EM Texture Segmentation of Mammographic Images

Reyer Zwiggelaar; Pere Planiol; Joan Marti; Robert Marti; Lilian Blot; Erika R. E. Denton; Caroline M. E. Rubin

doi:10.1007/978-3-642-59327-7_53

EM Texture Segmentation of Mammographic Images

Reyer Zwiggelaar, Pere Planiol, Joan Marti, Robert Marti, Lilian Blot, Erika R. E. Denton, Caroline M. E. Rubin

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

Abstract

We have investigated a combination of statistical modelling and expectation maximisation for a texture based approach to the segmentation of mammographic images. Texture modelling is based on the implicit incorporation of spatial information. Statistical modelling is used for data generalisation and noise removal purposes. Expectation maximisation modelling of the spatial information in combination with the statistical modelling are compared to expectation maximisation modelling based on single grey-level values. The developed segmentation results are used for two specific applications which are registration of mammographic images and mammographic risk assessment.

Original language	English
Title of host publication	Digital Mammography
Subtitle of host publication	IWDM 2002 — 6th International Workshop on Digital Mammography
Editors	Hainz-Otto Peitgen
Pages	223-227
Number of pages	5
ISBN (Electronic)	978-3-642-59327-7
DOIs	https://doi.org/10.1007/978-3-642-59327-7_53
Publication status	Published - 27 Sept 2011
Externally published	Yes
Event	6th International Workshop on Digital Mammography - Bremen, Germany Duration: 22 Jun 2002 → 25 Jun 2002

Workshop

Workshop	6th International Workshop on Digital Mammography
Abbreviated title	IWDM 2002
Country/Territory	Germany
City	Bremen
Period	22 Jun 2002 → 25 Jun 2002

Access to Document

10.1007/978-3-642-59327-7_53

Permanent link

Persistent link

Cite this

@inproceedings{dacacde92b904c8bba18d60ced8b68b5,

title = "EM Texture Segmentation of Mammographic Images",

abstract = "We have investigated a combination of statistical modelling and expectation maximisation for a texture based approach to the segmentation of mammographic images. Texture modelling is based on the implicit incorporation of spatial information. Statistical modelling is used for data generalisation and noise removal purposes. Expectation maximisation modelling of the spatial information in combination with the statistical modelling are compared to expectation maximisation modelling based on single grey-level values. The developed segmentation results are used for two specific applications which are registration of mammographic images and mammographic risk assessment.",

author = "Reyer Zwiggelaar and Pere Planiol and Joan Marti and Robert Marti and Lilian Blot and Denton, {Erika R. E.} and Rubin, {Caroline M. E.}",

year = "2011",

month = sep,

day = "27",

doi = "10.1007/978-3-642-59327-7_53",

language = "English",

isbn = "978-3-642-63936-4",

pages = "223--227",

editor = "Hainz-Otto Peitgen",

booktitle = "Digital Mammography",

note = "6th International Workshop on Digital Mammography, IWDM 2002 ; Conference date: 22-06-2002 Through 25-06-2002",

}

Zwiggelaar, R, Planiol, P, Marti, J, Marti, R, Blot, L, Denton, ERE & Rubin, CME 2011, EM Texture Segmentation of Mammographic Images. in H-O Peitgen (ed.), Digital Mammography: IWDM 2002 — 6th International Workshop on Digital Mammography. pp. 223-227, 6th International Workshop on Digital Mammography, Bremen, Germany, 22 Jun 2002. https://doi.org/10.1007/978-3-642-59327-7_53

TY - GEN

T1 - EM Texture Segmentation of Mammographic Images

AU - Zwiggelaar, Reyer

AU - Planiol, Pere

AU - Marti, Joan

AU - Marti, Robert

AU - Blot, Lilian

AU - Denton, Erika R. E.

AU - Rubin, Caroline M. E.

PY - 2011/9/27

Y1 - 2011/9/27

N2 - We have investigated a combination of statistical modelling and expectation maximisation for a texture based approach to the segmentation of mammographic images. Texture modelling is based on the implicit incorporation of spatial information. Statistical modelling is used for data generalisation and noise removal purposes. Expectation maximisation modelling of the spatial information in combination with the statistical modelling are compared to expectation maximisation modelling based on single grey-level values. The developed segmentation results are used for two specific applications which are registration of mammographic images and mammographic risk assessment.

AB - We have investigated a combination of statistical modelling and expectation maximisation for a texture based approach to the segmentation of mammographic images. Texture modelling is based on the implicit incorporation of spatial information. Statistical modelling is used for data generalisation and noise removal purposes. Expectation maximisation modelling of the spatial information in combination with the statistical modelling are compared to expectation maximisation modelling based on single grey-level values. The developed segmentation results are used for two specific applications which are registration of mammographic images and mammographic risk assessment.

U2 - 10.1007/978-3-642-59327-7_53

DO - 10.1007/978-3-642-59327-7_53

M3 - Conference Proceeding (Non-Journal item)

SN - 978-3-642-63936-4

SP - 223

EP - 227

BT - Digital Mammography

A2 - Peitgen, Hainz-Otto

T2 - 6th International Workshop on Digital Mammography

Y2 - 22 June 2002 through 25 June 2002

ER -

EM Texture Segmentation of Mammographic Images

Abstract

Workshop

Access to Document

Permanent link

Fingerprint

Cite this