Multiscale Monte Carlo simulations for dosimetry in x-ray breast imaging: Part I - Macroscopic scales.

BACKGROUND: X-ray breast imaging modalities are commonly employed for breast cancer detection, from screening programs to diagnosis. Thus, dosimetry studies are important for quality control and risk estimation since ionizing radiation is used. PURPOSE: To perform multiscale dosimetry assessments for different breast imaging modalities and for a variety of breast sizes and compositions. The first part of our study is focused on macroscopic scales (down to millimeters). METHODS: Nine anthropomorphic breast phantoms with a voxel resolution of 0.5 mm were computationally generated using the BreastPhantom software, representing three breast sizes with three distinct values of volume glandular fraction (VGF) for each size. Four breast imaging modalities were studied: digital mammography (DM), contrast-enhanced digital mammography (CEDM), digital breast tomosynthesis (DBT) and dedicated breast computed tomography (BCT). Additionally, the impact of tissue elemental compositions from two databases were compared. Monte Carlo (MC) simulations were performed with the MC-GPU code to obtain the 3D glandular dose distribution (GDD) for each case considered with the mean glandular dose (MGD) fixed at 4 mGy (to facilitate comparisons). RESULTS: The GDD within the breast is more uniform for CEDM and BCT compared to DM and DBT. For large breasts and high VGF, the ratio between the minimum/maximum glandular dose to MGD is 0.12/4.02 for DM and 0.46/1.77 for BCT; the corresponding results for a small breast and low VGF are 0.35/1.98 (DM) and 0.63/1.42 (BCT). The elemental compositions of skin, adipose and glandular tissue have a considerable impact on the MGD, with variations up to 30% compared to the baseline. The inclusion of tissues other than glandular and adipose within the breast has a minor impact on MGD, with differences below 2%. Variations in the final compressed breast thickness alter the shape of the GDD, with a higher compression resulting in a more uniform GDD. CONCLUSIONS: For a constant MGD, the GDD varies with imaging modality and breast compression. Elemental tissue compositions are an important factor for obtaining MGD values, being a source of systematic uncertainties in MC simulations and, consequently, in breast dosimetry.

Deep learning, radiomics and radiogenomics applications in the digital breast tomosynthesis: a systematic review.

BACKGROUND: Recent advancements in computing power and state-of-the-art algorithms have helped in more accessible and accurate diagnosis of numerous diseases. In addition, the development of de novo areas in imaging science, such as radiomics and radiogenomics, have been adding more to personalize healthcare to stratify patients better. These techniques associate imaging phenotypes with the related disease genes. Various imaging modalities have been used for years to diagnose breast cancer. Nonetheless, digital breast tomosynthesis (DBT), a state-of-the-art technique, has produced promising results comparatively. DBT, a 3D mammography, is replacing conventional 2D mammography rapidly. This technological advancement is key to AI algorithms for accurately interpreting medical images. OBJECTIVE AND METHODS: This paper presents a comprehensive review of deep learning (DL), radiomics and radiogenomics in breast image analysis. This review focuses on DBT, its extracted synthetic mammography (SM), and full-field digital mammography (FFDM). Furthermore, this survey provides systematic knowledge about DL, radiomics, and radiogenomics for beginners and advanced-level researchers. RESULTS: A total of 500 articles were identified, with 30 studies included as the set criteria. Parallel benchmarking of radiomics, radiogenomics, and DL models applied to the DBT images could allow clinicians and researchers alike to have greater awareness as they consider clinical deployment or development of new models. This review provides a comprehensive guide to understanding the current state of early breast cancer detection using DBT images. CONCLUSION: Using this survey, investigators with various backgrounds can easily seek interdisciplinary science and new DL, radiomics, and radiogenomics directions towards DBT.

Denoising digital breast tomosynthesis projections using deep learning with synthetic data as training set.

