Deformable Mapping Method to Relate Lesions in Dedicated Breast CT Images to Those in Automated Breast Ultrasound and Digital Breast Tomosynthesis Images

This work demonstrates the potential for using a deformable mapping method to register lesions between dedicated breast computed tomography (bCT) and both automated breast ultrasound (ABUS) and digital breast tomosynthesis (DBT) images (craniocaudal [CC] and mediolateral oblique [MLO] views). Two multi-modality breast phantoms with external fiducial markers attached were imaged by the three modalities. The DBT MLO view was excluded for the second phantom. The automated deformable mapping algorithm uses biomechanical modeling to determine corresponding lesions based on distances between their centers of mass (d_COM) in the deformed bCT model and the reference model (DBT or ABUS). For bCT to ABUS, the mean d_COM was 5.2 ± 2.6 mm. For bCT to DBT (CC), the mean d_COM was 5.1 ± 2.4 mm. For bCT to DBT (MLO), the mean d_COM was 4.7 ± 2.5 mm. This application could help improve a radiologist's efficiency and accuracy in breast lesion characterization, using multiple imaging modalities.     

几何模型匹配乳腺头足位与内外斜位X线片所示病灶

目的 观察几何模型(GM)匹配乳腺头足(CC)位与内外斜(MLO)位X线片所示病灶的价值。方法 回顾性分析493例接受乳腺CC位和MLO位X线摄影的乳腺病灶患者,共598个乳腺病灶,包括499个钙化灶和99个肿块。构建GM用于匹配CC与MLO位片所示乳腺病灶,再以环形法(AB)和直线法(SS)进行对比,分别计算匹配误差,包括GM匹配误差、AB径向误差及SS轴向误差;分析GM对CC及MLO位图像中同一病灶的匹配性能,评价其应用价值。结果 GM对乳腺钙化灶和肿块的匹配误差分别为2.85(1.45,5.08)及3.70(1.35,6.25)mm,差异无统计学意义(Z=-1.344,P=0.179)。对乳腺上部病灶,AB匹配的径向误差和SS匹配的轴向误差均大于下部病灶(P均<0.001);对乳腺外侧病灶,AB的径向误差和SS的轴向误差均大于内侧病灶(P均<0.05)。GM、AB及SS间匹配误差整体差异有统计学意义(H=93.012,P<0.001);两两比较差异均有统计学意义(P均<0.05),GM匹配性能明显优于AB和SS。GM匹配误差与摄片时乳腺压迫厚度无明显相关性...     

Breast Cancer Classification in Automated Breast Ultrasound Using Multiview Convolutional Neural Network with Transfer Learning

To assist radiologists in breast cancer classification in automated breast ultrasound (ABUS) imaging, we propose a computer-aided diagnosis based on a convolutional neural network (CNN) that classifies breast lesions as benign and malignant. The proposed CNN adopts a modified Inception-v3 architecture to provide efficient feature extraction in ABUS imaging. Because the ABUS images can be visualized in transverse and coronal views, the proposed CNN provides an efficient way to extract multiview features from both views. The proposed CNN was trained and evaluated on 316 breast lesions (135 malignant and 181 benign). An observer performance test was conducted to compare five human reviewers' diagnostic performance before and after referring to the predicting outcomes of the proposed CNN. Our method achieved an area under the curve (AUC) value of 0.9468 with five-folder cross-validation, for which the sensitivity and specificity were 0.886 and 0.876, respectively. Compared with conventional machine learning-based feature extraction schemes, particularly principal component analysis (PCA) and histogram of oriented gradients (HOG), our method achieved a significant improvement in classification performance. The proposed CNN achieved a >10% increased AUC value compared with PCA and HOG. During the observer performance test, the diagnostic results of all human reviewers had increased AUC values and sensitivities after referring to the classification results of the proposed CNN, and four of the five human reviewers’ AUCs were significantly improved. The proposed CNN employing a multiview strategy showed promise for the diagnosis of breast cancer, and could be used as a second reviewer for increasing diagnostic reliability.     

