Synthetic 2D Images Research Articles

Digital breast tomosynthesis in the diagnostic settings of breast cancer patients treated by conservation therapy

BackgroundBreast cancer survivors who underwent breast conserving therapy (BCT) are still 10% more likely to develop a second breast cancer at follow-up, Digital mammography (DM) was advised in every practical guideline for follow-up after BCT; however, it was difficult to distinguish between actual recurrence and scar development at DM.Our research objective is to assess the value of Digital Breast Tomosynthesis (DBT) on the final Breast Imaging Reporting and Data System lexicon (BI-RADS) assessment categories compared to diagnostic digital mammography (DM) of breast cancer patients treated with breast conservation therapy (BCT), and to determine the additive value of DBT to DM or the use of DBT with synthetic 2D images in the diagnostic workup following BCT.ResultsFour hundred and seventeen breast cancer patients who underwent breast-conserving therapy (BCT) and received diagnostic assessments, including digital mammography (DM), digital breast tomosynthesis (DBT), and reconstructed synthetic 2D images, were enrolled in the study.There is a significant reduction in the proportion of studies classified as probably benign BI-RADS 3 in synthetic 2D + DBT and DM + DBT compared with those in DM alone, and this was demonstrated by the two readers and at double reading (all P = 0.01). There is a significant increase in the PPV of malignancy in synthetic 2D + DBT and DM + DBT compared with those in DM alone for reader 2 (P = 0.01) and at double reading (P < 0.04) without a significant difference in the proportion of studies classified as BI-RADS category 4 or 5 for both readers and at double reading (p > 0.3).Regarding the mammographic abnormalities, we noted a significant reduction in asymmetry and a significant increase in architectural distortion in synthetic 2D + DBT and DM + DBT compared with those in DM alone, and this was demonstrated by the two readers and at double reading (all P = 0.01).ConclusionsThe addition of digital breast tomosynthesis (DBT) into the diagnostic process after breast-conserving therapy (BCT), either in conjunction with digital mammography (DM) or with synthetic 2D images in the diagnostic workup following BCT, significantly decreases the proportion of studies classified as probably benign, significantly increases the rate of studies classified as normal or benign, and significantly increase in the positive predictive value (PPV) of malignancy without significant difference in the patients classified as BI-RADS category 4 or 5. Moreover it improved diagnostic confidence in biopsy recommendations, regardless of mammographic density.

Synthetic Training Data in AI-Driven Quality Inspection: The Significance of Camera, Lighting, and Noise Parameters.

Industrial-quality inspections, particularly those leveraging AI, require significant amounts of training data. In fields like injection molding, producing a multitude of defective parts for such data poses environmental and financial challenges. Synthetic training data emerge as a potential solution to address these concerns. Although the creation of realistic synthetic 2D images from 3D models of injection-molded parts involves numerous rendering parameters, the current literature on the generation and application of synthetic data in industrial-quality inspection scarcely addresses the impact of these parameters on AI efficacy. In this study, we delve into some of these key parameters, such as camera position, lighting, and computational noise, to gauge their effect on AI performance. By utilizing Blender software, we procedurally introduced the "flash" defect on a 3D model sourced from a CAD file of an injection-molded part. Subsequently, with Blender's Cycles rendering engine, we produced datasets for each parameter variation. These datasets were then used to train a pre-trained EfficientNet-V2 for the binary classification of the "flash" defect. Our results indicate that while noise is less critical, using a range of noise levels in training can benefit model adaptability and efficiency. Variability in camera positioning and lighting conditions was found to be more significant, enhancing model performance even when real-world conditions mirror the controlled synthetic environment. These findings suggest that incorporating diverse lighting and camera dynamics is beneficial for AI applications, regardless of the consistency in real-world operational settings.

