Automated mammographic breast density estimation using a fully convolutional network

Lee, Juhun; Nishikawa, Robert M.

doi:10.1002/mp.12763

Cited by 82 publications

(52 citation statements)

References 29 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In clinical practice, the MGF can be determined using Breast Imaging Reporting and Data System (BI-RADS) or other clinically available software through mammogram analysis, such as the Laboratory for Individualized Breast Radiodensity Assessment (LIBRA) 28 and Volpara 29 . The glandular distribution pattern can be evaluated from the mediolateral view of mammograms by drawing regions of interest (ROIs) in the upper, middle, and lower regions inside the breast, respective.…”

Section: Discussionmentioning

confidence: 99%

Three-layer heterogeneous mammographic phantoms for Monte Carlo simulation of normalized glandular dose coefficients in mammography

Chang

Lai

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Normalized glandular dose (DgN) coefficients obtained using homogeneous breast phantoms are commonly used in breast dosimetry for mammography. However, glandular tissue is heterogeneously distributed in the breast. This study aimed to construct three-layer heterogeneous mammographic phantoms (THEPs) to examine the effect of glandular distribution on DgN coefficient. Each layer of THEPs was set to 25%, 50%, or 75% glandular fraction to emulate heterogeneous glandular distribution. Monte Carlo simulation was performed to attain mean glandular dose (MGD) and air kerma at 22-36 kVp and W/Al, W/Rh, and W/Ag target-filter combinations. The heterogeneous DgN coefficient was calculated as functions of the mean glandular fraction (MGF), breast thickness, tube voltage, and half-value layer. At 50% MGF, the heterogeneous DgN coefficients for W/Al, W/Rh, and W/Ag differed by 40.3%, 36.7%, and 31.2%. At 9-cm breast thickness, the DgN values of superior and inferior glandular distributions were 25.4% higher and 29.2% lower than those of uniform distribution. The proposed THEPs can be integrated with conventional breast dosimetry to consider the heterogeneous glandular distribution in clinical practice.Breast cancer is the second most common cause of cancer deaths among women 1 , and its incidence increases annually. X-ray mammography has become a primary tool for breast cancer screening because of the high sensitivity and specificity for microcalcification and mass detection 2 . Mammography is also adopted for the high effectiveness/cost ratio. However, the glandular tissue in the breast is sensitive to radiation. The radiation exposure during mammography may increase the risk of radiation-induced secondary breast malignancy 3 . Therefore, assessment of glandular dose in the breast is crucial.In modern breast dosimetry, one of the most critical parameters is the normalized glandular dose (DgN) coefficient, which is obtained using Monte Carlo simulation of mean glandular dose (MGD) in a breast phantom 4 . Boone used simple homogeneous breast phantoms to simulate the breast absorbed dose and applied a G factor to calculate MGD 5 . He further proposed monoenergetic DgN coefficients to facilitate rapid calculation of the DgN coefficient for arbitrary X-ray energy spectra 6 . Dance et al. 7 employed homogeneous semicylindrical phantoms to assess MGD conversion factor g, breast composition factor c, and X-ray spectrum factor s. The authors further supplemented the conversion factors for the high-energy spectra used for contrast enhanced digital mammography by using the same homogeneous breast model 8 . The above method is now the standard breast dosimetry protocols for mammography in the United Kingdom, European Union, and IAEA and is also adopted in the quality assurance protocols in USA. Sarno et al. 9 used homogeneous semi-cylinder and ad hoc shaped phantoms to exam MGD in mammography and calculate the t-factor (relative glandular dose, RGD) and T-factor in digital breast tomosynthesis (DBT).Three-dimensional (3D) imaging modalit...

show abstract

Section: Discussionmentioning

confidence: 99%

Three-layer heterogeneous mammographic phantoms for Monte Carlo simulation of normalized glandular dose coefficients in mammography

Chang

Lai

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In addition, deep convolutional neural network (CNN) is becoming state-of-the-art for many image analysis tasks, including segmentation. Although most previous publications on deep CNN for segmentation is on 2-D images, 32,33 3-D extensions exist. 34 Testing the proposed NS enhancement on other segmentation algorithms, such as breast CT-specific algorithms and more generic CNN algorithms, should be done in the future.…”

Section: Discussionmentioning

confidence: 99%

Neutrosophic segmentation of breast lesions for dedicated breast computed tomography

et al. 2018

Self Cite

View full text Add to dashboard Cite

We proposed the neutrosophic approach for segmenting breast lesions in breast computed tomography (bCT) images. The neutrosophic set considers the nature and properties of neutrality (or indeterminacy). We considered the image noise as an indeterminate component while treating the breast lesion and other breast areas as true and false components. We iteratively smoothed and contrast-enhanced the image to reduce the noise level of the true set. We then applied one existing algorithm for bCT images, the RGI segmentation, on the resulting noise-reduced image to segment the breast lesions. We compared the segmentation performance of the proposed method (named as NS-RGI) to that of the regular RGI segmentation. We used 122 breast lesions (44 benign and 78 malignant) of 111 noncontrast enhanced bCT cases. We measured the segmentation performances of the NS-RGI and the RGI using the Dice coefficient. The average Dice values of the NS-RGI and RGI were 0.82 and 0.80, respectively, and their difference was statistically significant ([Formula: see text]). We conducted a subsequent feature analysis on the resulting segmentations. The classifier performance for the NS-RGI ([Formula: see text]) improved over that of the RGI ([Formula: see text], [Formula: see text]).

show abstract

“…have found that women at high risk for breast cancer have dense breasts with parenchymal patterns that are coarse and low in contrast . DL is now being used to assess breast density . In addition, parenchymal characterization is being conducted using DL, in which the parenchymal patterns are related through the CNN architecture to groups of women using surrogate markers of risk.…”

Section: Application Areas In Radiological Imaging and Radiation Therapymentioning

confidence: 99%

Deep learning in medical imaging and radiation therapy

et al. 2018

View full text Add to dashboard Cite

The goals of this review paper on deep learning (DL) in medical imaging and radiation therapy are to (a) summarize what has been achieved to date; (b) identify common and unique challenges, and strategies that researchers have taken to address these challenges; and (c) identify some of the promising avenues for the future both in terms of applications as well as technical innovations. We introduce the general principles of DL and convolutional neural networks, survey five major areas of application of DL in medical imaging and radiation therapy, identify common themes, discuss methods for dataset expansion, and conclude by summarizing lessons learned, remaining challenges, and future directions.

show abstract

Automated mammographic breast density estimation using a fully convolutional network

Abstract: We have developed a new deep learning based algorithm for breast density segmentation and estimation. We showed that the proposed algorithm correlated well with BI-RADS density assessments by radiologists and outperformed an existing state of the art algorithm.

Cited by 82 publications

References 29 publications

Three-layer heterogeneous mammographic phantoms for Monte Carlo simulation of normalized glandular dose coefficients in mammography

Three-layer heterogeneous mammographic phantoms for Monte Carlo simulation of normalized glandular dose coefficients in mammography

Neutrosophic segmentation of breast lesions for dedicated breast computed tomography

Deep learning in medical imaging and radiation therapy

Contact Info

Product

Resources

About