Techniques for detection of spectral signatures in MR images and their applications

Clayton Chi Chang Chen, Chuin Mu Wang, Chein I. Chang, Ching Wen Yang, Jyh Wen Chai, Yi Nung Chung, Pau Choo Chung

Research output: Chapter in Book/Report/Conference proceedingChapter

2 Citations (Scopus)

Abstract

This chapter presents two major hyperspectral imaging techniques, orthogonal subspace projection (OSP) and constrained energy minimization (CEM), which have recently found success in spectral signature detection in magnetic resonance (MR) images. Unlike classical image classification techniques which are primarily designed on a pure pixel basis, the proposed OSP and CEM are mixed pixel classification techniques which estimate abundance fractions of different material substances assumed to be present in the image data so as to achieve classification. Three scenarios are considered for MR image classification based on different levels of object information provided a priori. The OSP is applied to the case that the complete prior object knowledge is known compared to the CEM which is used for the case that only desired object knowledge is given while other knowledge can be discarded. When no prior object knowledge is available, the OSP is extended to an unsupervised OSP (UOSP), which obtains the necessary object knowledge directly from the image data in an unsupervised manner for OSP classification. In order to demonstrate the utility of these three scenarios in MR image classification, a series of experiments are conducted and compared to the commonly used c-means method for performance evaluation. The results show that both OSP and CEM are promising and effective spectral techniques for MR image classification.

Original languageEnglish
Title of host publicationMedical Imaging Systems Technology
Subtitle of host publicationVolume 4: Methods in Diagnosis Optimization
PublisherWorld Scientific Publishing Co.
Pages297-328
Number of pages32
ISBN (Electronic)9789812701053
ISBN (Print)9812563644, 9789812569905
DOIs
Publication statusPublished - 2005 Jan 1

Fingerprint

Magnetic Resonance Spectroscopy

All Science Journal Classification (ASJC) codes

  • Medicine(all)

Cite this

Chen, C. C. C., Wang, C. M., Chang, C. I., Yang, C. W., Chai, J. W., Chung, Y. N., & Chung, P. C. (2005). Techniques for detection of spectral signatures in MR images and their applications. In Medical Imaging Systems Technology: Volume 4: Methods in Diagnosis Optimization (pp. 297-328). World Scientific Publishing Co.. https://doi.org/10.1142/9789812701053_0009
Chen, Clayton Chi Chang ; Wang, Chuin Mu ; Chang, Chein I. ; Yang, Ching Wen ; Chai, Jyh Wen ; Chung, Yi Nung ; Chung, Pau Choo. / Techniques for detection of spectral signatures in MR images and their applications. Medical Imaging Systems Technology: Volume 4: Methods in Diagnosis Optimization. World Scientific Publishing Co., 2005. pp. 297-328
@inbook{36a7f86939a6406bab2e3485190a9e94,
title = "Techniques for detection of spectral signatures in MR images and their applications",
abstract = "This chapter presents two major hyperspectral imaging techniques, orthogonal subspace projection (OSP) and constrained energy minimization (CEM), which have recently found success in spectral signature detection in magnetic resonance (MR) images. Unlike classical image classification techniques which are primarily designed on a pure pixel basis, the proposed OSP and CEM are mixed pixel classification techniques which estimate abundance fractions of different material substances assumed to be present in the image data so as to achieve classification. Three scenarios are considered for MR image classification based on different levels of object information provided a priori. The OSP is applied to the case that the complete prior object knowledge is known compared to the CEM which is used for the case that only desired object knowledge is given while other knowledge can be discarded. When no prior object knowledge is available, the OSP is extended to an unsupervised OSP (UOSP), which obtains the necessary object knowledge directly from the image data in an unsupervised manner for OSP classification. In order to demonstrate the utility of these three scenarios in MR image classification, a series of experiments are conducted and compared to the commonly used c-means method for performance evaluation. The results show that both OSP and CEM are promising and effective spectral techniques for MR image classification.",
author = "Chen, {Clayton Chi Chang} and Wang, {Chuin Mu} and Chang, {Chein I.} and Yang, {Ching Wen} and Chai, {Jyh Wen} and Chung, {Yi Nung} and Chung, {Pau Choo}",
year = "2005",
month = "1",
day = "1",
doi = "10.1142/9789812701053_0009",
language = "English",
isbn = "9812563644",
pages = "297--328",
booktitle = "Medical Imaging Systems Technology",
publisher = "World Scientific Publishing Co.",
address = "United States",

