Prediction of Cancer Masking in Screening Mammography Using Density and Textural Features

Clinical Significance of Combined Density and Deep-Learning-Based Texture Analysis for Stratifying the Risk of Short-Term and Long-Term Breast Cancer in Screening

Assessing a woman's risk of breast cancer is important for personalized screening. Mammographic density is a strong risk factor for breast cancer, but parenchymal texture patterns offer additional information which cannot be captured by density. We aimed to combine BI-RADS density score 4th Edition and a deep-learning-based texture score to stratify women in screening and compare rates among the combinations. This retrospective study cohort study included 216,564 women from a Danish populations-based screening program. Baseline mammograms were evaluated using BI-RADS density scores (1-4) and a deep-learning texture risk model, with scores categorized into four quartiles (1-4). The incidence rate ratio (IRR) for screen-detected, interval, and long-term cancer were adjusted for age, year of screening and screening clinic. Compared with subgroup B1-T1, the highest IRR for screen-detected cancer were within the T4 category (3.44 (95% CI: 2.43-4.82)-4.57 (95% CI: 3.66-5.76)). IRR for interval cancer was highest in the BI-RADS 4 category (95% CI: 5.36 (1.77-13.45)-16.94 (95% CI: 9.93-30.15)). IRR for long-term cancer increased both with increasing BI-RADS and increasing texture reaching 5.15 (4.31-6.16) for the combination of B4-T4 compared with B1-T1. Deep-learning-based texture analysis combined with BI-RADS density categories can reveal subgroups with increased rates beyond what density alone can ascertain, suggesting the potential of combining texture and density to improve risk stratification in breast cancer screening.

Impact of GAN artifacts for simulating mammograms on identifying mammographically occult cancer

Purpose

Generative adversarial networks (GANs) can synthesize various feasible-looking images. We showed that a GAN, specifically a conditional GAN (CGAN), can simulate breast mammograms with normal, healthy appearances and can help detect mammographically-occult (MO) cancer. However, similar to other GANs, CGANs can suffer from various artifacts, e.g., checkerboard artifacts, that may impact the quality of the final synthesized image, as well as the performance of detecting MO cancer. We explored the types of GAN artifacts that exist in mammogram simulations and their effect on MO cancer detection.

Approach

We first trained a CGAN using digital mammograms (FFDMs) of 1366 women with normal/healthy breasts. Then, we tested the trained CGAN on an independent MO cancer dataset with 333 women with dense breasts (97 MO cancers). We trained a convolutional neural network (CNN) on the MO cancer dataset, in which real and simulated mammograms were fused, to identify women with MO cancer. Then, a radiologist who was independent of the development of the CGAN algorithms evaluated the entire MO cancer dataset to identify and annotate artifacts in the simulated mammograms.

Results

We found four artifact types, including checkerboard, breast boundary, nipple-areola complex, and black spots around calcification artifacts, with an overall incidence rate over 69% (the individual incident rate ranged from 9% to 53%) from both normal and MO cancer samples. We then evaluated their potential impact on MO cancer detection. Even though various artifacts existed in the simulated mammogram, we found that it still provided complementary information for MO cancer detection when it was combined with the real mammograms.

Conclusions

We found that artifacts were pervasive in the CGAN-simulated mammograms. However, they did not negatively affect our MO cancer detection algorithm; the simulated mammograms still provided complementary information for MO cancer detection when combined with real mammograms.

The effect of variable labels on deep learning models trained to predict breast density

Purpose. High breast density is associated with reduced efficacy of mammographic screening and increased risk of developing breast cancer. Accurate and reliable automated density estimates can be used for direct risk prediction and passing density related information to further predictive models. Expert reader assessments of density show a strong relationship to cancer risk but also inter-reader variation. The effect of label variability on model performance is important when considering how to utilise automated methods for both research and clinical purposes.Methods. We utilise subsets of images with density labels from the same 13 readers and 12 reader pairs, and train a deep transfer learning model which is used to assess how label variability affects the mapping from representation to prediction. We then create two end-to-end models: one that is trained on averaged labels across the reader pairs and the second that is trained using individual reader scores, with a novel alteration to the objective function. The combination of these two end-to-end models allows us to investigate the effect of label variability on the model representation formed.Results. We show that the trained mappings from representations to labels are altered considerably by the variability of reader scores. Training on labels with distribution variation removed causes the Spearman rank correlation coefficients to rise from 0.751 ± 0.002 to either 0.815 ± 0.026 when averaging across readers or 0.844 ± 0.002 when averaging across images. However, when we train different models to investigate the representation effect we see little difference, with Spearman rank correlation coefficients of 0.846 ± 0.006 and 0.850 ± 0.006 showing no statistically significant difference in the quality of the model representation with regard to density prediction.Conclusions. We show that the mapping between representation and mammographic density prediction is significantly affected by label variability. However, the effect of the label variability on the model representation is limited.

