Differences in lesion interpretation between radiologists in two countries: Lessons from a digital breast tomosynthesis training test set.

Pathological outcome of sonographically occult architectural distortions (AD) visible only on digital breast tomosynthesis, and comparison with AD visible also on 2D mammography

We propose computer-aided detection (CADe) algorithm for microcalcification (MC) clusters in reconstructed digital breast tomosynthesis (DBT) images. The algorithm consists of prescreening, MC detection, clustering, and false-positive (FP) reduction steps. The DBT images containing the MC-like objects were enhanced by a multiscale Hessian-based three-dimensional (3D) objectness response function and a connected-component segmentation method was applied to extract the cluster seed objects as potential clustering centers of MCs. Secondly, a signal-to-noise ratio (SNR) enhanced image was also generated to detect the individual MC candidates and prescreen the MC-like objects. Each cluster seed candidate was prescreened by counting neighboring individual MC candidates nearby the cluster seed object according to several microcalcification clustering criteria. As a second step, we introduced bounding boxes for the accepted seed candidate, clustered all the overlapping cubes, and examined. After the FP reduction step, the average number of FPs per case was estimated to be 2.47 per DBT volume with a sensitivity of 83.3%.

P.698 Relationship between psychosocial functioning and quality of life of schizophrenic patients treated with second generation long acting injectable antipsychotics

Architectural distortion (AD) is one of the breast abnormal signs in digital breast tomosynthesis (DBT) and digital mammography (DM). It is hard to be detected because of its subtle appearance and similar intensity with surrounding tissue. Since DBT is a three-dimensional imaging, it can address the problem of tissue superimposition in DM, so as to reduce false positives and false negatives. Several clinical studies have confirmed that radiologists can detect more ADs in DBT than in DM. These conclusions are based on subjective experience. To explore whether the engineering model and the experience of radiologists are consistent in AD detection tasks, this study compared the AD detection performance of a deep-learning-based computer-aided detection (CADe) model in DBT and DM images of the same group of cases. 394 DBT volumes and their corresponding DM images were collected retrospectively from 99 breast cancer screening cases. Among them, 203 DBT volumes and DM images contained ADs and the remaining 191 ones were negative group without any AD. Ten-fold cross-validation was used to train and evaluate the models and mean true positive fraction (MTPF) was used as figure-of-merit. The results showed that the CADe model achieved significantly better detection performance in DBT than DM (MTPF: 0.7026±0.0394 for DBT vs. 0.5870±0.0407 for DM, p=0.002). Qualitative analysis illustrated that DBT indeed had the ability to overcome tissue superimposition and showed more details of breast tissue. It helped the CADe model detect more ADs, which was consistent with clinical experience.

Purpose To examine the outcomes of a breast cancer screening program based on digital breast tomosynthesis (DBT) plus synthesized two-dimensional (2D) mammography compared with those after full-field digital mammography (FFDM). Materials and Methods This prospective study included 16 666 asymptomatic women aged 50-69 years who were recruited in April 2015 through March 2016 for DBT plus synthetic 2D screening in the Verona screening program. A comparison cohort of women screened with FFDM (n = 14 423) in the previous year was included. Screening detection measures for the two groups were compared by calculating the proportions associated with each outcome, and the relative rates (RRs) were estimated with multivariate logistic regression. Results Cancer detection rate (CDR) for DBT plus synthetic 2D imaging was 9.30 per 1000 screening examinations versus 5.41 per 1000 screening examinations with FFDM (RR, 1.72; 95% confidence interval [CI]: 1.30, 2.29). CDR was significantly higher in patients screened with DBT plus synthetic 2D imaging than in those screened with FFDM among women classified as having low breast density (RR, 1.53; 95% CI: 1.13, 2.10) or high breast density (RR, 2.86; 95% CI: 1.42, 6.25). The positive predictive value (PPV) for recall was almost doubled with DBT plus synthetic 2D imaging: 23.3% versus 12.9% of recalled patients who were screened with FFDM (RR, 1.81; 95% CI: 1.34, 2.47). The recall rate was similar between groups (RR, 0.95; 95% CI: 0.84, 1.06), whereas the recall rate with invasive assessment was higher for DBT plus synthetic 2D imaging than for FFDM (RR, 1.93; 95% CI: 1.31, 2.03). The mean number of screening studies interpreted per hour was significantly lower for screening examinations performed with DBT plus synthetic 2D imaging (38.5 screens per hour) than with FFDM (60 screens per hour) (P < .001). Conclusion DBT plus synthetic 2D imaging increases CDRs with recall rates comparable to those of FFDM. DBT plus synthetic 2D imaging increased image reading time and the time needed for invasive assessments. © RSNA, 2017.

