Breast Density Classification with Deep Convolutional Neural Networks

Wu, Nan; Geras, Krzysztof J.; Shen, Yiqiu; Su, Jingyi; Kim, S. Gene; Kim, Eric; Wolfson, S. Sabina; Moy, Linda; Cho, Kyunghyun

doi:10.1109/icassp.2018.8462671

Cited by 64 publications

(43 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a study of 6081 patients, Brandt et al showed moderate agreement, with k scores of 0.46 for Quantra and 0.57 for Volpara (13). More recently, Wu et al used a DL model to assess density in a reader study of 100 mammograms and showed moderate agreement between their DL model and the assessment of an experienced breast imager, with a k score of 0.48 (12). In contrast, our DL model showed higher agreement, with k scores of 0.67 (95% CI: 0.66, 0.68) and 0.78 (95% CI: 0.73, 0.82) in our test set and reader study, respectively.…”

Section: Discussionmentioning

confidence: 99%

Mammographic Breast Density Assessment Using Deep Learning: Clinical Implementation

et al. 2019

View full text Add to dashboard Cite

ammographic breast density can mask cancers at mammography and is an independent risk factor for breast cancer (1-3). Legislation mandating patients be notified of mammographic breast density has passed in more than 30 states, and a federal bill is under consideration. Details of state legislation vary, but most states require direct reporting to the patient that breast density can mask cancers at mammography and that the patient may benefit from additional testing. Qualitative assessment of mammographic breast density is subjective and varies widely between radiologists (4-10). In a study of 83 radiologists who assessed breast density, Sprague et al (4) found extreme variation in qualitative density assessment per the Breast Imaging Reporting and Data System (BI-RADS), with 6%-85% of mammograms assessed as either heterogeneously or extremely dense depending on radiologist interpretation. In a study of 34 radiologists, the intraradiologist agreement of density assessments among women who underwent two examinations varied from 62% to 87% (6). Commercially available methods for automated assessment of breast density do exist; however, they yield mixed results in agreement with expert qualitative density assessments, with k scores of 0.32-0.61 (11,12). These methods tend to result in over-or underreporting of breast density when compared with qualitative assessment by radiologists (11,13). A recent study found significant differences in density assessments in the same 4170 women with two software programs (Volpara, Volpara Solutions, Wellington, New Zealand; Quantra, Hologic, Bedford, Mass), with the software programs showing 37% and 51%, respectively, of women had dense breast tissue. In the same set of mammograms, radiologists determined 43% of the women had dense breast tissue (13). Deep learning (DL) has been gaining traction in radiology (12,14-17). Specifically, there has been preliminary work with DL methods to assess breast density (12,18); however, none of these techniques have been implemented in clinical practice, raising questions about clinical acceptance by practicing radiologists and the effect on patient care. In contrast, our purpose was to develop a DL algorithm we could use to reliably assess breast density and to measure the acceptance of its predictions in real-time clinical practice. We hypothesize that DL models can be applied to assess breast density at the same level as experienced breast imagers and that they can be accepted into routine clinical practice.

show abstract

Section: Discussionmentioning

confidence: 99%

Mammographic Breast Density Assessment Using Deep Learning: Clinical Implementation

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Informed consent was waived. This dataset * is a larger and more carefully curated version of a dataset used in our earlier work (14,15). The dataset includes 229,426 digital screening mammography exams (1, To extract labels indicating whether each breast of the patient was found to have malignant or benign findings at the end of the diagnostic pipeline, we relied on pathology reports from biopsies.…”

Section: Datamentioning

confidence: 99%

Deep Neural Networks Improve Radiologists’ Performance in Breast Cancer Screening