Diagnostic Performance of Automated Breast Ultrasound as a Replacement for a Hand-Held Second-Look Ultrasound for Breast Lesions Detected Initially on Magnetic Resonance Imaging

To evaluate the diagnostic performance of automated breast ultrasound (ABUS) after breast magnetic resonance imaging (MRI) as a replacement for hand-held second-look ultrasound (HH-SLUS), we evaluated 58 consecutive patients with breast cancer who had additional suspicious lesions on breast MRI. All patients underwent HH-SLUS and ABUS. Three breast radiologists evaluated the detectability, location, characteristics and conspicuity of lesions on ABUS. We also evaluated inter-observer variability and compared the results with HH-SLUS results. Eighty additional suspicious lesions were identified on breast MRI. Fifteen of the 80 lesions (19%) were not detected on HH-SLUS. Eight of the 15 lesions (53%) were detected on ABUS, whereas the remaining 7 were not detected on ABUS. Among the 65 lesions detected on HH-SLUS, only 3 lesions were not detected on ABUS. The intra-class correlation coefficients for lesion location and size all exceeded 0.70, indicating high reliability. Moderate to fair agreement was found for mass shape, orientation, margin and Breast Imaging Reporting and Data System (BI-RADS) final assessment. Therefore, ABUS can reliably detect additional suspicious lesions identified on breast MRI and may help in the decision on biopsy guidance method (US vs. MRI) as a replacement tool for HH-SLUS.     

MRI to X-ray mammography registration using a volume-preserving affine transformation

X-ray mammography is routinely used in national screening programmes and as a clinical diagnostic tool. Magnetic Resonance Imaging (MRI) is commonly used as a complementary modality, providing functional information about the breast and a 3D image that can overcome ambiguities caused by the superimposition of fibro-glandular structures associated with X-ray imaging. Relating findings between these modalities is a challenging task however, due to the different imaging processes involved and the large deformation that the breast undergoes. In this work we present a registration method to determine spatial correspondence between pairs of MR and X-ray images of the breast, that is targeted for clinical use. We propose a generic registration framework which incorporates a volume-preserving affine transformation model and validate its performance using routinely acquired clinical data. Experiments on simulated mammograms from 8 volunteers produced a mean registration error of 3.8±1.6mm for a mean of 12 manually identified landmarks per volume. When validated using 57 lesions identified on routine clinical CC and MLO mammograms (n=113 registration tasks) from 49 subjects the median registration error was 13.1mm. When applied to the registration of an MR image to CC and MLO mammograms of a patient with a localisation clip, the mean error was 8.9mm. The results indicate that an intensity based registration algorithm, using a relatively simple transformation model, can provide radiologists with a clinically useful tool for breast cancer diagnosis.     

Computer-aided diagnosis for the classification of breast masses in automated whole breast ultrasound images

New automated whole breast ultrasound (ABUS) machines have recently been developed and the ultrasound (US) volume dataset of the whole breast can be acquired in a standard manner. The purpose of this study was to develop a novel computer-aided diagnosis system for classification of breast masses in ABUS images. One hundred forty-seven cases (76 benign and 71 malignant breast masses) were obtained by a commercially available ABUS system. Because the distance of neighboring slices in ABUS images is fixed and small, these continuous slices were used for reconstruction as three-dimensional (3-D) US images. The 3-D tumor contour was segmented using the level-set segmentation method. Then, the 3-D features, including the texture, shape and ellipsoid fitting were extracted based on the segmented 3-D tumor contour to classify benign and malignant tumors based on the logistic regression model. The Student’s t test, Mann-Whitney U test and receiver operating characteristic (ROC) curve analysis were used for statistical analysis. From the Az values of ROC curves, the shape features (0.9138) are better than the texture features (0.8603) and the ellipsoid fitting features (0.8496) for classification. The difference was significant between shape and ellipsoid fitting features (p = 0.0382). However, combination of ellipsoid fitting features and shape features can achieve a best performance with accuracy of 85.0% (125/147), sensitivity of 84.5% (60/71), specificity of 85.5% (65/76) and the area under the ROC curve Az of 0.9466. The results showed that ABUS images could be used for computer-aided feature extraction and classification of breast tumors. (E-mail: rfchang@csie.ntu.edu.tw)     