To compare the diagnostic performance of two-dimensional (2D) mammography, 2D mammography plus digital breast tomosynthesis (DBT), and synthetic 2D mammography plus DBT in depicting malignant radiographic features. In this multicenter, multireader, retrospective reading study (the TOMMY trial), after written informed consent was obtained, 8869 women (age range, 29-85 years; mean, 56 years) were recruited from July 2011 to March 2013 in an ethically approved study. From these women, a reading dataset of 7060 cases was randomly allocated for independent blinded review of (a) 2D mammography images, (b) 2D mammography plus DBT images, and (c) synthetic 2D mammography plus DBT images. Reviewers had no access to results of previous examinations. Overall sensitivities and specificities were calculated for younger women and those with dense breasts. Overall sensitivity was 87% for 2D mammography, 89% for 2D mammography plus DBT, and 88% for synthetic 2D mammography plus DBT. The addition of DBT was associated with a 34% increase in the odds of depicting cancer (odds ratio [OR] = 1.34, P = .06); however, this level did not achieve significance. For patients aged 50-59 years old, sensitivity was significantly higher (P = .01) for 2D mammography plus DBT than it was for 2D mammography. For those with breast density of 50% or more, sensitivity was 86% for 2D mammography compared with 93% for 2D mammography plus DBT (P = .03). Specificity was 57% for 2D mammography, 70% for 2D mammography plus DBT, and 72% for synthetic 2D mammography plusmDBT. Specificity was significantly higher than 2D mammography (P < .001in both cases) and was observed for all subgroups (P < .001 for all cases). The addition of DBT increased the sensitivity of 2D mammography in patients with dense breasts and the specificity of 2D mammography for all subgroups. The use of synthetic 2D DBT demonstrated performance similar to that of standard 2D mammography with DBT. DBT is of potential benefit to screening programs, particularly in younger women with dense breasts. (©) RSNA, 2015.

The purpose of this study was to develop a deep learning model to diagnose breast cancer by embedding a diagnostic algorithm that examines the asymmetry of bilateral breast tissue. This retrospective study was approved by the institutional review board. A total of 115 patients who underwent breast surgery and had pathologically confirmed breast cancer were enrolled in this study. Two image pairs [230 pairs of bilateral breast digital breast tomosynthesis (DBT) images with 115 malignant tumors and contralateral tissue (M/N), and 115 bilateral normal areas (N/N)] were generated from each patient enrolled in this study. The proposed deep learning model is called bilateral asymmetrical detection (BilAD), which is a modified convolutional neural network (CNN) model of Xception with two-dimensional tensors for bilateral breast images. BilAD was trained to classify the differences between pairs of M/N and N/N datasets. The results of the BilAD model were compared to those of the unilateral control CNN model (uCNN). The results of BilAD and the uCNN were as follows: accuracy, 0.84 and 0.75; sensitivity, 0.73 and 0.58; and specificity, 0.93 and 0.92, respectively. The mean area under the receiver operating characteristic curve of BilAD was significantly higher than that of the uCNN (p = 0.02): 0.90 and 0.84, respectively. The proposed deep learning model trained by embedding a diagnostic algorithm to examine the asymmetry of bilateral breast tissue improves the diagnostic accuracy for breast cancer.

