Statistical appearance models of mammographic masses

Michael Berks; Steven Caulkin; Rumana Rahim; Caroline Boggis; Susan Astley

doi:10.1007/978-3-540-70538-3_56

Statistical appearance models of mammographic masses

Michael Berks, Steven Caulkin, Rumana Rahim, Caroline Boggis, Susan Astley

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Abstract

We present a method for building generative statistical appearance models of mammographic masses. We address several key issues that limited the performance of previous methods. In particular, we use MDL optimization to generate more compact shape correspondences; we describe a technique for the accurate estimation of the background tissue on which a mass is superimposed; and we highlight the importance of choosing suitable weighting between shape, texture and scale components in the final combined model. Improvements in the ability of the model to characterize a set of 101 mammographic masses are quantified using leave-one-out testing, showing a reduction in mean square error per pixel from 3.109 using a previous method to 1.262 using the new appearance model. © 2008 Springer-Verlag Berlin Heidelberg.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)\|Lect. Notes Comput. Sci.
Pages	401-408
Number of pages	7
Volume	5116
DOIs	https://doi.org/10.1007/978-3-540-70538-3_56
Publication status	Published - 2008
Event	9th International Workshop on Digital Mammography, IWDM 2008 - Tucson, AZ Duration: 1 Jul 2008 → …

Conference

Conference	9th International Workshop on Digital Mammography, IWDM 2008
City	Tucson, AZ
Period	1/07/08 → …

Keywords

Appearance model
Breast cancer
Breast mass
Mammography
Shape analysis
Texture analysis
X-ray

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-540-70538-3_56

Cite this

@inbook{5cceef23538e44ab860428b57f1a3fc5,

title = "Statistical appearance models of mammographic masses",

abstract = "We present a method for building generative statistical appearance models of mammographic masses. We address several key issues that limited the performance of previous methods. In particular, we use MDL optimization to generate more compact shape correspondences; we describe a technique for the accurate estimation of the background tissue on which a mass is superimposed; and we highlight the importance of choosing suitable weighting between shape, texture and scale components in the final combined model. Improvements in the ability of the model to characterize a set of 101 mammographic masses are quantified using leave-one-out testing, showing a reduction in mean square error per pixel from 3.109 using a previous method to 1.262 using the new appearance model. {\textcopyright} 2008 Springer-Verlag Berlin Heidelberg.",

keywords = "Appearance model, Breast cancer, Breast mass, Mammography, Shape analysis, Texture analysis, X-ray",

author = "Michael Berks and Steven Caulkin and Rumana Rahim and Caroline Boggis and Susan Astley",

year = "2008",

doi = "10.1007/978-3-540-70538-3_56",

language = "English",

isbn = "3540705376",

volume = "5116",

pages = "401--408",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.",

note = "9th International Workshop on Digital Mammography, IWDM 2008 ; Conference date: 01-07-2008",

}

Berks, M, Caulkin, S, Rahim, R, Boggis, C & Astley, S 2008, Statistical appearance models of mammographic masses. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.. vol. 5116, pp. 401-408, 9th International Workshop on Digital Mammography, IWDM 2008, Tucson, AZ, 1/07/08. https://doi.org/10.1007/978-3-540-70538-3_56

TY - CHAP

T1 - Statistical appearance models of mammographic masses

AU - Berks, Michael

AU - Caulkin, Steven

AU - Rahim, Rumana

AU - Boggis, Caroline

AU - Astley, Susan

PY - 2008

Y1 - 2008

N2 - We present a method for building generative statistical appearance models of mammographic masses. We address several key issues that limited the performance of previous methods. In particular, we use MDL optimization to generate more compact shape correspondences; we describe a technique for the accurate estimation of the background tissue on which a mass is superimposed; and we highlight the importance of choosing suitable weighting between shape, texture and scale components in the final combined model. Improvements in the ability of the model to characterize a set of 101 mammographic masses are quantified using leave-one-out testing, showing a reduction in mean square error per pixel from 3.109 using a previous method to 1.262 using the new appearance model. © 2008 Springer-Verlag Berlin Heidelberg.

AB - We present a method for building generative statistical appearance models of mammographic masses. We address several key issues that limited the performance of previous methods. In particular, we use MDL optimization to generate more compact shape correspondences; we describe a technique for the accurate estimation of the background tissue on which a mass is superimposed; and we highlight the importance of choosing suitable weighting between shape, texture and scale components in the final combined model. Improvements in the ability of the model to characterize a set of 101 mammographic masses are quantified using leave-one-out testing, showing a reduction in mean square error per pixel from 3.109 using a previous method to 1.262 using the new appearance model. © 2008 Springer-Verlag Berlin Heidelberg.

KW - Appearance model

KW - Breast cancer

KW - Breast mass

KW - Mammography

KW - Shape analysis

KW - Texture analysis

KW - X-ray

U2 - 10.1007/978-3-540-70538-3_56

DO - 10.1007/978-3-540-70538-3_56

M3 - Chapter

SN - 3540705376

SN - 9783540705376

VL - 5116

SP - 401

EP - 408

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.

T2 - 9th International Workshop on Digital Mammography, IWDM 2008

Y2 - 1 July 2008

ER -

Statistical appearance models of mammographic masses

Abstract

Conference

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this