Grant: Engage Grants Program
Simon Fraser University
$16,250.00
Enhanced algorithms for mammography image segmentation
EF 2010-11
Résumé
Breast cancer is a leading cause of death for women and digital mammography screening is a widely accepted
procedure for identifying this cancer at its early development stage. Modern imaging centers use digital
mammography devices to acquire images as part of screening procedures. The images are send to review
workstations where radiologists review the images by using visualization software connected to high resolution
5 mega pixel (MP) monitors.
The spatial resolution of mammography images is usually very high and usually bigger than the resolution of
the 5 MP monitors. Each mammography study includes a number of images and certain rules govern the
presentation of the images to the radiologist. Radiologists are required to review the images in their full
acquired resolution, and since the images do not fit into a single monitor radiologists have to pan the images to
make sure all portions of the image are observed/reviewed. Such manual user interaction is inconvenient and is
considered as a problem with negative performance on reading workflow and patient care.
A general solution to the above problem is based on automatic detection of the breast in the mammography
images, requiring ways to identify the existence of breast soft tissue in a given region of interest. Usually about
half of a mammography image consists of background which can be excluded from the diagnostic analysis. A
meaningful automatic detection of the breast tissue of interest will therefore allow a big image to be divided
into regions with clinical information so that only clinical relevant parts of the image are displayed in full
resolution (while possibly background is excluded).
In addition, automatic detection allows the viewer application to position text overlay and orientation
information at appropriate locations so that there is no interference with radiologists' reading.
This software will increase radiologists' efficiency and may lead to improved patient diagnosis.
Programme
Engage Grants Program
Bénéficiaire
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
403154-2010
Name nom
Atkins, Stella
Department département
Computing Science
Organization id
5
Institution établissement
Simon Fraser University
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2010
Competition year année de concours
2010
Award amount
16250
Program id
EGP
Program name en
Engage Grants Program
Program name fr
Programme de subventions d'engagement partenariat
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1554
Committee name en
Pacific Internal Decision Committee
Committee name fr
Comité de décision interne pour le Pacifique
Area of application code
518
Area of application group en
Manufacturing processes and products
Area of application group fr
Procédés et produits industriels
Area of application en
Medical equipment and apparatus
Area of application fr
Appareils et instruments médicaux
Research subject code
2705
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Software and development
Research subject fr
Logiciels et développement
Partie
1
Nb partie
1
Application title
Enhanced algorithms for mammography image segmentation
Keyword
NSERC
Application summary
Breast cancer is a leading cause of death for women and digital mammography screening is a widely accepted
procedure for identifying this cancer at its early development stage. Modern imaging centers use digital
mammography devices to acquire images as part of screening procedures. The images are send to review
workstations where radiologists review the images by using visualization software connected to high resolution
5 mega pixel (MP) monitors.
The spatial resolution of mammography images is usually very high and usually bigger than the resolution of
the 5 MP monitors. Each mammography study includes a number of images and certain rules govern the
presentation of the images to the radiologist. Radiologists are required to review the images in their full
acquired resolution, and since the images do not fit into a single monitor radiologists have to pan the images to
make sure all portions of the image are observed/reviewed. Such manual user interaction is inconvenient and is
considered as a problem with negative performance on reading workflow and patient care.
A general solution to the above problem is based on automatic detection of the breast in the mammography
images, requiring ways to identify the existence of breast soft tissue in a given region of interest. Usually about
half of a mammography image consists of background which can be excluded from the diagnostic analysis. A
meaningful automatic detection of the breast tissue of interest will therefore allow a big image to be divided
into regions with clinical information so that only clinical relevant parts of the image are displayed in full
resolution (while possibly background is excluded).
In addition, automatic detection allows the viewer application to position text overlay and orientation
information at appropriate locations so that there is no interference with radiologists' reading.
This software will increase radiologists' efficiency and may lead to improved patient diagnosis.