自动乳腺超声诊断系统结合计算机辅助检测乳腺恶性肿瘤敏感度的影响因素分析

目的评估影响计算机辅助检测（CAD）识别自动乳腺超声诊断系统（ABUS）乳腺恶性肿瘤敏感度的因素。 方法收集自2016年1月至2017年2月于空军军医大学西京医院行ABUS检查并经外科手术或组织学活检病理证实的乳腺恶性肿瘤患者232例，共240个恶性病灶。所有病例均经CAD软件检测，统计CAD对病灶的总敏感度，并统计分析病灶组织学类型、最大径、距乳头距离、距皮肤距离及象限等因素与CAD敏感度之间的关系。以外科手术或组织学活检病理结果为诊断"金标准"，采用χ²检验分析病灶组织学类型、最大径、距乳头距离、距皮肤距离、象限、病灶边缘特征等因素与CAD敏感度的关系。 结果CAD对恶性病灶的总敏感度为85%（204/240），对不同病理学类型的敏感度分别为：浸润性导管癌89.0%（186/209）、导管原位癌53.9%（14/29）、黏液癌75.0%（3/4）、恶性叶状肿瘤100%（1/1），差异有统计学意义（χ²=18.836，P<0.001）。病灶最大径、距乳头距离、距皮肤距离及象限均与CAD敏感度之间比较，差异无统计学意义（P>0.05）。病灶距皮肤距离、病灶边缘特征与CAD对浸润性导管癌的敏感度之间比较，差异有统计学意义（P<0.05）。 结论CAD对恶性病灶的敏感度较高（85.0%），尤其是对浸润性导管癌的检出（89.0%），医师在借助CAD读图时，应注意是否有遗漏的导管原位癌、位置深或边缘模糊的浸润性导管癌。     

Auto-DenseUNet: Searchable neural network architecture for mass segmentation in 3D automated breast ultrasound

Accurate segmentation of breast mass in 3D automated breast ultrasound (ABUS) plays an important role in breast cancer analysis. Deep convolutional networks have become a promising approach in segmenting ABUS images. However, designing an effective network architecture is time-consuming, and highly relies on specialist’s experience and prior knowledge. To address this issue, we introduce a searchable segmentation network (denoted as Auto-DenseUNet) based on the neural architecture search (NAS) to search the optimal architecture automatically for the ABUS mass segmentation task. Concretely, a novel search space is designed based on a densely connected structure to enhance the gradient and information flows throughout the network. Then, to encourage multiscale information fusion, a set of searchable multiscale aggregation nodes between the down-sampling and up-sampling parts of the network are further designed. Thus, all the operators within the dense connection structure or between any two aggregation nodes can be searched to find the optimal structure. Finally, a novel decoupled search training strategy during architecture search is also introduced to alleviate the memory limitation caused by continuous relaxation in NAS. The proposed Auto-DenseUNet method has been evaluated on our ABUS dataset with 170 volumes (from 107 patients), including 120 training volumes and 50 testing volumes split at patient level. Experimental results on testing volumes show that our searched architecture performed better than several human-designed segmentation models on the 3D ABUS mass segmentation task, indicating the effectiveness of our proposed method.     

Multi-task learning for segmentation and classification of tumors in 3D automated breast ultrasound images

Tumor classification and segmentation are two important tasks for computer-aided diagnosis (CAD) using 3D automated breast ultrasound (ABUS) images. However, they are challenging due to the significant shape variation of breast tumors and the fuzzy nature of ultrasound images (e.g., low contrast and signal to noise ratio). Considering the correlation between tumor classification and segmentation, we argue that learning these two tasks jointly is able to improve the outcomes of both tasks. In this paper, we propose a novel multi-task learning framework for joint segmentation and classification of tumors in ABUS images. The proposed framework consists of two sub-networks: an encoder-decoder network for segmentation and a light-weight multi-scale network for classification. To account for the fuzzy boundaries of tumors in ABUS images, our framework uses an iterative training strategy to refine feature maps with the help of probability maps obtained from previous iterations. Experimental results based on a clinical dataset of 170 3D ABUS volumes collected from 107 patients indicate that the proposed multi-task framework improves tumor segmentation and classification over the single-task learning counterparts.     