Mouse Mammary Gland Whole Mount Density Assessment across Different Morphologies Using a Bifurcated Program for Image Processing

Mammographic density is associated with increased breast cancer risk. Conventional visual assessment of murine mouse models does not include quantified total density analysis. A bifurcated method was sufficient to obtain relative density scores on a broad range of two-dimensional whole mount images that contained both normal and abnormal findings. Image processing techniques, including a ridge operator and a gaussian denoising method, were used to isolate background away from mammary epithelium and use mean pixel intensity to represent mammary density on genetically engineered mouse models for breast cancer in mice 4 to 29 months of age. The bifurcated method allowed for application of an optimal image processing approach for the structural elements present in the whole mount images. Gaussian denoising was the optimal approach when more dense lobular growth and tertiary branching dominate and a ridge operator when epithelial growth was more sparse and secondary branching was the more dominant structural feature. The two processing approaches were combined in a single experimental flow program using an initial image density measurement as the decision point between the two approaches. Higher density was associated with lobular growth, tertiary branching, fibrotic stroma, and presence of cancer. The significance of the study is development of a readily accessible program for digital assessment of mammary gland whole mount density across a range of mammary gland morphologies.

Breast Density: Current Knowledge, Assessment Methods, and Clinical Implications

Breast density is an accepted independent risk factor for the future development of breast cancer, and greater breast density has the potential to mask malignancies on mammography, thus lowering the sensitivity of screening mammography. The risk associated with dense breast tissue has been shown to be modifiable with changes in breast density. Numerous studies have sought to identify factors that influence breast density, including age, genetic, racial/ethnic, prepubertal, adolescent, lifestyle, environmental, hormonal, and reproductive history factors. Qualitative, semiquantitative, and quantitative methods of breast density assessment have been developed, but to date there is no consensus assessment method or reference standard for breast density. Breast density has been incorporated into breast cancer risk models, and there is growing consciousness of the clinical implications of dense breast tissue in both the medical community and public arena. Efforts to improve breast cancer screening sensitivity for women with dense breasts have led to increased attention to supplemental screening methods in recent years, prompting the American College of Radiology to publish Appropriateness Criteria for supplemental screening based on breast density.

Identifying Women With Mammographically- Occult Breast Cancer Leveraging GAN-Simulated Mammograms

Our objective is to show the feasibility of using simulated mammograms to detect mammographically-occult (MO) cancer in women with dense breasts and a normal screening mammogram who could be triaged for additional screening with magnetic resonance imaging (MRI) or ultrasound. We developed a Conditional Generative Adversarial Network (CGAN) to simulate a mammogram with normal appearance using the opposite mammogram as the condition. We used a Convolutional Neural Network (CNN) trained on Radon Cumulative Distribution Transform (RCDT) processed mammograms to detect MO cancer. For training CGAN, we used screening mammograms of 1366 women. For MO cancer detection, we used screening mammograms of 333 women (97 MO cancer) with dense breasts. We simulated the right mammogram for normal controls and the cancer side for MO cancer cases. We created two RCDT images, one from a real mammogram pair and another from a real-simulated mammogram pair. We finetuned a VGG16 on resulting RCDT images to classify the women with MO cancer. We compared the classification performance of the CNN trained on fused RCDT images, CNN_Fused to that of trained only on real RCDT images, CNN_Real, and to that of trained only on simulated RCDT images, CNN_Simulated. The test AUC for CNN_Fused was 0.77 with a 95% confidence interval (95CI) of [0.71, 0.83], which was statistically better (p-value < 0.02) than the CNN_Real AUC of 0.70 with a 95CI of [0.64, 0.77] and CNN_Simulated AUC of 0.68 with a 95CI of [0.62, 0.75]. It showed that CGAN simulated mammograms can help MO cancer detection.

