Simultaneous Estimation and Segmentation of T1 Map for Breast Parenchyma Measurement

Abstract

Breast density has been shown to be an independent risk factor for breast cancer. In order to segment breast parenchyma, which has been proposed as a biomarker of breast cancer risk, we present an integrated algorithm for simultaneous T1 map estimation and segmentation, using a series of magnetic resonance (MR) breast images. The advantage of using this algorithm is that the step of T1 map estimation (E-Step) and the step of T1 map based tissue segmentation (S-Step) can benefit each other. Since the estimated T1 map can be noisy due to the complexity of T1 estimation method, the tentative tissue segmentation results from S-Step can help perform the edge-preserving smoothing on the estimated T1 map in E-Step, thus removing noises and also preserving tissue boundaries. On the other hand, the improved estimation of T1 map from E-Step can help segment breast tissues in a more accurate and less noisy way. Therefore, by repeating these steps, we can simultaneously obtain better results for both T1 map estimation and segmentation. Experimental results show the effectiveness of the proposed algorithm in breast tissue segmentation and parenchyma volume measurement. Keywords ABSTRACTBreast density has been shown to be an independent risk factor for breast cancer. In order to segment breast parenchyma, which has been proposed as a biomarker of breast cancer risk, we present an integrated algorithm for simultaneous T 1 map estimation and segmentation, using a series of magnetic resonance (MR) breast images. The advantage of using this algorithm is that the step of T 1 map estimation (E-Step) and the step of T 1 map based tissue segmentation (S-Step) can benefit each other. Since the estimated T 1 map can be noisy due to the complexity of T 1 estimation method, the tentative tissue segmentation results from S-Step can help perform the edge-preserving smoothing on the estimated T 1 map in E-Step, thus removing noises and also preserving tissue boundaries. On the other hand, the improved estimation of T 1 map from E-Step can help segment breast tissues in a more accurate and less noisy way. Therefore, by repeating these steps, we can simultaneously obtain better results for both T 1 map estimation and segmentation. Experimental results show the effectiveness of the proposed algorithm in breast tissue segmentation and parenchyma volume measurement.

Deformable registration for quantifying longitudinal tumor changes during neoadjuvant chemotherapy

“…As tumor changes are often temporally smooth, registering a full 4‐D spectrum of longitudinal images at the same time can better maintain the temporal smoothness. In addition, we can incorporate some task‐specific prior knowledge to further aid the registration, such as the removal of image background and the chest region [e.g., ], the automatically detected landmarks [e.g., ], the segmentation of various breast tissue types [e.g., tumor vs. normal fibroglandular tissue ], and even the explicit segmentation of tumors [e.g., ]. Such modifications could further aid the registration to focus on ROI, and could largely reduce the negative impacts from structures that are not of interest to the specific application.…”

A comparative study of volumetric breast density estimation in digital mammography and magnetic resonance imaging: results from a high-risk population

“…New-generation breast imaging modalities provide the opportunity for multimodality breast density estimation, including the ability to measure volumetric breast density 16, 17 . Digital imaging, in particular, allows the implementation of fully-automated computerized methods that can provide objective quantitative measures 5,13 . Such automated methods can alleviate the subjectivity of the currently used semi-automated methods (Fig.…”

“…The estimated fibroglandular tissue volume is then divided by the total breast volume to calculate the volumetric percentage of fibroglandular tissue in the breast 12 . The MR data were analyzed using a custom segmentation method 13,14 , in which the breast boundary is semi-automatically outlined using an active contour algorithm. The fibroglandular parenchyma (FP) is then segmented using a fuzzy-C-means (FCM) algorithm based on the T 1 map, which is estimated by fitting the IR-SPRG data to the Bloch equation 13 .…”

“…The MR data were analyzed using a custom segmentation method 13,14 , in which the breast boundary is semi-automatically outlined using an active contour algorithm. The fibroglandular parenchyma (FP) is then segmented using a fuzzy-C-means (FCM) algorithm based on the T 1 map, which is estimated by fitting the IR-SPRG data to the Bloch equation 13 . FCM provides a continuous voxel-wise probabilistic membership value ranging from 0 (i.e., pure fat) to 1 (i.e., pure parenchyma), which is important for dealing with the partial volume effect in the breast MR images.…”

Review of Printed Fabric Pattern Segmentation Analysis and Application

“…Several improved segmentation methodologies for printed fabric patterns have been projected over the years which demonstrated inherently appropriate in textiles quality control evaluation. Fuzzy segmentation algorithms, exclusively the fuzzy c-means (FCM) algorithm, have been generally used in the image segmentation [12].…”