Breast density analysis of digital breast tomosynthesis

Mammography shifted to digital breast tomosynthesis (DBT) in the US. An automated percentage of breast density (PD) technique designed for two-dimensional (2D) applications was evaluated with DBT using several breast cancer risk prediction measures: normalized-volumetric; dense volume; applied to the volume slices and averaged (slice-mean); and applied to synthetic 2D images. Volumetric measures were derived theoretically. PD was modeled as a function of compressed breast thickness (CBT). The mean and standard deviation of the pixel values were investigated. A matched case–control (CC) study (n = 426 pairs) was evaluated. Odd ratios (ORs) were estimated with 95% confidence intervals. ORs were significant for PD: identical for volumetric and slice-mean measures [OR = 1.43 (1.18, 1.72)] and [OR = 1.44 (1.18, 1.75)] for synthetic images. A 2nd degree polynomial (concave-down) was used to model PD as a function of CBT: location of the maximum PD value was similar across CCs, occurring at 0.41 × CBT, and PD was significant [OR = 1.47 (1.21, 1.78)]. The means from the volume and synthetic images were also significant [ORs ~ 1.31 (1.09, 1.57)]. An alternative standardized 2D synthetic image was constructed, where each pixel value represents the percentage of breast density above its location. Several measures were significant and an alternative method for constructing a standardized 2D synthetic image was produced.

Diagnostic accuracy of automated ACR BI-RADS breast density classification using deep convolutional neural networks

ObjectivesHigh breast density is a well-known risk factor for breast cancer. This study aimed to develop and adapt two (MLO, CC) deep convolutional neural networks (DCNN) for automatic breast density classification on synthetic 2D tomosynthesis reconstructions.MethodsIn total, 4605 synthetic 2D images (1665 patients, age: 57 ± 37 years) were labeled according to the ACR (American College of Radiology) density (A-D). Two DCNNs with 11 convolutional layers and 3 fully connected layers each, were trained with 70% of the data, whereas 20% was used for validation. The remaining 10% were used as a separate test dataset with 460 images (380 patients). All mammograms in the test dataset were read blinded by two radiologists (reader 1 with two and reader 2 with 11 years of dedicated mammographic experience in breast imaging), and the consensus was formed as the reference standard. The inter- and intra-reader reliabilities were assessed by calculating Cohen’s kappa coefficients, and diagnostic accuracy measures of automated classification were evaluated.ResultsThe two models for MLO and CC projections had a mean sensitivity of 80.4% (95%-CI 72.2–86.9), a specificity of 89.3% (95%-CI 85.4–92.3), and an accuracy of 89.6% (95%-CI 88.1–90.9) in the differentiation between ACR A/B and ACR C/D. DCNN versus human and inter-reader agreement were both “substantial” (Cohen’s kappa: 0.61 versus 0.63).ConclusionThe DCNN allows accurate, standardized, and observer-independent classification of breast density based on the ACR BI-RADS system.Key Points• A DCNN performs on par with human experts in breast density assessment for synthetic 2D tomosynthesis reconstructions.• The proposed technique may be useful for accurate, standardized, and observer-independent breast density evaluation of tomosynthesis.

Difference-of-Gaussian generative adversarial network for segmenting breast arterial calcifications in mammograms

Breast arterial calcifications (BACs) are amongst the different types of benign calcifications observed on mammograms. BACs have been found to correlate with cardiovascular risk factors, cardiovascular mortality, and coronary artery disease (CAD). Considering that women are recommended to undergo routine screening mammography for the early detection of breast cancer, identifying BACs on mammograms could help identify women at risk of cardiovascular diseases (CVD) without the additional cost or radiation of other tests, such as coronary artery computed tomography (CT). In this paper, we present a difference of Gaussian generative adversarial network (DoG-GAN) model for segmenting BACs in mammograms. It combines the multi-scale difference of Gaussian (DoG) pyramid with the U-net as a generator of the GAN. This approach allows the model to explore image details at each scale and exploit the edge information to effectively segment BACs. We evaluated the performance of our model on a set of synthetic 2D images from digital breast tomosynthesis exams (DBT) collected and prepared for this task. The experimental results show that our model outperforms the state-of-the-art methods.

