Toward expert knowledge representation for automatic breast cancer detection

Marina Velikova; Maurice Samulski; Nico Karssemeijer; Peter Lucas

doi:10.1007/978-3-540-85776-1_28

Toward expert knowledge representation for automatic breast cancer detection

Marina Velikova^*, Maurice Samulski, Nico Karssemeijer, Peter Lucas

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

3 Citations (Scopus)

Abstract

In reading mammograms, radiologists judge for the presence of a lesion by comparing at least two breast projections (views) as a lesion is to be observed in both of them. Most computer-aided detection (CAD) systems, on the other hand, treat single views independently and thus they fail to account for the interaction between the breast views. Following the radiologist's practice, in this paper, we develop a Bayesian network framework for automatic multi-view mammographic analysis based on causal independence models and the regions detected as suspicious by a single-view CAD system. We have implemented two versions of the framework based on different definitions of multi-view correspondences. The proposed approach is evaluated and compared against the single-view CAD system in an experimental study with real-life data. The results show that using expert knowledge helps to increase the cancer detection rate at a patient level.

Original language	English
Title of host publication	Artificial Intelligence
Subtitle of host publication	Methodology, Systems, and Applications - 13th International Conference, AIMSA 2008, Varna, Bulgaria, September 4-6, 2008. Proceedings
Editors	Danail Dochev, Marco Pistore, Paolo Traverso
Publisher	Springer
Pages	333-344
Number of pages	12
ISBN (Print)	3-540-85775-3, 978-3-540-85775-4
DOIs	https://doi.org/10.1007/978-3-540-85776-1_28
Publication status	Published - 2008
Externally published	Yes
Event	13th International Conference on Artificial Intelligence: Methodology, Systems, and Applications, AIMSA 2008 - Varna, Bulgaria Duration: 4 Sept 2008 → 6 Sept 2008 Conference number: 13

Publication series

Name	Lecture Notes in Artificial Intelligence
Publisher	Springer
Volume	5253
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	13th International Conference on Artificial Intelligence: Methodology, Systems, and Applications, AIMSA 2008
Abbreviated title	AIMSA
Country/Territory	Bulgaria
City	Varna
Period	4/09/08 → 6/09/08

Keywords

n/a OA procedure
Causal independence model
Mammography
Multi-view breast cancer detection
Bayesian network

UN SDGs

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

Access to Document

10.1007/978-3-540-85776-1_28

Cite this

Velikova, M., Samulski, M., Karssemeijer, N., & Lucas, P. (2008). Toward expert knowledge representation for automatic breast cancer detection. In D. Dochev, M. Pistore, & P. Traverso (Eds.), Artificial Intelligence: Methodology, Systems, and Applications - 13th International Conference, AIMSA 2008, Varna, Bulgaria, September 4-6, 2008. Proceedings (pp. 333-344). (Lecture Notes in Artificial Intelligence; Vol. 5253). Springer. https://doi.org/10.1007/978-3-540-85776-1_28

Velikova, Marina ; Samulski, Maurice ; Karssemeijer, Nico et al. / Toward expert knowledge representation for automatic breast cancer detection. Artificial Intelligence: Methodology, Systems, and Applications - 13th International Conference, AIMSA 2008, Varna, Bulgaria, September 4-6, 2008. Proceedings. editor / Danail Dochev ; Marco Pistore ; Paolo Traverso. Springer, 2008. pp. 333-344 (Lecture Notes in Artificial Intelligence).

@inproceedings{65c3cdf66a0c4740afdf75d99d02fd26,

title = "Toward expert knowledge representation for automatic breast cancer detection",

abstract = "In reading mammograms, radiologists judge for the presence of a lesion by comparing at least two breast projections (views) as a lesion is to be observed in both of them. Most computer-aided detection (CAD) systems, on the other hand, treat single views independently and thus they fail to account for the interaction between the breast views. Following the radiologist's practice, in this paper, we develop a Bayesian network framework for automatic multi-view mammographic analysis based on causal independence models and the regions detected as suspicious by a single-view CAD system. We have implemented two versions of the framework based on different definitions of multi-view correspondences. The proposed approach is evaluated and compared against the single-view CAD system in an experimental study with real-life data. The results show that using expert knowledge helps to increase the cancer detection rate at a patient level.",

keywords = "n/a OA procedure, Causal independence model, Mammography, Multi-view breast cancer detection, Bayesian network",

author = "Marina Velikova and Maurice Samulski and Nico Karssemeijer and Peter Lucas",

year = "2008",

doi = "10.1007/978-3-540-85776-1_28",

language = "English",

isbn = "3-540-85775-3",

series = "Lecture Notes in Artificial Intelligence",

publisher = "Springer",

pages = "333--344",

editor = "Danail Dochev and Marco Pistore and Paolo Traverso",

booktitle = "Artificial Intelligence",

address = "Germany",

note = "13th International Conference on Artificial Intelligence: Methodology, Systems, and Applications, AIMSA 2008, AIMSA ; Conference date: 04-09-2008 Through 06-09-2008",