Whole-Breast Ultrasound for Breast Screening and Archiving

The incidence of breast cancer is increasing worldwide, reinforcing the importance of breast screening. Conventional hand-held ultrasound (HHUS) for breast screening is efficient and relatively easy to perform; however, it lacks systematic recording and localization. This study investigated an electromagnetic tracking-based whole-breast ultrasound (WBUS) system to facilitate the use of HHUS for breast screening. One-hundred nine breast masses were collected, and the detection of suspicious breast lesions was compared between the WBUS system, HHUS and a commercial automated breast ultrasound (ABUS) system. The positioning error between WBUS and ABUS (1.39 ± 0.68 cm) was significantly smaller than that between HHUS and ABUS (1.62 ± 0.91 cm, p = 0.014) and HHUS and WBUS (1.63 ± 0.9 cm, p = 0.024). WBUS is a practical clinical tool for breast screening that can be used instead of the often unavailable and costly ABUS.     

不同年资医师应用自动乳腺超声诊断系统结合CAD对乳腺恶性病灶的诊断价值

目的探讨不同年资医师应用计算机辅助诊断系统(CAD)辅助自动乳腺超声诊断系统(ABUS)对于诊断乳腺恶性病灶的价值。方法收集行ABUS检查的乳腺病灶患者1452例,其中,恶性270例,共282个病灶;良性674例,共695个病灶;阴性508例。比较6名医师(3名低年资医师与3名高年资医师)使用CAD系统前后的诊断敏感性、特异性、受试者工作特征(ROC)曲线下面积及平均阅读时间。结果应用CAD前,低年资医师与高年资医师诊断恶性病灶的敏感性分别为87%、93%,使用CAD后均提高至94%,低年资医师使用CAD前、后诊断敏感性比较差异有统计学意义(P<0.05),高年资医师差异无统计学意义。6名医师在使用CAD系统前后诊断特异性无变化。低年资医师在使用CAD系统后的诊断准确率有所提高,曲线下面积由0.85提高至0.89,差异有统计学意义(P<0.05);而高年资医师在使用CAD系统后,虽然ROC曲线下面积由0.91提高至0.92,但差异无统计学意义。所有医师使用CAD后的平均阅读时间均有不同程度的延长,差异有统计学意义(P<0.05)。结论虽然使用CAD后的平均阅读时间有所延长,但在可接受范围内,ABUS结合CAD能大大提高超声医师诊断乳腺恶性病灶的准确率和敏感性,且对低年资医师帮助更大。     

自动乳腺超声诊断系统（ABUS）结合计算机辅助诊断系统（CAD）前后对乳腺恶性病灶的诊断价值对比

目的 本研究旨在探讨应用自动乳腺超声诊断系统（ABUS）结合计算机辅助诊断系统（CAD）对于不同经验医师诊断乳腺恶性病灶的价值。 方法 收集行ABUS检查的乳腺病例1452例,结果均经病理或随访1年以上证实。比较6名医师（3名低年资医师和3名高年资医师）使用CAD系统前后的诊断敏感度、特异度、AUC及平均阅读时间。结果 1452例病例中,270例为恶性,共有282个恶性病灶,674例为良性,共有695个良性病灶,508例为阴性。应用CAD系统前,低年资与高年资医师诊断乳腺癌的敏感度分别为87%、93%,使用CAD后提高到94%、94%,低年资医师的诊断敏感度前后差异具有统计学意义（P<0.05）,高年资医师差异无统计学意义（P＞0.05）。6名医师在使用CAD系统前后诊断特异度均略有下降,但差异均无统计学意义（P＞0.05）。低年资医师在使用CAD系统前后的诊断符合率有所提高,ROC曲线下面积由0.85提高到0.89,差异具有统计学意义（P<0.05）。而高年资医师组,虽然ROC曲线下面积由0.91提高到0.92,但差异不具有统计学意义（P＞0.05）。所有医师使用CAD后的平均阅读时间均有不同程度的延长,差异具有统计学意义（P<0.05）。结论 虽然使用CAD后的平均阅读时间有所延长,但在可接受范围内,借助ABUS-CAD的阅读模式能大大提高医生诊断的准确度和敏感度,对于低年资医师帮助更大。     