The EMory BrEast imaging Dataset (EMBED): A Racially Diverse, Granular Dataset of 3.4 Million Screening and Diagnostic Mammographic Images.

Supplemental material is available for this article. Keywords: Mammography, Breast, Machine Learning © RSNA, 2023.

2D or Synthetic 2D? A Reader Study of Visualization of Amorphous Calcifications.

Some vendors have created algorithms that generate synthetic 2D (s2D) images from a digital breast tomosynthesis (DBT) dataset to reduce the radiation from obtaining a separate 2D digital mammography (DM). This study evaluated the visibility of amorphous calcifications on 2D DM versus s2D on screening mammography. This IRB-approved, retrospective, reader study included screening mammograms from 36 women who received screening DBT exams where both 2D DM and s2D images were obtained: 28 screening mammograms that were eventually given BI-RADS category 4 or 5 for amorphous calcifications and 8 BI-RADS category 1 or 2 screening exams. Two rounds of interpretation were conducted with a six-week washout period. Cases were randomized to display either the 2D DM or s2D images, which were then alternated in the second round. Four fellowship-trained breast radiologists determined whether a study merited recall for calcifications. If so, they rated calcification visibility on a scale of 1 to 5. McNemar chi-square tests were conducted to assess differences in recall rates and Wilcoxon signed rank tests were used to examine shifts in visibility. There was no difference in detection rates of amorphous calcifications between 2D DM and s2D, which were 75.9% and 75.0%, respectively (P = 1.000). Collectively, amorphous calcifications were more visible on s2D than 2D DM, with mean visibility scores of 3.4 versus 3.0, respectively (P = 0.005). Synthetic 2D did not change identification of amorphous calcifications compared to 2D DM, and readers considered them more visible on average.

OD123 - Small-size details detection performance of digital breast tomosynthesis, synthetic 2D and conventional full-field digital mammography images for different mammography systems: a multicenter study

OD123 - Small-size details detection performance of digital breast tomosynthesis, synthetic 2D and conventional full-field digital mammography images for different mammography systems: a multicenter study

Power Spectrum Analysis of Breast Parenchyma with Digital Breast Tomosynthesis Images in a Longitudinal Screening Cohort from Two Vendors

To quantitatively compare breast parenchymal texture between two Digital Breast Tomosynthesis (DBT) vendors using images from the same patients. This retrospective study included consecutive patients who had normal screening DBT exams performed in January 2018 from GE and normal screening DBT exams in adjacent years from Hologic. Power spectrum analysis was performed within the breast tissue region. The slope of a linear function between log-frequency and log-power, β, was derived as a quantitative measure of breast texture and compared within and across vendors along with secondary parameters (laterality, view, year, image format, and breast density) with correlation tests and t-tests. A total of 24,339 DBT slices or synthetic 2D images from 85 exams in 25 women were analyzed. Strong power-law behavior was verified from all images. Values of β d did not differ significantly for laterality, view, or year. Significant differences of β were observed across vendors for DBT images (Hologic: 3.4±0.2 vs GE: 3.1±0.2, 95% CI on difference: 0.27 to 0.30) and synthetic 2D images (Hologic: 2.7±0.3 vs GE: 3.0±0.2, 95% CI on difference: -0.36 to -0.27), and density groups with each vendor: scattered (GE: 3.0±0.3, Hologic: 3.3±0.3) vs. heterogeneous (GE: 3.2±0.2, Hologic: 3.4±0.1), 95% CI (-0.27, -0.08) and (-0.21, -0.05), respectively. There are quantitative differences in the presentation of breast imaging texture between DBT vendors and across breast density categories. Our findings have relevance and importance for development and optimization of AI algorithms related to breast density assessment and cancer detection.