While deep learning (DL) models have shown promise in breast cancer diagnosis using digital breast tomosynthesis (DBT) images, the impact of varying matrix sizes and image interpolation methods on diagnostic accuracy remains unclear. Understanding these effects is essential to optimize preprocessing steps for DL models, which can lead to more efficient training processes, improved diagnostic accuracy, and better utilization of computational resources. Our institutional review board approved this retrospective study and waived the requirement for informed consent from the patients. In this study, 499 patients (29-90 years old, mean age 50.5 years) who underwent breast tomosynthesis were included. We performed downsampling to 256 × 256, 128 × 128, 64 × 64, and 32 × 32 using five image interpolation methods: Nearest (NN), Bilinear (BL), Bicubic (BC), Hamming (HM), and Lanczos (LC). The diagnostic accuracy of the DL model was assessed by mean AUC with its 95% confidence interval (CI). DL models with downsampled images to 256 × 256 pixels using the LC interpolation method showed a significantly lower AUC than the original 512 × 512 pixels model. This decrease was also observed with the 128 × 128 pixels DL models using HM and LC methods. All interpolation methods showed a significant decrease in AUC for the 64 × 64 and 32 × 32 pixels DL models. Our results highlight the significant impact of downsampling size and interpolation methods on the diagnostic performance of DL models. Understanding these effects is essential for optimizing preprocessing steps, which can enhance the accuracy and reliability of breast cancer diagnosis using DBT images.

The purpose of this study was to analyze the clinicopathologic and immunohistochemical features of invasive breast cancers detected only with digital breast tomosynthesis (DBT), compared with those of cancers detected with both DBT and full-field digital mammography (FFDM). The medical records of 261 women (108 without and 153 with symptoms) with invasive breast cancers who underwent FFDM and DBT between April 2015 and June 2016 were retrospectively reviewed. To assess detectability, all DBT and FFDM images were reviewed independently by three radiologists blinded to clinicopathologic information. The reference standard was established by an unblinded consensus review of all images. Clinicopathologic and immunohistochemical features were analyzed according to the detectability status. Of the 261 cancers, 223 (85.4%) were detected with both DBT and FFDM (both-detected group). Twenty-four cancers (9.2%) not detected with FFDM (DBT-only group) were classified by DBT as a mass (58.3%), architectural distortion (33.3%), or asymmetry (8.3%). The remaining 14 cancers (5.4%) were not detected with either DBT or FFDM (both-occult group). On multivariate analysis, a dense breast parenchyma (p = 0.007), small tumor size (≤ 2 cm; p = 0.027), and luminal A-like subtype (estrogen receptor positive or progesterone receptor positive or both, human epidermal growth factor receptor 2 negative, and Ki-67 expression < 14%; p = 0.008) were significantly associated with the DBT-only group. For 108 screening-detected cancers, a dense breast parenchyma (p = 0.007) and luminal A-like subtype (p = 0.008) also maintained significance. The addition of DBT to FFDM in screening would aid in the detection of less-aggressive subtypes of invasive breast cancers in women with dense breasts.

Mammographic breast density is a known breast cancer risk factor. Studies have shown the potential to automate breast density estimation by using computerized texture-based segmentation of the dense tissue in mammograms. Digital breast tomosynthesis (DBT) is a tomographic x-ray breast imaging modality that could allow volumetric breast density estimation. We evaluated the feasibility of distinguishing between dense and fatty breast regions in DBT using computer-extracted texture features. Our long-term hypothesis is that DBT texture analysis can be used to develop 3D dense tissue segmentation algorithms for estimating volumetric breast density. DBT images from 40 women were analyzed. The dense tissue area was delineated within each central source projection (CSP) image using a thresholding technique (Cumulus, Univ. Toronto). Two (2.5cm) 2 ROIs were manually selected: one within the dense tissue region and another within the fatty region. Corresponding (2.5cm) 3 ROIs were placed within the reconstructed DBT images. Texture features, previously used for mammographic dense tissue segmentation, were computed. Receiver operating characteristic (ROC) curve analysis was performed to evaluate feature classification performance. Different texture features appeared to perform best in the 3D reconstructed DBT compared to the 2D CSP images. Fractal dimension was superior in DBT (AUC=0.90), while contrast was best in CSP images (AUC=0.92). We attribute these differences to the effects of tissue superimposition in CSP and the volumetric visualization of the breast tissue in DBT. Our results suggest that novel approaches, different than those conventionally used in projection mammography, need to be investigated in order to develop DBT dense tissue segmentation algorithms for estimating volumetric breast density.