Preliminary Clinical Experience with a Combined Automated Breast Ultrasound and Digital Breast Tomosynthesis System

We analyzed the performance of a mammographically configured, automated breast ultrasound (McABUS) scanner combined with a digital breast tomosynthesis (DBT) system. The GE Invenia ultrasound system was modified for integration with GE DBT systems. Ultrasound and DBT imaging were performed in the same mammographic compression. Our small preliminary study included 13 cases, six of whom had contained invasive cancers. From analysis of these cases, current limitations and corresponding potential improvements of the system were determined. A registration analysis was performed to compare the ease of McABUS to DBT registration for this system with that of two systems designed previously. It was observed that in comparison to data from an earlier study, the McABUS-to-DBT registration alignment errors for both this system and a previously built combined system were smaller than those for a previously built standalone McABUS system.     

Diagnostic Performance Using Automated Breast Ultrasound System for Breast Cancer in Chinese Women Aged 40 Years or Older: A Comparative Study

The purpose of this study was to investigate the diagnostic performance of the automated breast ultrasound system (ABUS) compared with hand-held ultrasonography (HHUS) and mammography (MG) for breast cancer in women aged 40 y or older. A total of 594 breasts in 385 patients were enrolled in the study. HHUS, ABUS and MG exams were performed for these patients. Follow-up and pathologic findings were used as the reference standard. Based on the reference standard, 519 units were benign or normal and 75 were malignant. The sensitivity, specificity, accuracy and Youden index were 97.33%, 89.79%, 90.74% and 0.87 for HHUS; 90.67%, 92.49%, 92.26% and 0.83 for ABUS; 84.00%, 92.87%, 91.75% and 0.77 for MG, respectively. The specificity of ABUS was significantly superior to that of HHUS (p = 0.024). The area under the receiver operating characteristic curve was 0.936 for HHUS, which was the highest, followed by 0.916 for ABUS and 0.884 for MG. However, the difference was not statistically significant (p > 0.05). In conclusion, the diagnostic performance of ABUS for breast cancer was equivalent to HHUS and MG and potentially can be used as an alternative method for breast cancer diagnosis.     

Local diagnostic reference levels (LDRLs) for full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT) procedures in Morocco

IntroductionThe objective of this study was to establish local diagnostic reference levels (LDRLs) for the full-field digital mammography (FFDM) and tomosynthesis (DBT) in Moroccan health facilities.MethodsData from 146 women were collected from three facilities. The proposed DRLs were defined as the 75th percentile of the mean average glandular dose (AGD) distribution.ResultsThe mean AGD recorded in this study for the three centers was 1.47 mGy for all centers, and 1.42 mGy and 1.64 mGy for the CC and MLO projections, respectively. The mean compressed breast thickness (CBT) values recorded in this current study were 55 mm, the LDRLs reported for all centers was 1.7 mGy, the CC projection was 1.6 mGy, and the MLO projection was 1.8 mGy. In addition, the LDRLs reported in the current study were compared with those from previous studies for other countries, including the United Kingdom, Japan, Ghana, and Sri Lanka.ConclusionThis work provides an assessment of local DRLs for mammography in Morocco and is suggested as a starting point that will allow professionals to evaluate and optimize their practice. Furthermore, the definition of national DRLs is a necessary process in optimizing Moroccan medical exposures.     