Purpose: Image denoising based on deep neural networks (DNN) needs a big dataset containing digital breast tomosynthesis (DBT) projections acquired in different radiation doses to be trained, which is impracticable. Therefore, we propose extensively investigating the use of synthetic data generated by software for training DNNs to denoise DBT real data. Approach: The approach consists of generating a synthetic dataset representative of the DBT sample space by software, containing noisy and original images. Synthetic data were generated in two different ways: (a) virtual DBT projections generated by OpenVCT and (b) noisy images synthesized from photography regarding noise models used in DBT (e.g., Poisson-Gaussian noise). Then, DNN-based denoising techniques were trained using a synthetic dataset and tested for denoising physical DBT data. Results were evaluated in quantitative (PSNR and SSIM measures) and qualitative (visual analysis) terms. Furthermore, a dimensionality reduction technique (t-SNE) was used for visualization of sample spaces of synthetic and real datasets. Results: The experiments showed that training DNN models with synthetic data could denoise DBT real data, achieving competitive results to traditional methods in quantitative terms but showing a better balance between noise filtering and detail preservation in a visual analysis. T-SNE enables us to visualize if synthetic and real noises are in the same sample space. Conclusion: We propose a solution for the lack of suitable training data to train DNN models for denoising DBT projections, showing that we just need the synthesized noise to be in the same sample space as the target image.

Artificial Intelligence in Breast Imaging: A Special Focus on Advances in Digital Mammography & Digital Breast Tomosynthesis.

Breast cancer accounts for 30% of female cancers and is the second leading cause of cancerrelated deaths in women. The rate is rising at 0.4% per year. Early detection is crucial to improve treatment efficacy and overall survival of women diagnosed with breast cancer. Digital Mammography and Digital Breast Tomosynthesis have widely demonstrated their role as a screening tool. However, screening mammography is limited by radiologist's experience, unnecessarily high recalls, overdiagnosis, overtreatment and, in the case of Digital Breast Tomosynthesis, long reporting time. This is compounded by an increasing shortage of manpower and resources issue, especially among breast imaging specialists. Recent advances in image analysis with the use of artificial intelligence (AI) in breast imaging have the potential to overcome some of these needs and address the clinical challenges in cancer detection, assessment of treatment response, and monitoring disease progression. This article focuses on the most important clinical implication and future application of AI in the field of digital mammography and digital breast tomosynthesis, providing the readers with a comprehensive overview of AI impact in cancer detection, diagnosis, reduction of workload and breast cancer risk stratification.

Proposal of an algorithm to evaluate geometric distortion and artifact spreading in digital breast tomosynthesis.

BACKGROUND: According to the European Reference Organization for Quality Assurance Breast Screening and European Diagnostic Services, the spatial accuracy of reconstructed images and reconstruction artifacts must be evaluated in digital breast tomosynthesis (DBT) quality control procedures. PURPOSE: To propose a computational algorithm to evaluate the geometric distortion and artifact spreading (GDAS) in DBT images. MATERIAL AND METHODS: The proposed algorithm analyzed tomosynthesis images of a phantom that contains aluminum spheres (1 mm in diameter) arranged in a rectangular matrix spaced 5 cm apart that was inserted in 5-mm-thick polymethylmethacrylate (PMMA). RESULTS: The obtained results were compared with the values provided by the algorithm developed by the National Coordinating Center for the Physics of Mammography (NCCPM). In the comparison, the results depended on the dimensions of the region of interest (ROI). This dependence proves the benefit of the proposed algorithm because it allows the user to select the ROI. CONCLUSION: The computational algorithm proved to be useful for the evaluation of GDAS in DBT images, in the same way as the reference algorithm (NCCPM), as well as allowing the selection of the ROI dimensions that best suit the spreading of the artifact in the analyzed images.

Novel Perlin-based Phantoms Using 3D Models of Compressed Breast Shape and Fractal Noise.