Effect of glandularity on the detection of simulated cancers in planar, tomosynthesis, and synthetic 2D imaging of the breast using a hybrid virtual clinical trial.

The purpose of this study was to measure the threshold diameter of calcifications and masses for 2D imaging, digital breast tomosynthesis (DBT), and synthetic 2D images, for a range of breast glandularities. This study shows the limits of detection for each of the technologies and the strengths and weaknesses of each in terms of visualizing the radiological features of small cancers. Mathematical voxel breast phantoms with glandularities by volume of 9%, 18%, and 30% with a thickness of 53mm were created. Simulated ill-defined masses and calcification clusters with a range of diameters were inserted into some of these breast models. The imaging characteristics of a Siemens Inspiration X-ray system were measured for a 29kV, tungsten/rhodium anode/filter combination. Ray tracing through the breast models was undertaken to create simulated 2D and DBT projection images. These were then modified to adjust the image sharpness, and to add scatter and noise. The mean glandular doses for the images were 1.43, 1.47, and 1.47 mGy for 2D and 1.92, 1.97, and 1.98 mGy for DBT for the three glandularities. The resultant images were processed to create 2D, DBT planes and synthetic 2D images. Patches of the images with or without a simulated lesion were extracted, and used in a four-alternative forced choice study to measure the threshold diameters for each imaging mode, lesion type, and glandularity. The study was undertaken by six physicists. The threshold diameters of the lesions were 6.2, 4.9, and 6.7mm (masses) and 225, 370, and 399μm, (calcifications) for 2D, DBT, and synthetic 2D, respectively, for a breast glandularity of 18%. The threshold diameter of ill-defined masses is significantly smaller for DBT than for both 2D (p≤0.006) and synthetic 2D (p≤0.012) for all glandularities. Glandularity has a significant effect on the threshold diameter of masses, even for DBT where there is reduced background structure in the images. The calcification threshold diameters for 2D images were significantly smaller than for DBT and synthetic 2D for all glandularities. There were few significant differences for the threshold diameter of calcifications between glandularities, indicating that the background structure has little effect on the detection of calcifications. We measured larger but nonsignificant differences in the threshold diameters for synthetic 2D imaging than for 2D imaging for masses in the 9% (p=0.059) and 18% (p=0.19) glandularities. The threshold diameters for synthetic 2D imaging were larger than for 2D imaging for calcifications (p<0.001) for all glandularities. We have shown that glandularity has only a small effect on the detection of calcifications, but the threshold diameter of masses was significantly larger for higher glandularity for all of the modalities tested. We measured nonsignificantly larger threshold diameters for synthetic 2D imaging than for 2D imaging for masses at the 9% (p=0.059) and 18% (p=0.19) glandularities and significantly larger diameters for calcifications (p<0.001) for all glandularities. The lesions simulated were very subtle and further work is required to examine the clinical effect of not seeing the smallest calcifications in clusters.

Virtual clinical trial to compare cancer detection using combinations of 2D mammography, digital breast tomosynthesis and synthetic 2D imaging

This study was designed to compare the detection of subtle lesions (calcification clusters or masses) when using the combination of digital breast tomosynthesis (DBT) and synthetic mammography (SM) with digital mammography (DM) alone or combined with DBT. A set of 166 cases without cancer was acquired on a DBT mammography system. Realistic subtle calcification clusters and masses in the DM images and DBT planes were digitally inserted into 104 of the acquired cases. Three study arms were created: DM alone, DM with DBT and SM with DBT. Five mammographic readers located the centre of any lesion within the images that should be recalled for further investigation and graded their suspiciousness. A JAFROC figure of merit (FoM) and lesion detection fraction (LDF) were calculated for each study arm. The visibility of the lesions in the DBT images was compared with SM and DM images. For calcification clusters, there were no significant differences (p > 0.075) in FoM or LDF. For masses, the FoM and LDF were significantly improved in the arms using DBT compared to DM alone (p < 0.001). On average, both calcification clusters and masses were more visible on DBT than on DM and SM images. This study demonstrated that masses were detected better with DBT than with DM alone and there was no significant difference (p = 0.075) in LDF between DM&DBT and SM&DBT for calcifications clusters. Our results support previous studies that it may be acceptable to not acquire digital mammography alongside tomosynthesis for subtle calcification clusters and ill-defined masses. • The detection of masses was significantly better using DBT than with digital mammography alone. • The detection of calcification clusters was not significantly different between digital mammography and synthetic 2D images combined with tomosynthesis. • Our results support previous studies that it may be acceptable to not acquire digital mammography alongside tomosynthesis for subtle calcification clusters and ill-defined masses for the imaging technology used.