Radiation dose from digital breast tomosynthesis screening – A comparison with full field digital mammography

ObjectivesTo compare Mean Glandular Dose (MGD) and effective dose from digital breast tomosynthesis (DBT) screening with that from full field digital mammography (FFDM) screening.MethodTo simulate compressed breasts, two Perspex-polyethylene breast phantoms were used, one phantom for compressed breast in craniocaudal and the other for compressed breast in mediolateral oblique. An adult ATOM dosimetry phantom was loaded with high sensitivity thermoluminescence dosimeters; the phantom was then positioned on Hologic Selenia Dimensions mammographic machine to imitate DBT and 4-view FFDM screening. Organ radiation doses were measured from 4-view DBT and 4-view FFDM (craniocaudal and mediolateral oblique views for each breast). Organ radiation doses were used to calculate effective dose from one screening session.ResultsMGD for DBT was 3.6 mGy; MGD for FFDM was 2.8 mGy. For DBT, other organs (e.g. thymus, lungs, salivary glands, thyroid, contralateral breast and bone marrow) radiation dose was also higher than for FFDM. The use of DBT for breast cancer screening increases the effective dose (E) of one screening session by 22%. E for DBT was 0.44 mSv; E for FFDM was 0.34 mSv.ConclusionThe use of DBT for breast cancer screening increases the radiation dose to screening clients.     

Towards improved breast mass detection using dual-view mammogram matching

Breast cancer screening benefits from the visual analysis of multiple views of routine mammograms. As for clinical practice, computer-aided diagnosis (CAD) systems could be enhanced by integrating multi-view information. In this work, we propose a new multi-tasking framework that combines craniocaudal (CC) and mediolateral-oblique (MLO) mammograms for automatic breast mass detection. Rather than addressing mass recognition only, we exploit multi-tasking properties of deep networks to jointly learn mass matching and classification, towards better detection performance. Specifically, we propose a unified Siamese network that combines patch-level mass/non-mass classification and dual-view mass matching to take full advantage of multi-view information. This model is exploited in a full image detection pipeline based on You-Only-Look-Once (YOLO) region proposals. We carry out exhaustive experiments to highlight the contribution of dual-view matching for both patch-level classification and examination-level detection scenarios. Results demonstrate that mass matching highly improves the full-pipeline detection performance by outperforming conventional single-task schemes with 94.78% as Area Under the Curve (AUC) score and a classification accuracy of 0.8791. Interestingly, mass classification also improves the performance of mass matching, which proves the complementarity of both tasks. Our method further guides clinicians by providing accurate dual-view mass correspondences, which suggests that it could act as a relevant second opinion for mammogram interpretation and breast cancer diagnosis.     

Automatic multimodal 2D/3D breast image registration using biomechanical FEM models and intensity-based optimization

Due to their different physical origin, X-ray mammography and Magnetic Resonance Imaging (MRI) provide complementary diagnostic information. However, the correlation of their images is challenging due to differences in dimensionality, patient positioning and compression state of the breast. Our automated registration takes over part of the correlation task. The registration method is based on a biomechanical finite element model, which is used to simulate mammographic compression. The deformed MRI volume can be compared directly with the corresponding mammogram. The registration accuracy is determined by a number of patient-specific parameters. We optimize these parameters – e.g. breast rotation – using image similarity measures. The method was evaluated on 79 datasets from clinical routine. The mean target registration error was 13.2 mm in a fully automated setting. On basis of our results, we conclude that a completely automated registration of volume images with 2D mammograms is feasible. The registration accuracy is within the clinically relevant range and thus beneficial for multimodal diagnosis.     

乳腺癌的钼靶X线诊断

目的；探讨乳腺癌的x线征象．提高钼靶x线对乳腺癌诊断准确性。方法：87例乳腺癌患者均行x线检查，检查体位常规采用轴位（CC位）、侧斜位（MLO位）。结果；87例患者中，病变位于左侧51例，右侧35例，双乳1例,1例为多中心性病灶，其余均为单发病灶。60例肿块直径小于临床触诊。结论：钼靶x线摄影是诊断乳腺癌的首选检查方法，结合乳腺癌的各种X线征象,一般可正确诊断。