RSNA 2012 

Abstract Archives of the RSNA, 2012


LL-INS-TU2B

Semi-automatic, User-driven Breast Fibroglandular Tissue Segmentation, Using Combined Breast MRI and Fat-saturated Breast MRI Image Sets

Scientific Informal (Poster) Presentations

Presented on November 27, 2012
Presented as part of LL-INS-TU: Informatics Lunch Hour CME Posters  

Participants

Ya Wang, Presenter: Nothing to Disclose
Valencia King MD, Abstract Co-Author: Nothing to Disclose
Malcolm Pike, Abstract Co-Author: Nothing to Disclose
Jennifer Brooks PhD, Abstract Co-Author: Nothing to Disclose
Elizabeth A. Morris MD, Abstract Co-Author: Nothing to Disclose
Eve Burstein, Abstract Co-Author: Nothing to Disclose
Joseph Owen Deasy PhD, Abstract Co-Author: Nothing to Disclose

CONCLUSION

We have developed a semi-automatic human-guided breast segmentation method for MRI based on dual MRI and fat-saturated MRI image sets. Although other image information (local texture analysis, etc.) might be usefully added, we believe this is a reliable and robust starting point for FGT segmentation.

BACKGROUND

In breast cancer MRI screening with a contrast agent, the volume of breast tissue that increases in intensity, and the signal intensity increase, are referred to as background parenchymal enhancement (BPE). Recent results indicate that BPE is predictive of future breast cancer risk (King et al., Radiology (2011) 260:50-60), but reported a dependence on reader classification of enhancement. In addition, the volume of fibroglandular tissue (FGT) has also been implicated in breast cancer risk. Future studies to further refine these relationships, and potential clinical applications, require reliable auto- or semi-automatic segmentation of FGT, which is currently an unsolved problem that we address.

DISCUSSION

Missing

EVALUATION

We have developed a semi-automatic human-guided breast segmentation method for MRI based on dual MRI and fat-saturated MRI image sets. Although other image information (local texture analysis, etc.) might be usefully added, we believe this is a reliable and robust starting point for FGT segmentation.

Cite This Abstract

Wang, Y, King, V, Pike, M, Brooks, J, Morris, E, Burstein, E, Deasy, J, Semi-automatic, User-driven Breast Fibroglandular Tissue Segmentation, Using Combined Breast MRI and Fat-saturated Breast MRI Image Sets.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12028927.html