Conceptual design of the dual X-ray absorptiometry health informatics prediction system for osteoporosis care

Osteoporotic fractures are a major and growing public health problem, which is strongly associated with other illnesses and multi-morbidity. Big data analytics has the potential to improve care for osteoporotic fractures and other non-communicable diseases (NCDs), reduces healthcare costs and improves healthcare decision-making for patients with multi-disorders. However, robust and comprehensive utilization of healthcare big data in osteoporosis care practice remains unsatisfactory. In this paper, we present a conceptual design of an intelligent analytics system, namely, the dual X-ray absorptiometry (DXA) health informatics prediction (HIP) system, for healthcare big data research and development. Comprising data source, extraction, transformation, loading, modelling and application, the DXA HIP system was applied in an osteoporosis healthcare context for fracture risk prediction and the investigation of multi-morbidity risk. Data was sourced from four DXA machines located in three healthcare centres in Ireland. The DXA HIP system is novel within the Irish context as it enables the study of fracture-related issues in a larger and more representative Irish population than previous studies. We propose this system is applicable to investigate other NCDs which have the potential to improve the overall quality of patient care and substantially reduce the burden and cost of all NCDs.

Opportunities in cancer imaging: risk-adapted breast imaging in screening

In the UK, women between 50-70 years are invited for 3-yearly mammography screening irrespective of their likelihood of developing breast cancer. The only risk adaption is for women with >30% lifetime risk who are offered annual magnetic resonance imaging (MRI) and mammography, and annual mammography for some moderate-risk women. Using questionnaires, breast density, and polygenic risk scores, it is possible to stratify the population into the lowest 20% risk, who will develop <4% of cancers and the top 4%, who will develop 18% of cancers. Mammography is a good screening test but has low sensitivity of 60% in the 9% of women with the highest category of breast density (BIRADS D) who have a 2.5- to fourfold breast cancer risk. There is evidence that adding ultrasound to the screening mammogram can increase the cancer detection rate and reduce advanced stage interval and next round cancers. Similarly, alternative tests such as contrast-enhanced mammography (CESM) or abbreviated MRI (ABB-MRI) are much more effective in detecting cancer in women with dense breasts. Scintimammography has been shown to be a viable alternative for dense breasts or for follow-up in those with a personal history of breast cancer and scarring as result of treatment. For supplemental screening to be worthwhile in these women, new technologies need to reduce the number of stage II cancers and be cost effective when tested in large scale trials. This article reviews the evidence for supplemental imaging and examines whether a risk-stratified approach is feasible.

Radiomics: A Primer for Breast Radiologists

Radiomics has a long-standing history in breast imaging with computer-aided detection (CAD) for screening mammography developed in the late 20th century. Although conventional CAD had widespread adoption, the clinical benefits for experienced breast radiologists were debatable due to high false-positive marks and subsequent increased recall rates. The dramatic growth in recent years of artificial intelligence-based analysis, including machine learning and deep learning, has provided numerous opportunities for improved modern radiomics work in breast imaging. There has been extensive radiomics work in mammography, digital breast tomosynthesis, MRI, ultrasound, PET-CT, and combined multimodality imaging. Specific radiomics outcomes of interest have been diverse, including CAD, prediction of response to neoadjuvant therapy, lesion classification, and survival, among other outcomes. Additionally, the radiogenomics subfield that correlates radiomics features with genetics has been very proliferative, in parallel with the clinical validation of breast cancer molecular subtypes and gene expression assays. Despite the promise of radiomics, there are important challenges related to image normalization, limited large unbiased data sets, and lack of external validation. Much of the radiomics work to date has been exploratory using single-institution retrospective series for analysis, but several promising lines of investigation have made the leap to clinical practice with commercially available products. As a result, breast radiologists will increasingly be incorporating radiomics-based tools into their daily practice in the near future. Therefore, breast radiologists must have a broad understanding of the scope, applications, and limitations of radiomics work.

