Robustness of two methods for segmenting salient features in screening mammograms

Fei Ma; Mariusz Bajger; Murk Bottema

doi:10.1109/DICTA.2007.4426784

Robustness of two methods for segmenting salient features in screening mammograms

Fei Ma, Mariusz Bajger, Murk Bottema

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

7 Citations (Scopus)

Abstract

The performance of two image segmentation methods are compared according to robustness of the segmentation to image distortion. This criterion is crucial for temporal analysis of screening mammograms where natural changes in the breast plus inherent deformation of soft tissue during image acquisition result in severe image registration problems. A method based on minimum spanning trees (MST) is found to be more robust to the distortions studied than a method based on adaptive pyramids (AP). Although segmentation leads to great differences in segmentation in distorted images for many components of low saliency, salient components (those of primary interest) are found to be segmented consistently regardless of distortion.

Original language	English
Title of host publication	Proceedings
Subtitle of host publication	Digital Image Computing Techniques and Applications 9th Biennial Conference of the Australian Pattern Recognition Society
Place of Publication	Piscataway, NJ
Publisher	Institute of Electrical and Electronics Engineers
Pages	112-117
Number of pages	6
ISBN (Print)	0-7695-3067-2 , 978-0-7695-3067-3
DOIs	https://doi.org/10.1109/DICTA.2007.4426784
Publication status	Published - 3 Dec 2007
Event	9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications (DICTA) - Adelaide, Australia Duration: 3 Dec 2007 → 5 Dec 2007 Conference number: 9th

Conference

Conference	9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications (DICTA)
Abbreviated title	DICTA 2007
Country/Territory	Australia
City	Adelaide
Period	3/12/07 → 5/12/07

Access to Document

10.1109/DICTA.2007.4426784Licence: In Copyright

Cite this

Ma, F., Bajger, M., & Bottema, M. (2007). Robustness of two methods for segmenting salient features in screening mammograms. In Proceedings: Digital Image Computing Techniques and Applications 9th Biennial Conference of the Australian Pattern Recognition Society (pp. 112-117). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/DICTA.2007.4426784

@inproceedings{bfcb7cc057264c9fbb6a9323d8e9fd56,

title = "Robustness of two methods for segmenting salient features in screening mammograms",

abstract = "The performance of two image segmentation methods are compared according to robustness of the segmentation to image distortion. This criterion is crucial for temporal analysis of screening mammograms where natural changes in the breast plus inherent deformation of soft tissue during image acquisition result in severe image registration problems. A method based on minimum spanning trees (MST) is found to be more robust to the distortions studied than a method based on adaptive pyramids (AP). Although segmentation leads to great differences in segmentation in distorted images for many components of low saliency, salient components (those of primary interest) are found to be segmented consistently regardless of distortion.",

author = "Fei Ma and Mariusz Bajger and Murk Bottema",

year = "2007",

month = dec,

day = "3",

doi = "10.1109/DICTA.2007.4426784",

language = "English",

isbn = "0-7695-3067-2 ",

pages = "112--117",

booktitle = "Proceedings",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications (DICTA), DICTA 2007 ; Conference date: 03-12-2007 Through 05-12-2007",

}

Ma, F, Bajger, M & Bottema, M 2007, Robustness of two methods for segmenting salient features in screening mammograms. in Proceedings: Digital Image Computing Techniques and Applications 9th Biennial Conference of the Australian Pattern Recognition Society . Institute of Electrical and Electronics Engineers, Piscataway, NJ , pp. 112-117, 9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications (DICTA), Adelaide, South Australia, Australia, 3/12/07. https://doi.org/10.1109/DICTA.2007.4426784

Robustness of two methods for segmenting salient features in screening mammograms. / Ma, Fei; Bajger, Mariusz ; Bottema, Murk.
Proceedings: Digital Image Computing Techniques and Applications 9th Biennial Conference of the Australian Pattern Recognition Society . Piscataway, NJ : Institute of Electrical and Electronics Engineers, 2007. p. 112-117.

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

TY - GEN

T1 - Robustness of two methods for segmenting salient features in screening mammograms

AU - Ma, Fei

AU - Bajger, Mariusz

AU - Bottema, Murk

N1 - Conference code: 9th

PY - 2007/12/3

Y1 - 2007/12/3

N2 - The performance of two image segmentation methods are compared according to robustness of the segmentation to image distortion. This criterion is crucial for temporal analysis of screening mammograms where natural changes in the breast plus inherent deformation of soft tissue during image acquisition result in severe image registration problems. A method based on minimum spanning trees (MST) is found to be more robust to the distortions studied than a method based on adaptive pyramids (AP). Although segmentation leads to great differences in segmentation in distorted images for many components of low saliency, salient components (those of primary interest) are found to be segmented consistently regardless of distortion.

AB - The performance of two image segmentation methods are compared according to robustness of the segmentation to image distortion. This criterion is crucial for temporal analysis of screening mammograms where natural changes in the breast plus inherent deformation of soft tissue during image acquisition result in severe image registration problems. A method based on minimum spanning trees (MST) is found to be more robust to the distortions studied than a method based on adaptive pyramids (AP). Although segmentation leads to great differences in segmentation in distorted images for many components of low saliency, salient components (those of primary interest) are found to be segmented consistently regardless of distortion.

U2 - 10.1109/DICTA.2007.4426784

DO - 10.1109/DICTA.2007.4426784

M3 - Conference contribution

SN - 0-7695-3067-2

SN - 978-0-7695-3067-3

SP - 112

EP - 117

BT - Proceedings

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway, NJ

T2 - 9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications (DICTA)

Y2 - 3 December 2007 through 5 December 2007

ER -

Robustness of two methods for segmenting salient features in screening mammograms

Abstract

Conference

Access to Document

Fingerprint

Cite this