Automated percentage of breast density measurements for full field digital mammography

10497,117

15/988143

May 24, 2018

An automated percentage of breast density measure (PDa) that analyzes signal dependent noise (SDN) based on a wavelet expansion using full field digital mammography (FFDM). A matched case-control dataset is used with images acquired from a specific direct x-ray capture FFDM system. PDa is applied to the raw and clinical display representation images. Transforming (pixel mapping) of the raw image to another representation (raw-transformed) is performed using differential evolution optimization and investigated the influence of lowering the native spatial resolution of the image by a one-half. When controlling for body mass index, the quartile specific ORs for the associations of PDa with breast cancer varied with representation and resolution. PDa is a valid automated breast density measurement for a specific FFDM technology and compares well against PD (operator-assisted or the standard) when applied to either the raw-transformed or clinical display images from this FFDM technology.

H. LEE MOFFITT CANCER CENTER AND RESEARCH INSTITUTE, INC. (Tampa, FL, US); MAYO FOUNDATION FOR MEDICAL EDUCATION AND RESEARCH (Rochester, MN, US)

H. Lee Moffitt Cancer Center & Research Institute, Inc. (Tampa, FL, US), Mayo Foundation for Medical Education and Research (Rochester, MN, US)

1. A computer implemented method of determining breast density, comprising: receiving digital image data including a plurality of pixels; filtering the digital image data; estimating a global reference variance signal in the digital image data; performing a statistical analysis on the digital image data to determine areas of higher density tissue in the digital image data by: applying a wavelet expansion to the digital image data to produce a multiresolutional representation comprising difference images (di images) and a low-resolution image; estimating a global reference variance signal from all pixels of a di image that corresponds to a breast region of the areas of higher spatial variation having signal dependent noise (SDN) signatures; determining a local variance using a moving search window, and if the local variance for each search window location deviates from the global reference, labeling a respective search window location with a first designation, otherwise labeling the respective search window location with a second designation; and determining that the locations labeled with the second designation are adipose image regions to provide the measure of risk for breast cancer; and associating the statistically analyzed digital image data with a measure of risk for breast cancer.

2. The computer implemented method of claim 1 , wherein the first indication represents a dense breast tissue region and where in the second indication represents a breast tissue region having other characteristics than dense.

2009/0041373 February 2009 Incesu
2011/0013819 January 2011 Raundahl
2014/0219552 August 2014 Porikli
WO-2012082994 June 2012

edspgr.10497117