RICE: A method for quantitative mammographic image enhancement

We introduce Region of Interest Contrast Enhancement (RICE) to identify focal densities in mammograms. It aims to help radiologists: 1) enhancing the contrast of mammographic images; and 2) detecting regions of interest (such as focal densities) that are candidate masses potentially masked behind dense parenchyma. Cancer masking is an unsolved issue, particularly in breast density categories BI-RADS C and D. RICE suppresses normal breast parenchyma in order to highlight focal densities. Unlike methods that enhance mammograms by modifying the dynamic range of an image; RICE relies on the actual tissue composition of the breast. It segments Volumetric Breast Density (VBD) maps into smaller regions and then applies a recursive mechanism to estimate the 'neighbourhood' for each segment. The method then subtracts and updates the neighbourhood, or the encompassing tissue, from each piecewise constant component of the breast image. This not only enhances the appearance of a candidate mass but also helps in estimating the mass density. In extensive experiments, RICE enhances focal densities in all breast density types including the most challenging category BI-RADS D. Suitably adapted, RICE can be used as a precursor to any computer-aided diagnostics and detection system.

Use of Low-Dose Tamoxifen to Increase Mammographic Screening Sensitivity in Premenopausal Women

Increased breast density decreases mammographic sensitivity due to masking of cancers by dense tissue. Tamoxifen exposure reduces mammographic density and, therefore, should improve screening sensitivity. We modelled how low-dose tamoxifen exposure could be used to increase mammographic sensitivity. Mammographic sensitivity was calculated using the KARMA prospective screening cohort. Two models were fitted to estimate screening sensitivity and detected tumor size based on baseline mammographic density. BI-RADS-dependent sensitivity was estimated. The results of the 2.5 mg tamoxifen arm of the KARISMA trial were used to define expected changes in mammographic density after six months exposure and to predict changes in mammographic screening sensitivity and detected tumor size. Rates of interval cancers and detection of invasive tumors were estimated for women with mammographic density relative decreases by 10-50%. In all, 517 cancers in premenopausal women were diagnosed in KARMA: 287 (56%) screen-detected and 230 (44%) interval cancers. Screening sensitivities prior to tamoxifen, were 76%, 69%, 53%, and 46% for BI-RADS density categories A, B, C, and D, respectively. After exposure to tamoxifen, modelled screening sensitivities were estimated to increase by 0% (p = 0.35), 2% (p < 0.01), 5% (p < 0.01), and 5% (p < 0.01), respectively. An estimated relative density decrease by ≥20% resulted in an estimated reduction of interval cancers by 24% (p < 0.01) and reduction in tumors >20 mm at detection by 4% (p < 0.01). Low-dose tamoxifen has the potential to increase mammographic screening sensitivity and thereby reduce the proportion of interval cancers and larger screen-detected cancers.

Sensitivity of screening mammography by density and texture: a cohort study from a population-based screening program in Denmark

Background

Screening mammography works better in fatty than in dense breast tissue. Computerized assessment of parenchymal texture is a non-subjective method to obtain a refined description of breast tissue, potentially valuable in addition to breast density scoring for the identification of women in need of supplementary imaging. We studied the sensitivity of screening mammography by a combination of radiologist-assessed Breast Imaging Reporting and Data System (BI-RADS) density score and computer-assessed parenchymal texture marker, mammography texture resemblance (MTR), in a population-based screening program.

Methods

Breast density was coded according to the fourth edition of the BI-RADS density code, and MTR marker was divided into quartiles from 1 to 4. Screening data were followed up for the identification of screen-detected and interval cancers. We calculated sensitivity and specificity with 95% confidence intervals (CI) by BI-RADS density score, MTR marker, and combination hereof.

Results

Density and texture were strongly correlated, but the combination led to the identification of subgroups with different sensitivity. Sensitivity was high, about 80%, in women with BI-RADS density score 1 and MTR markers 1 or 2. Sensitivity was low, 67%, in women with BI-RADS density score 2 and MTR marker 4. For women with BI-RADS density scores 3 and 4, the already low sensitivity was further decreased for women with MTR marker 4. Specificity was 97–99% in all subgroups.

Conclusion

Our study showed that women with low density constituted a heterogenous group. Classifying women for extra imaging based on density only might be a too crude approach. Screening sensitivity was systematically high in women with fatty and homogenous breast tissue.