Virtual clinical trials (VCTs) have been used widely to evaluate digital breast tomosynthesis (DBT) systems. VCTs require realistic simulations of the breast anatomy (phantoms) to characterize lesions and to estimate risk of masking cancers. This study introduces the use of Perlin-based phantoms to optimize the acquisition geometry of a novel DBT prototype. These phantoms were developed using a GPU implementation of a novel library called Perlin-CuPy. The breast anatomy is simulated using 3D models under mammography cranio-caudal compression. In total, 240 phantoms were created using compressed breast thickness, chest-wall to nipple distance, and skin thickness that varied in a {[35, 75], [59, 130), [1.0, 2.0]} mm interval, respectively. DBT projections and reconstructions of the phantoms were simulated using two acquisition geometries of our DBT prototype. The performance of both acquisition geometries was compared using breast volume segmentations of the Perlin phantoms. Results show that breast volume estimates are improved with the introduction of posterior-anterior motion of the x-ray source in DBT acquisitions. The breast volume is overestimated in DBT, varying substantially with the acquisition geometry; segmentation errors are more evident for thicker and larger breasts. These results provide additional evidence and suggest that custom acquisition geometries can improve the performance and accuracy in DBT. Perlin phantoms help to identify limitations in acquisition geometries and to optimize the performance of the DBT prototypes.

Multiclass Segmentation of Suspicious Findings in Simulated Breast Tomosynthesis Images Using a U-Net.

Our lab has built a next-generation tomosynthesis (NGT) system utilizing scanning motions with more degrees of freedom than clinical digital breast tomosynthesis systems. We are working toward designing scanning motions that are customized around the locations of suspicious findings. The first step in this direction is to demonstrate that these findings can be detected with a single projection image, which can guide the remainder of the scan. This paper develops an automated method to identify findings that are prone to be masked. Perlin-noise phantoms and synthetic lesions were used to simulate masked cancers. NGT projections of phantoms were simulated using ray-tracing software. The risk of masking cancers was mapped using the ground-truth labels of phantoms. The phantom labels were used to denote regions of low and high risk of masking suspicious findings. A U-Net model was trained for multiclass segmentation of phantom images. Model performance was quantified with a receiver operating characteristic (ROC) curve using area under the curve (AUC). The ROC operating point was defined to be the point closest to the upper left corner of ROC space. The output predictions showed an accurate segmentation of tissue predominantly adipose (mean AUC of 0.93). The predictions also indicate regions of suspicious findings; for the highest risk class, mean AUC was 0.89, with a true positive rate of 0.80 and a true negative rate of 0.83 at the operating point. In summary, this paper demonstrates with virtual phantoms that a single projection can indeed be used to identify suspicious findings.

Breast Cancer Screening Among Medically Underserved Women in New Mexico: Potential for Lower Recall Rates with Digital Breast Tomosynthesis.

Introduction: Digital breast tomosynthesis (DBT) may decrease recall rates (RRs) and improve positive predictive values (PPVs) and cancer detection rates (CDRs) versus full-field digital mammography (FFDM). The value of DBT has not been assessed in New Mexico's rural and minority population. Objectives of this study were to compare RRs, CDRs, and PPVs using FFDM+DBT versus FFDM in screening mammograms at the University of New Mexico between 2013 and 2016 and to qualitatively evaluate patient decision-making regarding DBT. Materials and Methods: RRs, CDRs, and PPVs with 95% confidence intervals and relative risk were calculated from 35,147 mammograms. The association between relative risk and mammography approach was tested using Pearson's chi-square test. Twenty women undergoing screening were interviewed for qualitative evaluation of decision-making. Results: From 2013 to 2016, RRs were 8.4% and 11.1% for FFDM+DBT and FFDM, respectively. The difference in RRs became more pronounced with time. No significant difference was observed in PPVs or CDRs. Qualitative interviews revealed that the majority had limited prior knowledge of DBT and relied on provider recommendations. Conclusion: In New Mexico women undergoing screening mammography, a 30% relative risk reduction in RRs was observed with FFDM+DBT. Qualitative interviews suggest that women are aware of and receptive to DBT, assuming adequate educational support. Clinical Trials.gov ID: NCT03979729.

Characterization of the imaging settings in screening mammography using a tracking and reporting system: A multi-center and multi-vendor analysis.