Evaluation of tightly- and loosely-coupled approaches in CNN-based pose estimation systems for uncooperative spacecraft

The relative pose estimation of an inactive spacecraft by an active servicer spacecraft is a critical task in the design of current and planned space missions, due to its relevance for close-proximity operations, such as In-Orbit Servicing and Active Debris Removal. This paper introduces a novel framework to enable robust monocular pose estimation for close-proximity operations around an uncooperative spacecraft, which combines a Convolutional Neural Network (CNN) for feature detection with a Covariant Efficient Procrustes Perspective-n-Points (CEPPnP) solver and a Multiplicative Extended Kalman Filter (MEKF). The performance of the proposed method is evaluated at different levels of the pose estimation system. A Single-stack Hourglass CNN is proposed for the feature detection step in order to decrease the computational load of the Image Processing (IP), and its accuracy is compared to the standard, more complex High-Resolution Net (HRNet). Subsequently, heatmaps-derived covariance matrices are included in the CEPPnP solver to assess the pose estimation accuracy prior to the navigation filter. This is done in order to support the performance evaluation of the proposed tightly-coupled approach against a loosely-coupled approach, in which the detected features are converted into pseudomeasurements of the relative pose prior to the filter. The performance results of the proposed system indicate that a tightly-coupled approach can guarantee an advantageous coupling between the rotational and translational states within the filter, whilst reflecting a representative measurements covariance. This suggest a promising scheme to cope with the challenging demand for robust navigation in close-proximity scenarios. Synthetic 2D images of the European Space Agency’s Envisat spacecraft are used to generate datasets for training, validation and testing of the CNN. Likewise, the images are used to recreate a representative close-proximity scenario for the validation of the proposed filter.

Keeping Pace With Technology Advances in Breast Cancer Screening: Synthetic 2D Images Outperform Digital Mammography.

Keeping Pace With Technology Advances in Breast Cancer Screening: Synthetic 2D Images Outperform Digital Mammography.

Comparing recall rates following implementation of digital breast tomosynthesis to synthetic 2D images and digital mammography on women with breast-conserving surgery.