Detecting mammographically occult cancer in women with dense breasts using deep convolutional neural network and Radon Cumulative Distribution Transform

We have applied the Radon Cumulative Distribution Transform (RCDT) as an image transformation to highlight the subtle difference between left and right mammograms to detect mammographically occult (MO) cancer in women with dense breasts and negative screening mammograms. We developed deep convolutional neural networks (CNNs) as classifiers for estimating the probability of having MO cancer. We acquired screening mammograms of 333 women (97 unilateral MO cancer) with dense breasts and at least two consecutive mammograms and used the immediate prior mammograms, which radiologists interpreted as negative. We used fivefold cross validation to divide our dataset into a training and independent test sets with ratios of 0.8:0.2. We set aside 10% of the training set as a validation set. We applied RCDT on the left and right mammograms of each view. We applied inverse Radon transform to represent the resulting RCDT images in the image domain. We then fine-tuned a VGG16 network pretrained on ImageNet using the resulting images per each view. The CNNs achieved mean areas under the receiver operating characteristic (AUC) curve of 0.73 (standard error, SE = 0.024) and 0.73 (SE = 0.04) for the craniocaudal and mediolateral oblique views, respectively. We combined the scores from two CNNs by training a logistic regression classifier and it achieved a mean AUC of 0.81 (SE = 0.032). In conclusion, we showed that inverse Radon-transformed RCDT images contain information useful for detecting MO cancers and that deep CNNs could learn such information.

Investigating the feasibility of stratified breast cancer screening using a masking risk predictor

BACKGROUND

Women with dense breasts face a double risk for breast cancer; they are at a higher risk for development of breast cancer than those with less dense breasts, and there is a greater chance that mammography will miss detection of a cancer in dense breasts due to the masking effect of surrounding fibroglandular tissue. These women may be candidates for supplemental screening. In this study, a masking risk model that was previously developed is tested on a cohort of cancer-free women to assess potential efficiency of stratification.

METHODS

Three masking risk models based on (1) BI-RADS density, (2) volumetric breast density (VBD), and (3) a combination of VBD and detectability were applied to stratify the mammograms of 1897 cancer-free women. The fraction of cancer-free women whose mammograms were deemed by the algorithm to be masked and who would be considered for supplemental imaging was computed as was the corresponding fraction in a screened population of interval (masked) cancers that would be potentially detected by supplemental imaging.

RESULTS

Of the models tested, the combined VBD/detectability model offered the highest efficiency for stratification to supplemental imaging. It predicted that 725 supplemental screens would be performed per interval cancer potentially detected, at an operating point that allowed detection of 64% of the interval cancers. In comparison, stratification based on the upper two BI-RADS density categories required 1117 supplemental screenings per interval cancer detected to capture 64% of interval cancers.

CONCLUSION

The combined VBD/detectability models perform better than BI-RADS and offer a continuum of operating points, suggesting that this model may be effective in guiding a stratified screening environment.

Emergence of "Big Data" and Its Potential and Current Limitations in Medical Imaging

Although electronic imaging was performed in the early 1950s in nuclear medicine, it was the introduction of computed tomography in 1972 that caused a revolution in medical imaging in that it marked the beginning of the inevitable transformation to digital imaging. This transformation is now more or less complete. While initially these CT images were relatively small, comprised of only about 6400 pixels per slice, the steady move toward higher spatial resolution, multislice imaging, digital radiography, and fluoroscopy rapidly increased the size of images and the amount of data required to be stored, processed, displayed, and moved about in a medical imaging department. The more recent introduction of digital pathology with submicron-sized pixels and the need for color further increases these demands. Rising work volumes in hospital, a push for cost containment, and a move toward greater precision in diagnosis and treatment of disease all work together to motivate the development of automated image analysis algorithms and techniques to improve efficiencies in in vivo imaging and pathology. This may require bringing together information from different imaging and nonimaging sources within the institution. While technological development has provided practical means for storage of the burgeoning data load and the use of multiple processors and high-speed networks has enabled more sophisticated analysis locally or in the cloud, challenges remain in terms of the ability to integrate data from different systems, the development of appropriately annotated image bases for training and testing of algorithms, and issues around privacy and ownership in obtaining access to patient-related data.