Phang

Park

et al. 2020

IEEE Trans. Med. Imaging

Self Cite

504

397

View full text Add to dashboard Cite

We present a deep convolutional neural network for breast cancer screening exam classification, trained and evaluated on over 200,000 exams (over 1,000,000 images). Our network achieves an AUC of 0.895 in predicting whether there is a cancer in the breast, when tested on the screening population. We attribute the high accuracy of our model to a two-stage training procedure, which allows us to use a very high-capacity patch-level network to learn from pixel-level labels alongside a network learning from macroscopic breast-level labels. To validate our model, we conducted a reader study with 14 readers, each reading 720 screening mammogram exams, and find our model to be as accurate as experienced radiologists when presented with the same data. Finally, we show that a hybrid model, averaging probability of malignancy predicted by a radiologist with a prediction of our neural network, is more accurate than either of the two separately. To better understand our results, we conduct a thorough analysis of our network's performance on different subpopulations of the screening population, model design, training procedure, errors, and properties of its internal representations.deep learning | deep convolutional neural networks | breast cancer screening | mammography B reast cancer is the second leading cancer-related cause of death among women in the US. In 2014, over 39 million screening and diagnostic mammography exams were performed in the US. It is estimated that in 2015 232,000 women were diagnosed with breast cancer and approximately 40,000 died from it (1). Although mammography is the only imaging test that has reduced breast cancer mortality (2-4), there has been discussion regarding the potential harms of screening, including false positive recalls and associated false positive biopsies. The vast majority of the 10-15% of women asked to return following an inconclusive screening mammogram undergo another mammogram and/or ultrasound for clarification. After the additional imaging exams, many of these findings are determined as benign and only 10-20% are recommended to undergo a needle biopsy for further work-up. Among these, only 20-40% yield a diagnosis of cancer (5). Evidently, there is an unmet need to shift the balance of routine breast cancer screening towards more benefit and less harm.Traditional computer-aided detection (CAD) in mammography is routinely used by radiologists to assist with image interpretation, despite multicenter studies showing these CAD programs do not improve their diagnostic performance (6).These CAD programs typically use handcrafted features to mark sites on a mammogram that appear distinct from normal tissue structures. The radiologist decides whether to recall these findings, determining clinical significance and actionability. Recent developments in deep learning (7)-in particular, deep convolutional neural networks (CNNs) (8-12)-open possibilities for creating a new generation of CAD-like tools.This paper makes several contributions. Primarily, we train and evaluate a set of stro...

show abstract

“…With the advent of deep learning in image processing and classification, the technique has been gaining more attention in medical image processing [34,35] worldwide for segmentation and classification processes. Various studies have been performed using deep learning for density classification resulting in promising classification models [36][37][38][39]. But most of the deep learning methods focused on binary classification (fatty or dense) of mammographic breast density.…”

Section: Introductionmentioning

confidence: 99%

Comparative Study on Local Binary Patterns for Mammographic Density and Risk Scoring

George

Zwiggelaar

2019

J. Imaging

View full text Add to dashboard Cite

Breast density is considered to be one of the major risk factors in developing breast cancer. High breast density can also affect the accuracy of mammographic abnormality detection due to the breast tissue characteristics and patterns. We reviewed variants of local binary pattern descriptors to classify breast tissue which are widely used as texture descriptors for local feature extraction. In our study, we compared the classification results for the variants of local binary patterns such as classic LBP (Local Binary Pattern), ELBP (Elliptical Local Binary Pattern), Uniform ELBP, LDP (Local Directional Pattern) and M-ELBP (Mean-ELBP). A wider comparison with alternative texture analysis techniques was studied to investigate the potential of LBP variants in density classification. In addition, we investigated the effect on classification when using descriptors for the fibroglandular disk region and the whole breast region. We also studied the effect of the Region-of-Interest (ROI) size and location, the descriptor size, and the choice of classifier. The classification results were evaluated based on the MIAS database using a ten-run ten-fold cross validation approach. The experimental results showed that the Elliptical Local Binary Pattern descriptors and Local Directional Patterns extracted most relevant features for mammographic tissue classification indicating the relevance of directional filters. Similarly, the study showed that classification of features from ROIs of the fibroglandular disk region performed better than classification based on the whole breast region.

show abstract

Breast Density Classification with Deep Convolutional Neural Networks

Cited by 64 publications

References 14 publications

Mammographic Breast Density Assessment Using Deep Learning: Clinical Implementation

Mammographic Breast Density Assessment Using Deep Learning: Clinical Implementation

Deep Neural Networks Improve Radiologists’ Performance in Breast Cancer Screening

Comparative Study on Local Binary Patterns for Mammographic Density and Risk Scoring

Contact Info

Product

Resources

About