A tracking and reporting system was developed to monitor radiation dose in X-ray breast imaging. We used our tracking system to characterize and compare the mammographic practices of five breast imaging centers located in the United States and Brazil. Clinical data were acquired using eight mammography systems comprising three modalities: computed radiography (CR), full-field digital mammography (FFDM), and digital breast tomosynthesis (DBT). Our database consists of metadata extracted from 334,234 images. We analyzed distributions and correlations of compressed breast thickness (CBT), compression force, target-filter combinations, X-ray tube voltage, and average glandular dose (AGD). AGD reference curves were calculated based on AGD distributions as a function of CBT. These curves represent an AGD reference for a particular population and system. Differences in AGD and imaging settings were attributed to a combination of factors, such as improvements in technology, imaging protocol, and patient demographics. The tracking system allows the comparison of various imaging settings used in screening mammography, as well as the tracking of patient- and population-specific breast data collected from different populations.

Technical Note: Noise models for virtual clinical trials of digital breast tomosynthesis.

PURPOSE: To investigate the use of an affine-variance noise model, with correlated quantum noise and spatially dependent quantum gain, for the simulation of noise in virtual clinical trials (VCT) of digital breast tomosynthesis (DBT). METHODS: Two distinct technologies were considered: an amorphous-selenium (a-Se) detector with direct conversion and a thallium-doped cesium iodide (CsI(Tl)) detector with indirect conversion. A VCT framework was used to generate noise-free projections of a uniform three-dimensional simulated phantom, whose geometry and absorption match those of a polymethyl methacrylate (PMMA) uniform physical phantom. The noise model was then used to generate noisy observations from the simulated noise-free data, while two clinically available DBT units were used to acquire projections of the PMMA physical phantom. Real and simulated projections were then compared using the signal-to-noise ratio (SNR) and normalized noise power spectrum (NNPS). RESULTS: Simulated images reported errors smaller than 4.4% and 7.0% in terms of SNR and NNPS, respectively. These errors are within the expected variation between two clinical units of the same model. The errors increase to 65.8% if uncorrelated models are adopted for the simulation of systems featuring indirect detection. The assumption of spatially independent quantum gain generates errors of 11.2%. CONCLUSIONS: The investigated noise model can be used to accurately reproduce the noise found in clinical DBT. The assumption of uncorrelated noise may be adopted if the system features a direct detector with minimal pixel crosstalk.

Evaluation of denoising digital breast tomosynthesis data in both projection and image domains and a study of noise model on digital breast tomosynthesis image domain.

Digital breast tomosynthesis (DBT) is an imaging technique created to visualize 3-D mammary structures for the purpose of diagnosing breast cancer. This imaging technique is based on the principle of computed tomography. Due to the use of a dangerous ionizing radiation, the "as low as reasonably achievable" (ALARA) principle should be respected, aiming at minimizing the radiation dose to obtain an adequate examination. Thus, a noise filtering method is a fundamental step to achieve the ALARA principle, as the noise level of the image increases as the radiation dose is reduced, making it difficult to analyze the image. In our work, a double denoising approach for DBT is proposed, filtering in both projection (prereconstruction) and image (postreconstruction) domains. First, in the prefiltering step, methods were used for filtering the Poisson noise. To reconstruct the DBT projections, we used the filtered backprojection algorithm. Then, in the postfiltering step, methods were used for filtering Gaussian noise. Experiments were performed on simulated data generated by open virtual clinical trials (OpenVCT) software and on a physical phantom, using several combinations of methods in each domain. Our results showed that double filtering (i.e., in both domains) is not superior to filtering in projection domain only. By investigating the possible reason to explain these results, it was found that the noise model in DBT image domain could be better modeled by a Burr distribution than a Gaussian distribution. Finally, this important contribution can open a research direction in the DBT denoising problem.

Comparison of mean glandular dose values provided by a digital breast tomosynthesis system in Brazil.

Studies are needed to determine the radiation dose of patients that are undergoing Digital breast tomosynthesis (DBT) procedures. Mean glandular dose (DG) values were derived from the incident air kerma (Ki) measurements and tabulated conversion coefficients. Ki values were obtained through an ionization chamber positioned in a Hologic Selenia Dimensions system using appropriate exposure parameters. This work contributes to determine the reliable radiation dose received by the patients and compare DG values provided by this DBT system images.