic)	979-8-3503-1333-8
ISBN (Print)	979-8-3503-1334-5
DOIs	https://doi.org/10.1109/ISBI56570.2024.10635440
Publication status	Published - 2024
Event	21st IEEE International Symposium on Biomedical Imaging (ISBI 2024) - Megaron Athens International Conference Centre, Athens, Greece Duration: 27 May 2024 → 30 May 2024 https://biomedicalimaging.org/2024/

Publication series

Name
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	21st IEEE International Symposium on Biomedical Imaging (ISBI 2024)
Place	Greece
City	Athens
Period	27/05/24 → 30/05/24
Internet address	https://biomedicalimaging.org/2024/

Funding

This work is supported by HKRGC GRF grants CityU1101120, CityU11309922, CRF grant C1013-21GF, and HKRGC-NSFC Grant NCityU214/19.

Research Keywords

CEST MRI
Denoising
Neural Network

RGC Funding Information

RGC-funded

Access Output

10.1109/ISBI56570.2024.10635440

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

GRF: Efficient Mathematical and Data-driven Models for Closed-loop Adaptive Optics Systems for Ground-based Astronomy
CHAN, H. F. R. (Principal Investigator / Project Coordinator), Ke, R. (Co-Investigator), Ramlau, R. (Co-Investigator) & Wagner, R. (Co-Investigator)
1/01/23 → 13/12/24
Project: Research
NSFC: Mathematical Modeling and Analysis of Deep Neural Networks for Solving Structured Differential and Integral Models
CHAN, H. F. R. (Principal Investigator / Project Coordinator), Tang, T. (Co-Investigator) & YANG, J. (Co-Investigator)
1/01/20 → 31/12/24
Project: Research

Cite this

@inproceedings{2d6cc07a233f4f2882a65c5162b967c3,

title = "Implicit Regression in Subspace for High-Sensitivity CEST Imaging",

abstract = "Chemical Exchange Saturation Transfer (CEST) MRI demonstrates its capability in significantly enhancing the detection of proteins and metabolites with low concentrations through exchangeable protons. The clinical application of CEST, however, is constrained by its low contrast and low signal-to-noise ratio (SNR) in the acquired data. Denoising, as one of the post-processing stages for CEST data, can effectively improve the accuracy of CEST quantification. In this work, by modeling spatial variant z-spectrums into low-dimensional subspace, we introduce Implicit Regression in Subspace (IRIS), which is an unsupervised denoising algorithm utilizing the excellent property of implicit neural representation for continuous mapping. Experiments conducted on both synthetic and in-vivo data demonstrate that our proposed method surpasses other CEST denoising methods regarding both qualitative and quantitative performance. ",

keywords = "CEST MRI, Denoising, Neural Network",

author = "Chu Chen and Yang Liu and Park, \{Se Weon\} and Jizhou Li and Chan, \{Kannie W. Y.\} and Chan, \{Raymond H. F.\}",

year = "2024",

doi = "10.1109/ISBI56570.2024.10635440",

language = "English",

isbn = "979-8-3503-1334-5",

publisher = "IEEE",

booktitle = "IEEE International Symposium on Biomedical Imaging",

address = "United States",

note = "21st IEEE International Symposium on Biomedical Imaging (ISBI 2024) ; Conference date: 27-05-2024 Through 30-05-2024",

url = "https://biomedicalimaging.org/2024/",

}

TY - GEN

T1 - Implicit Regression in Subspace for High-Sensitivity CEST Imaging

AU - Chen, Chu

AU - Liu, Yang

AU - Park, Se Weon

AU - Li, Jizhou

AU - Chan, Kannie W. Y.

AU - Chan, Raymond H. F.

PY - 2024

Y1 - 2024

N2 - Chemical Exchange Saturation Transfer (CEST) MRI demonstrates its capability in significantly enhancing the detection of proteins and metabolites with low concentrations through exchangeable protons. The clinical application of CEST, however, is constrained by its low contrast and low signal-to-noise ratio (SNR) in the acquired data. Denoising, as one of the post-processing stages for CEST data, can effectively improve the accuracy of CEST quantification. In this work, by modeling spatial variant z-spectrums into low-dimensional subspace, we introduce Implicit Regression in Subspace (IRIS), which is an unsupervised denoising algorithm utilizing the excellent property of implicit neural representation for continuous mapping. Experiments conducted on both synthetic and in-vivo data demonstrate that our proposed method surpasses other CEST denoising methods regarding both qualitative and quantitative performance.

AB - Chemical Exchange Saturation Transfer (CEST) MRI demonstrates its capability in significantly enhancing the detection of proteins and metabolites with low concentrations through exchangeable protons. The clinical application of CEST, however, is constrained by its low contrast and low signal-to-noise ratio (SNR) in the acquired data. Denoising, as one of the post-processing stages for CEST data, can effectively improve the accuracy of CEST quantification. In this work, by modeling spatial variant z-spectrums into low-dimensional subspace, we introduce Implicit Regression in Subspace (IRIS), which is an unsupervised denoising algorithm utilizing the excellent property of implicit neural representation for continuous mapping. Experiments conducted on both synthetic and in-vivo data demonstrate that our proposed method surpasses other CEST denoising methods regarding both qualitative and quantitative performance.

KW - CEST MRI

KW - Denoising

KW - Neural Network

U2 - 10.1109/ISBI56570.2024.10635440

DO - 10.1109/ISBI56570.2024.10635440

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 979-8-3503-1334-5

BT - IEEE International Symposium on Biomedical Imaging

PB - IEEE

T2 - 21st IEEE International Symposium on Biomedical Imaging (ISBI 2024)

Y2 - 27 May 2024 through 30 May 2024

ER -

Implicit Regression in Subspace for High-Sensitivity CEST Imaging

Abstract

Publication series

Conference

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Fingerprint

Projects

GRF: Efficient Mathematical and Data-driven Models for Closed-loop Adaptive Optics Systems for Ground-based Astronomy

NSFC: Mathematical Modeling and Analysis of Deep Neural Networks for Solving Structured Differential and Integral Models

Cite this