Previously, the authors developed a dual-energy (DE) acquisition technique for a photon-counting digital breast tomosynthesis (DBT) imaging system. Low-energy (LE) and high-energy (HE) images are acquired in a single scan by covering alternate slits of a multislit prepatient collimator with Sn and Cu, respectively. A theoretical model was used to optimize the technique. In this article, an experimental validation of this technique is presented. Experiments were performed on a prototype DBT system. LE and HE projection images were acquired sequentially; either a Sn or a Cu filter was positioned in the filter holder at the exit window of the x-ray tube. Sn filters from 0.113 to 0.242 mm thick and Cu filters from 0.103 to 0.267 mm were used. The images were acquired with a W target at 49 kV. Tomographic images, hereafter referred to as DBT images, were reconstructed using a shift-and-add algorithm. DE-DBT images were obtained by weighted logarithmic subtraction of the LE and HE images. Weighting factors w(t) that optimally cancel breast tissues with two different glandularities were assessed for 20-80 mm thick phantoms with 0%, 50%, and 100% glandularity. The mean and standard deviation in the per-pixel signal intensity (SI) were calculated in the DBT images. These data were used to calculate signal-difference-to-noise ratios (SDNRs) between iodine enhanced and nonenhanced polymethyl methacrylate backgrounds. To illustrate the feasibility of the technique, DE-DBT images of a structured phantom containing iodine disks were assessed. The experimental results were compared against the values obtained from a theoretical model of the imaging system. The average difference between theoretical and experimental w(t) was found to range from 8% to 21%. Experimental w(t) values increase with phantom thickness and Cu thickness, depend somewhat on Sn thickness, and vary more as a function of breast composition in thick breasts than in thin breasts. Theoretical and experimental mean and standard deviation in the per-pixel SI differ by -7% to 10% and by -3% to 4%. Theoretical and experimental SDNR values differ, on average, by 1.5%. Iodine concentrations can be predicted from SDNR; the relationship can be accurately fit to a quadratic. In the images of the structured phantom, iodine concentrations of 1 mg/cm2 and larger are discernable. The strong agreement between experimental and theoretical results in this article indicates that the authors' computer model is accurate.

Our study was to compare the clarity with which calcifications are seen on conventional mammography (CM) with the same calcifications on digital breast tomosynthesis (DBT). We define clarity as the sharpness, contrast, and diagnostic quality by which the calcifications were depicted. In a HIPPA compliant Institutional Review Board approved study, 3,000 women volunteered to have both a screening mammogram and a DBT study. A total of 119 sequential cases with relevant calcifications (not clearly benign) were reviewed. Two board certified, dedicated, breast imaging radiologists reviewed the CM and DBT images in an unblinded paired comparison. Only the mediolateral oblique (MLO) projection was available for the DBT studies. The MLO and craniocaudal projections were reviewed using the 2D images. Window and leveling, and electronic zoom were permitted. Unlimited time was allowed to provide a subjective assessment as too how well the calcifications were seen, from a diagnostic perspective, when the two studies were evaluated side-by-side. In 41.6% of the cases, the readers felt that calcifications were seen with superior clarity on DBT. In 50.4% of the cases, the visibility of calcifications was the same for DBT and CM, and in 8% of the cases, calcifications were seen with greater clarity on CM than DBT. In 92% of the cases, the clarity with which calcifications were seen on DBT was equal to or better than for CM and in almost half, the clarity on DBT was judged to be better than for CM. Our analysis shows that calcifications can be demonstrated with equal or greater clarity on DBT as on CM, thus allowing for comparable, and, perhaps, improved interpretive analysis of detected calcifications.

