Quantitative analysis of breast DCE-MR images based on ICA and an empirical model

Sebastian Goebl; Claudia Plant; Marc Lobbes M.D.; Anke Meyer-Bäse

doi:10.1117/12.921952

10 May 2012 Quantitative analysis of breast DCE-MR images based on ICA and an empirical model

Sebastian Goebl, Claudia Plant, Marc Lobbes M.D., Anke Meyer-Bäse

Proceedings Volume 8401, Independent Component Analyses, Compressive Sampling, Wavelets, Neural Net, Biosystems, and Nanoengineering X; 84010Z (2012) https://doi.org/10.1117/12.921952
Event: SPIE Defense, Security, and Sensing, 2012, Baltimore, Maryland, United States

Abstract

DCE-MRI represents an important tool for detecting subtle kinetic changes in breast lesion tissue. Non-masslike breast lesions exhibit an atypical dynamical behavior compared to mass-like lesions and pose a challenge to a computer-aided diagnosis system. Yet the correct diagnosis of these tumors represents an important step towards early prevention. We apply Independent Component Analysis (ICA) on DCE-MRI images to extract kinetic tumor curves. We use a known empirical mathematical model to automatically identify the tumor curves from the ICA result. Filtering out noise, our technique is superior to traditional ROI-based analysis in capturing the kinetic characteristics of the tumor curves. These typical characteristics enable us to nd out the optimal number of independent components for ICA. Another benet of our method is the segmentation of tumor tissue which is superior to the segmentation from MR subtraction images. Our aim is a optimal extraction of tumor curves to provide a better basis for kinetic analysis and to distinguish between benign and malignant lesions, especially for the challenging non-mass-like breast lesions.

Citation Download Citation

Sebastian Goebl, Claudia Plant, Marc Lobbes M.D., and Anke Meyer-Bäse "Quantitative analysis of breast DCE-MR images based on ICA and an empirical model", Proc. SPIE 8401, Independent Component Analyses, Compressive Sampling, Wavelets, Neural Net, Biosystems, and Nanoengineering X, 84010Z (10 May 2012); https://doi.org/10.1117/12.921952

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available