}

Velikova, M, Samulski, M, Karssemeijer, N & Lucas, P 2008, Toward expert knowledge representation for automatic breast cancer detection. in D Dochev, M Pistore & P Traverso (eds), Artificial Intelligence: Methodology, Systems, and Applications - 13th International Conference, AIMSA 2008, Varna, Bulgaria, September 4-6, 2008. Proceedings. Lecture Notes in Artificial Intelligence, vol. 5253, Springer, pp. 333-344, 13th International Conference on Artificial Intelligence: Methodology, Systems, and Applications, AIMSA 2008, Varna, Bulgaria, 4/09/08. https://doi.org/10.1007/978-3-540-85776-1_28

Toward expert knowledge representation for automatic breast cancer detection. / Velikova, Marina; Samulski, Maurice; Karssemeijer, Nico et al.
Artificial Intelligence: Methodology, Systems, and Applications - 13th International Conference, AIMSA 2008, Varna, Bulgaria, September 4-6, 2008. Proceedings. ed. / Danail Dochev; Marco Pistore; Paolo Traverso. Springer, 2008. p. 333-344 (Lecture Notes in Artificial Intelligence; Vol. 5253).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Toward expert knowledge representation for automatic breast cancer detection

AU - Velikova, Marina

AU - Samulski, Maurice

AU - Karssemeijer, Nico

AU - Lucas, Peter

N1 - Conference code: 13

PY - 2008

Y1 - 2008

N2 - In reading mammograms, radiologists judge for the presence of a lesion by comparing at least two breast projections (views) as a lesion is to be observed in both of them. Most computer-aided detection (CAD) systems, on the other hand, treat single views independently and thus they fail to account for the interaction between the breast views. Following the radiologist's practice, in this paper, we develop a Bayesian network framework for automatic multi-view mammographic analysis based on causal independence models and the regions detected as suspicious by a single-view CAD system. We have implemented two versions of the framework based on different definitions of multi-view correspondences. The proposed approach is evaluated and compared against the single-view CAD system in an experimental study with real-life data. The results show that using expert knowledge helps to increase the cancer detection rate at a patient level.

AB - In reading mammograms, radiologists judge for the presence of a lesion by comparing at least two breast projections (views) as a lesion is to be observed in both of them. Most computer-aided detection (CAD) systems, on the other hand, treat single views independently and thus they fail to account for the interaction between the breast views. Following the radiologist's practice, in this paper, we develop a Bayesian network framework for automatic multi-view mammographic analysis based on causal independence models and the regions detected as suspicious by a single-view CAD system. We have implemented two versions of the framework based on different definitions of multi-view correspondences. The proposed approach is evaluated and compared against the single-view CAD system in an experimental study with real-life data. The results show that using expert knowledge helps to increase the cancer detection rate at a patient level.

KW - n/a OA procedure

KW - Causal independence model

KW - Mammography

KW - Multi-view breast cancer detection

KW - Bayesian network

UR - http://www.scopus.com/inward/record.url?scp=52149120077&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-85776-1_28

DO - 10.1007/978-3-540-85776-1_28

M3 - Conference contribution

AN - SCOPUS:52149120077

SN - 3-540-85775-3

SN - 978-3-540-85775-4

T3 - Lecture Notes in Artificial Intelligence

SP - 333

EP - 344

BT - Artificial Intelligence

A2 - Dochev, Danail

A2 - Pistore, Marco

A2 - Traverso, Paolo

PB - Springer

T2 - 13th International Conference on Artificial Intelligence: Methodology, Systems, and Applications, AIMSA 2008

Y2 - 4 September 2008 through 6 September 2008

ER -

Velikova M, Samulski M, Karssemeijer N, Lucas P. Toward expert knowledge representation for automatic breast cancer detection. In Dochev D, Pistore M, Traverso P, editors, Artificial Intelligence: Methodology, Systems, and Applications - 13th International Conference, AIMSA 2008, Varna, Bulgaria, September 4-6, 2008. Proceedings. Springer. 2008. p. 333-344. (Lecture Notes in Artificial Intelligence). doi: 10.1007/978-3-540-85776-1_28

Toward expert knowledge representation for automatic breast cancer detection

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this