Project

Back to overview

Role of tumor shape and background in breast imaging: from digital mammography to breast tomosynthesis

English title Role of tumor shape and background in breast imaging: from digital mammography to breast tomosynthesis
Applicant Bochud François
Number 113863
Funding scheme Project funding (Div. I-III)
Research institution Institut de Radiophysique Département de Radiologie Université de Lausanne/CHUV
Institution of higher education University of Lausanne - LA
Main discipline Clinical Cancer Research
Start/End 01.04.2007 - 31.03.2009
Approved amount 97'268.00
Show all

Keywords (5)

mammography; image texture; model observers; psychophysical experiments; tomosynthesis

Lay Summary (English)

Lead
Lay summary
This study is aimed at continuing grant FN3252BO-104273, still in progress at the time of writing, and which has been centered on the classification and the statistical characterization of digital mammograms, the generation of realistic textures reproducing their visual and mathematical aspects, and the determination of statistical features, computed from the anatomical noise, influencing human observer performance in the detection task.

In the present study, we propose to characterize the role of signal variations in a detection task on realistic digital mammographic background, and to objectively study and predict the input of breast tomosynthesis technique applied to this task. These aims could be reached in three steps.

First, we will study the ability of existing model observers to predict human performance in detecting a synthetic signal known exactly -a projected sphere- on realistic background. Additionally, we will try to develop a new model observer based on statistical properties of the background influencing the most human observers. We will be using the results of psychophysical experiments currently in progress in the framework of our previously cited SNF grant.

Second, we want to quantify the changes when model and human observers are confronted to varying signal parameters. Signal modifications will include shape changes, tumor type changes, and contrast variation between the background and the signal to be detected.

Finally, we propose to objectively compare digital mammography to breast tomosynthesis through model observers performances in the same signal known exactly experiment as in the first part of the project. Model observers would provide an unbiased tool for optimizing breast tomosynthesis acquisition and reconstruction parameters.

This work would be mainly conducted in Lausanne at the Institut Universitaire de Radiophysique Appliquée. However, it would greatly benefit from the close collaboration with the Vision & Image Understanding Laboratory of University of California Santa Barbara. From the medical point of view, and because this work would be centered on mammography, the collaboration of Karen Kinkel, MD from Geneva University Hospital and the Clinique des Grangettes would guarantee to stay close to the real practice.
Direct link to Lay Summary Last update: 21.02.2013

Responsible applicant and co-applicants

Employees

Associated projects

Number Title Start Funding scheme
104273 Breast texture synthesis and estimation of the role of the anatomy in the radiological detection process 01.08.2004 Project funding (Div. I-III)

-