In digital breast tomosynthesis (DBT) imaging, a microcalcification (MC) cluster may span across different slices and blurring exists in the out-of-focus slices. We developed a radiomics approach to extract features from focus slice and combine multiple spatial domains to reduce false positives (FPs) in an automated pipeline of detecting MC clusters. We performed a retrospective study on a cohort of 290 Chinese women patients with a total of 580 DBT volumes. We developed an automated MC detection pipeline that consists of two stages: an initial detection to identify a set of MC candidates that may include many FPs, followed by a radiomics-based classification model to identify and reduce the FPs. We extract both two-dimensional (2D) and three-dimensional (3D) radiomics features from multiple spatial domains, including a focus slice, projection image, and tomographic volume. A linear discriminant classifier was used coupled with a sequential forward feature selection procedure. The free-response operating characteristics (FROC) curve and partial area under the FROC curve (pAUC) in the FP rate range of 0 to 2 per DBT volume were used to evaluate the model's performance. At a sensitivity of 90%, the FP rate was reduced from 1.3 to 0.2 per DBT volume after applying the multi-domain-based classification on the initial detections. The multi-domain yielded a significantly higher pAUC compared to the initial detection (increase of pAUC=0.2278, P<0.0001), focus slice (increase of pAUC=0.0345, P=0.0152), project image (increase of pAUC=0.1043, P<0.0001), and tomographic volume (increase of pAUC=0.0791, P=0.0032). The radiomic features extracted from the three domains may provide complementary information and their integration can significantly reduce FPs in automated detection of MCs in DBT volumes on a large Chinese women population.

Digital breast tomosynthesis (DBT) has superior detection performance than mammography (DM) for population-based breast cancer screening, but the higher number of images that must be reviewed poses a challenge for its implementation. This may be ameliorated by creating a twodimensional synthetic mammographic image (SM) from the DBT volume, containing the most relevant information. When creating a SM, it is of utmost importance to have an accurate lesion localization detection algorithm, while segmenting fibroglandular tissue could also be beneficial. These tasks encounter an extra challenge when working with images in the medio-lateral oblique view, due to the presence of the pectoral muscle, which has similar radiographic density. In this work, we present an automatic pectoral muscle segmentation model based on a u-net deep learning architecture, trained with 136 DBT images acquired with a single system (different BIRADS ® densities and pathological findings). The model was tested on 36 DBT images from that same system resulting in a dice similarity coefficient (DSC) of 0.977 (0.967-0.984). In addition, the model was tested on 125 images from two different systems and three different modalities (DBT, SM, DM), obtaining DSCs between 0.947 and 0.970, a range determined visually to provide adequate segmentations. For reference, a resident radiologist independently annotated a mix of 25 cases obtaining a DSC of 0.971. The results suggest the possibility of using this model for inter-manufacturer DBT, DM and SM tasks that benefit from the segmentation of the pectoral muscle, such as SM generation, computer aided detection systems, or patient dosimetry algorithms.

Digital breast tomosynthesis (DBT) mammography is a promising imaging technique for detecting early-stage breast cancers. However, DBT imaging usually contains truncation artifacts around the chest wall area, where many recurring breast tumors or masses are located. Extrapolation or interpolation and weighting techniques are used to suppress artifacts in existing methods, but these are inappropriate for DBT due to its high spatial resolution requirement. To solve this problem, we propose a practical truncation correction method, which can realize simultaneous truncation artifact reduction and image reconstruction (STARIR) for DBT. In contrast to extrapolation or interpolation estimations and weighting strategy, this integrated method renews each truncated voxel by considering the information of all measured projections rather than that of one projection at each reconstruction iteration. Qualitative and quantitative studies were performed on simulated and realistic DBT data to validate the proposed method. Results showed that STARIR reduces truncation artifacts and preserves tissue details in reconstructed images more effectively than do the current methods. The proposed method can be used for simultaneous and accurate truncation artifact reduction and image reconstruction in DBT.