To evaluate the recall rates of digital mammography (DM) and synthetic images after adding digital breast tomosynthesis (DBT) in patients with breast-conserving surgery. From November 2015 to April 2017, 229 women with breast-conserving surgery due to breast cancer who underwent DBT after surgery were included (mean interval, 12.9 ± 1.4 months). All women underwent combo-mode DBT examinations including full-field DM, tomosynthesis, and reconstructed synthetic 2D images. Three board-certified breast radiologists reviewed the images sequentially: synthetic 2D+DBT and, 1 month later, DM and then DM+DBT. Recall rates and the abnormality type causing the recall were calculated and compared for each mammographic modality and breast density. Of the 229 patients included, 230 mammography images were reviewed. One patient (0.4%) developed locoregional recurrences during follow-up (mean duration, 25.8 ± 4.5 months). Recall rates for synthetic 2D+DBT were significantly lower than for DM alone (4.1% (2.6-6.2) vs. 11.6% (9.2-14.5), respectively; p < 0.001). Recall rates did not differ between synthetic 2D+DBT and DM+DBT (4.1% (2.6-6.2) vs. 2.9% (1.9-4.5), respectively; p = 0.234). Recall rates of synthetic 2D+DBT and DM+DBT were significantly lower than those of DM alone, regardless of mammographic breast density (all p < 0.05, respectively). Adding DBT to synthetic 2D images or DM shows significant reduction in recall rates compared with DM alone for women who undergo breast-conserving surgery for breast cancer, regardless of mammographic density. • Recall rates for synthetic 2D+DBT were significantly lower than those of DM alone (4.1% (2.6-6.2) vs. 11.6% (9.2-14.5), respectively; p < 0.001). • No significant differences were seen in recall rates between synthetic 2D+DBT and DM+DBT (4.1 (2.6-6.2) vs. 2.9 (1.9-4.5), respectively; p = 0.234). • Reader-averaged recall rates after adding DBT to synthetic 2D or DM were significantly lower than those of DM alone, regardless of mammographic breast density (all p < 0.05, respectively).

A comparison of full-field digital mammograms versus 2D synthesized mammograms for detection of microcalcifications on screening

ObjectivesTo compare the observer agreement of microcalcification detection on synthetic 2D images to full field digital mammography (FFDM) at screening and determine if calcifications can be detected to the same degree and given the same BI-RADS assessment. Material and methodsTwo-experienced radiologists retrospectively reviewed synthetic 2D images and FFDM, in separate sessions, to detect microcalcifications and provide a BIRADS assessment. A third experienced breast radiologist reviewed the cases that were disagreed upon and gave a final assessment.Between March 2016–December 2016, 414 women obtained a screening mammogram with tomosynthesis and acquisition of FFDM. 71 cases had combined FFDM and tomosynthesis images, calcifications visible on FFDM or no calcifications present, which comprised the study group. Synthetic 2D images were constructed from the DBT images. During session 1, all synthetic 2D images for the 71 cases were reviewed. During session 2, all the FFDM images for the 71 cases were reviewed. Tomosynthesis images were not reviewed. The agreement of detection of calcifications and BIRADS assessment between radiologists for FFDM and synthetic 2D images were assessed using Cohen’s kappa test. Fisher’s exact test was used to detect the differences in calcification identification among various breast densities on FFDM and synthetic 2D images. ResultsFor the detection of calcifications between synthetic 2D images and FFDM, there was moderate to substantial agreement (p-values < 0.0001) for the two radiologist. For the BIRADS assessments, the agreement between synthetic 2D imaging and FFDM was moderate (p-values < 0.0001). The inter-reader agreement for detection of calcifications was fair for using synthetic 2D and moderate for using FFDM (p-value < 0.0001). The final inter-reader agreement between FFDM and synthetic 2D images for the detection of calcifications was moderate (p-values < 0.0001) with the addition of the third reader. For the final BI-RADS assessment, there was moderate agreement between synthetic 2D imaging and FFDM (p-value < 0.0001). The two readers did not demonstrate a significant difference in the detection of microcalcifications for those who were dense or non-dense (p-value range 0.076–0.302). ConclusionRadiologist interpreting synthetic 2D imaging and FFDM have similar frequency for detection of calcifications and BIRADS assessment. A synthetic 2D mammogram may be a sufficient replacement for FFDM at screening.