}

Chen, CCC, Wang, CM, Chang, CI, Yang, CW, Chai, JW, Chung, YN & Chung, PC 2005, Techniques for detection of spectral signatures in MR images and their applications. in Medical Imaging Systems Technology: Volume 4: Methods in Diagnosis Optimization. World Scientific Publishing Co., pp. 297-328. https://doi.org/10.1142/9789812701053_0009

Techniques for detection of spectral signatures in MR images and their applications. / Chen, Clayton Chi Chang; Wang, Chuin Mu; Chang, Chein I.; Yang, Ching Wen; Chai, Jyh Wen; Chung, Yi Nung; Chung, Pau Choo.

Medical Imaging Systems Technology: Volume 4: Methods in Diagnosis Optimization. World Scientific Publishing Co., 2005. p. 297-328.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Techniques for detection of spectral signatures in MR images and their applications

AU - Chen, Clayton Chi Chang

AU - Wang, Chuin Mu

AU - Chang, Chein I.

AU - Yang, Ching Wen

AU - Chai, Jyh Wen

AU - Chung, Yi Nung

AU - Chung, Pau Choo

PY - 2005/1/1

Y1 - 2005/1/1

N2 - This chapter presents two major hyperspectral imaging techniques, orthogonal subspace projection (OSP) and constrained energy minimization (CEM), which have recently found success in spectral signature detection in magnetic resonance (MR) images. Unlike classical image classification techniques which are primarily designed on a pure pixel basis, the proposed OSP and CEM are mixed pixel classification techniques which estimate abundance fractions of different material substances assumed to be present in the image data so as to achieve classification. Three scenarios are considered for MR image classification based on different levels of object information provided a priori. The OSP is applied to the case that the complete prior object knowledge is known compared to the CEM which is used for the case that only desired object knowledge is given while other knowledge can be discarded. When no prior object knowledge is available, the OSP is extended to an unsupervised OSP (UOSP), which obtains the necessary object knowledge directly from the image data in an unsupervised manner for OSP classification. In order to demonstrate the utility of these three scenarios in MR image classification, a series of experiments are conducted and compared to the commonly used c-means method for performance evaluation. The results show that both OSP and CEM are promising and effective spectral techniques for MR image classification.

AB - This chapter presents two major hyperspectral imaging techniques, orthogonal subspace projection (OSP) and constrained energy minimization (CEM), which have recently found success in spectral signature detection in magnetic resonance (MR) images. Unlike classical image classification techniques which are primarily designed on a pure pixel basis, the proposed OSP and CEM are mixed pixel classification techniques which estimate abundance fractions of different material substances assumed to be present in the image data so as to achieve classification. Three scenarios are considered for MR image classification based on different levels of object information provided a priori. The OSP is applied to the case that the complete prior object knowledge is known compared to the CEM which is used for the case that only desired object knowledge is given while other knowledge can be discarded. When no prior object knowledge is available, the OSP is extended to an unsupervised OSP (UOSP), which obtains the necessary object knowledge directly from the image data in an unsupervised manner for OSP classification. In order to demonstrate the utility of these three scenarios in MR image classification, a series of experiments are conducted and compared to the commonly used c-means method for performance evaluation. The results show that both OSP and CEM are promising and effective spectral techniques for MR image classification.

UR - http://www.scopus.com/inward/record.url?scp=84967579129&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84967579129&partnerID=8YFLogxK

U2 - 10.1142/9789812701053_0009

DO - 10.1142/9789812701053_0009

M3 - Chapter

AN - SCOPUS:84967579129

SN - 9812563644

SN - 9789812569905

SP - 297

EP - 328

BT - Medical Imaging Systems Technology

PB - World Scientific Publishing Co.

ER -

Chen CCC, Wang CM, Chang CI, Yang CW, Chai JW, Chung YN et al. Techniques for detection of spectral signatures in MR images and their applications. In Medical Imaging Systems Technology: Volume 4: Methods in Diagnosis Optimization. World Scientific Publishing Co. 2005. p. 297-328 https://doi.org/10.1142/9789812701053_0009