A new divergence measure for medical image registration

S. Martin, T.S. Durrani

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

A new type of divergence measure for the registration of medical images is introduced that exploits the properties of the modified Bessel functions of the second kind. The properties of the proposed divergence coefficient are analysed and compared with those of the classic measures, including Kullback-Leibler, Renyi, and Ialpha divergences. To ensure its effectiveness and widespread applicability to any arbitrary set of data types, the performance of the new measure is analysed for Gaussian, exponential, and other advanced probability density functions. The results verify its robustness. Finally, the new divergence measure is used in the registration of CT to MR medical images to validate the improvement in registration accuracy
Original languageEnglish
Pages (from-to)957-966
Number of pages9
JournalIEEE Transactions on Image Processing
Volume16
Issue number4
DOIs
Publication statusPublished - Apr 2007

Fingerprint

Bessel functions
Image registration
Probability density function

Keywords

  • divergence measure
  • image registration
  • modifiedBessel function

Cite this

@article{a606479909a045ab8852db026b3a6814,
title = "A new divergence measure for medical image registration",
abstract = "A new type of divergence measure for the registration of medical images is introduced that exploits the properties of the modified Bessel functions of the second kind. The properties of the proposed divergence coefficient are analysed and compared with those of the classic measures, including Kullback-Leibler, Renyi, and Ialpha divergences. To ensure its effectiveness and widespread applicability to any arbitrary set of data types, the performance of the new measure is analysed for Gaussian, exponential, and other advanced probability density functions. The results verify its robustness. Finally, the new divergence measure is used in the registration of CT to MR medical images to validate the improvement in registration accuracy",
keywords = "divergence measure, image registration, modifiedBessel function",
author = "S. Martin and T.S. Durrani",
year = "2007",
month = "4",
doi = "10.1109/TIP.2007.891772",
language = "English",
volume = "16",
pages = "957--966",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
number = "4",

}

A new divergence measure for medical image registration. / Martin, S.; Durrani, T.S.

In: IEEE Transactions on Image Processing, Vol. 16, No. 4, 04.2007, p. 957-966.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A new divergence measure for medical image registration

AU - Martin, S.

AU - Durrani, T.S.

PY - 2007/4

Y1 - 2007/4

N2 - A new type of divergence measure for the registration of medical images is introduced that exploits the properties of the modified Bessel functions of the second kind. The properties of the proposed divergence coefficient are analysed and compared with those of the classic measures, including Kullback-Leibler, Renyi, and Ialpha divergences. To ensure its effectiveness and widespread applicability to any arbitrary set of data types, the performance of the new measure is analysed for Gaussian, exponential, and other advanced probability density functions. The results verify its robustness. Finally, the new divergence measure is used in the registration of CT to MR medical images to validate the improvement in registration accuracy

AB - A new type of divergence measure for the registration of medical images is introduced that exploits the properties of the modified Bessel functions of the second kind. The properties of the proposed divergence coefficient are analysed and compared with those of the classic measures, including Kullback-Leibler, Renyi, and Ialpha divergences. To ensure its effectiveness and widespread applicability to any arbitrary set of data types, the performance of the new measure is analysed for Gaussian, exponential, and other advanced probability density functions. The results verify its robustness. Finally, the new divergence measure is used in the registration of CT to MR medical images to validate the improvement in registration accuracy

KW - divergence measure

KW - image registration

KW - modifiedBessel function

UR - http://ieeexplore.ieee.org/iel5/83/4130412/04130432.pdf?tp=&arnumber=4130432&isnumber=4130412

UR - http://dx.doi.org/10.1109/TIP.2007.891772

U2 - 10.1109/TIP.2007.891772

DO - 10.1109/TIP.2007.891772

M3 - Article

VL - 16

SP - 957

EP - 966

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 4

ER -