OA211] A single phantom, a single statistical method, a single figure of merit for all X-ray equipment

Purpose The aim of this study is to collect Statistical Low Contrast Detectability Threshold (LCth) data from different equipment using the same phantom and the same software for the analysis. All data will be expressed with same unit, that is mmAl. Equipment types include: Digital radiography (DR), Computed Radiography (CR), 2D Digital Mammography, Tomosynthesis and Syntetic 2D Mammography, Fluoroscopy, Fluorography and Digital Subtracted Angiography (DSA) from Angiography. Data from this study will define the reference value and the tolerance for quality control tests. Methods and materials A phantom consisting of a square aluminium target and a 5-step-wedge (0.25–1.25 mmAl) was built. A Matlab software automatically calculates LCth values in terms of Pixel Value (PV) and it also converts PV to mmAl, providing the LCth(mmAl). Standard setting used for different modality: RQA5 beam at 2.5 μGy kerma entrance at the detector for DR and CR; automatic acquisition with 16 cm of PMMA filtration for angiography and 5 cm PMMA phantom for mammography setting. The images analysed in this study are both processed and unprocessed. Results This method is able to collect low contrast detectability information from different modalities and equipment by different vendors. Preliminary results are summarized by detail. For example, for 0.5 mm diameter detail results in the range of: 0.05–0.08, 0.04–0.093, 0.018–0.023, 0.26–0.3 and 0.36–0.41 mmAl LCth for tomosynthesis images, synthetic 2D, 2D mammography, DR and CR images, respectively. Furthermore, from the intercomparison activity of AIFM working group on angiographic equipment, for 2.5 mm diameter detail, results have been derived in the range of: 0.35–0.88, 0.31–0.58, 0.35–0.49, 0.046–0.59 LCth expressed in mmAl unit for low dose and normal dose fluoroscopy, coronary fluorography, cerebral DSA, respectively. Conclusion The Low Contrast Detectability phantom and statistical method can be easily implemented for all modalities. Data from this study can be used to define the reference value and the tolerance for low contrast threshold quality control in digital radiography.

A Method for Generation of Synthetic 2D and 3D Cryo-EM Images

Cryo-electron microscopy (cryo-EM) is widely used in structural biology for resolving 3D models of particles with Angstrom resolution. The most popular techniques for such high-resolution model reconstruction are single-particle cryo-EM and cryo-electron tomography (cryo-ET). The cornerstone of both techniques is the registration of images: 2D images in cryo-EM and 3D images in cryo-ET. There are several registration methods for 2D and 3D cryo-EM images; however, it is hard to evaluate these methods due to the lack of ground truth for real data. Moreover, evaluation of image registration methods on real data is complicated by a high level of noise. In this paper, we propose image synthesis methods for generating realistic 2D single-particle cryo-EM images and 3D cryo-ET subtomogram images. The proposed algorithms model the artifacts typical of the real EM image acquisition pipeline: EM-specific noise, missing wedge effect, 2D projection, and contrast transfer function. We also present some examples of the 2D and 3D synthetic images generated.

A Method for Generation of Synthetic 2D and 3D Cryo-EM Images

A Method for Generation of Synthetic 2D and 3D Cryo-EM Images

Radiation dose with digital breast tomosynthesis compared to digital mammography: per-view analysis.

To compare radiation dose delivered by digital mammography (FFDM) and breast tomosynthesis (DBT) for a single view. 4,780 FFDM and 4,798 DBT images from 1,208 women enrolled in a screening trial were used to ground dose comparison. Raw images were processed by an automatic software to determine volumetric breast density (VBD) and were used together with exposure data to compute the mean glandular dose (MGD) according to Dance's model. DBT and FFDM were compared in terms of operation of the automatic exposure control (AEC) and MGD level. Statistically significant differences were found between FFDM and DBT MGDs for all views (CC: MGDFFDM=1.366 mGy, MGDDBT=1.858 mGy; p<0.0001; MLO: MGDFFDM=1.374 mGy, MGDDBT=1.877 mGy; p<0.0001). Considering the 4,768 paired views, Bland-Altman analysis showed that the average increase of DBT dose compared to FFDM is 38 %, and a range between 0 % and 75 %. Our findings show a modest increase of radiation dose to the breast by tomosynthesis compared to FFDM. Given the emerging role of DBT, its use in conjunction with synthetic 2D images should not be deterred by concerns regarding radiation burden, and should draw on evidence of potential clinical benefit. • Most studies compared tomosynthesis in combination with mammography vs. mammography alone. • There is some concern about the dose increase with tomosynthesis. • Clinical data show a small increase in radiation dose with tomosynthesis. • Synthetic 2D images from tomosynthesis at zero dose reduce potential harm. • The small dose increase should not be a barrier to use of tomosynthesis.

Breast cancer screening with tomosynthesis (3D mammography) with acquired or synthetic 2D mammography compared with 2D mammography alone (STORM-2): a population-based prospective study

Breast tomosynthesis (pseudo-3D mammography) improves breast cancer detection when added to 2D mammography. In this study, we examined whether integrating 3D mammography with either standard 2D mammography acquisitions or with synthetic 2D images (reconstructed from 3D mammography) would detect more cases of breast cancer than 2D mammography alone, to potentially reduce the radiation burden from the combination of 2D plus 3D acquisitions. The Screening with Tomosynthesis Or standard Mammography-2 (STORM-2) study was a prospective population-based screening study comparing integrated 3D mammography (dual-acquisition 2D-3D mammography or 2D synthetic-3D mammography) with 2D mammography alone. Asymptomatic women aged 49 years or older who attended population-based screening in Trento, Italy were recruited for the study. All participants underwent digital mammography with 2D and 3D mammography acquisitions, with the use of software that allowed synthetic 2D mammographic images to be reconstructed from 3D acquisitions. Mammography screen-reading was done in two parallel double-readings conducted sequentially for 2D acquisitions followed by integrated acquisitions. Recall based on a positive mammography result was defined as recall at any screen read. Primary outcome measures were a comparison between integrated (2D-3D or 2D synthetic-3D) mammography and 2D mammography alone of the number of cases of screen-detected breast cancer, the cancer detection rate per 1000 screens, the incremental cancer detection rate, and the number and percentage of false-positive recalls. Between May 31, 2013, and May 29, 2015, 10 255 women were invited to participate, of whom 9672 agreed to participate and were screened. In these 9672 participants (median age 58 years [IQR 53-63]), screening detected 90 cases of breast cancer, including 74 invasive breast cancers, in 85 women (five women had bilateral breast cancer). To account for these bilateral cancers in cancer detection rate estimates, the number of screens used for analysis was 9677. Both 2D-3D mammography (cancer detection rate 8·5 per 1000 screens [82 cancers detected in 9677 screens]; 95% CI 6·7-10·5) and 2D synthetic-3D mammography (8·8 per 1000 [85 in 9677]; 7·0-10·8) had significantly higher rates of breast cancer detection than 2D mammography alone (6·3 per 1000 [61 in 9677], 4·8-8·1; p<0·0001 for both comparisons). The cancer detection rate did not differ significantly between 2D-3D mammography and 2D synthetic-3D mammography (p=0·58). Compared with 2D mammography alone, the incremental cancer detection rate from 2D-3D mammography was 2·2 per 1000 screens (95% CI 1·2-3·3) and that from 2D synthetic-3D mammography was 2·5 per 1000 (1·4-3·8). Compared with the proportion of false-positive recalls from 2D mammography alone (328 of 9587 participants not found to have cancer at assessment) [3·42%; 95% CI 3·07-3·80]), false-positive recall was significantly higher for 2D-3D mammography (381 of 9587 [3·97%; 3·59-4·38], p=0·00063) and for 2D synthetic-3D mammography (427 of 9587 [4·45%; 4·05-4·89], p<0·0001). Integration of 3D mammography (2D-3D or 2D synthetic-3D) detected more cases of breast cancer than 2D mammography alone, but increased the percentage of false-positive recalls in sequential screen-reading. These results should be considered in the context of the trade-off between benefits and harms inherent in population breast cancer screening, including that significantly increased breast cancer detection from integrating 3D mammography into screening has the potential to augment screening benefit and also possibly contribute to overdiagnosis. None.