Quantitative characterization of breast lesions and normal fibroglandular tissue using compartmentalized diffusion-weighted model: comparison of intravoxel incoherent motion and restriction spectrum imaging

BACKGROUND

To compare the compartmentalized diffusion-weighted models, intravoxel incoherent motion (IVIM) and restriction spectrum imaging (RSI), in characterizing breast lesions and normal fibroglandular tissue.

METHODS

This prospective study enrolled 152 patients with 157 histopathologically verified breast lesions (41 benign and 116 malignant). All patients underwent a full-protocol preoperative breast MRI, including a multi-b-value DWI sequence. The diffusion parameters derived from the mono-exponential model (ADC), IVIM model (Dt, Dp, f), and RSI model (C1, C2, C3, C1C2, F1, F2, F3, F1F2) were quantitatively measured and then compared among malignant lesions, benign lesions and normal fibroglandular tissues using Kruskal-Wallis test. The Mann-Whitney U-test was used for the pairwise comparisons. Diagnostic models were built by logistic regression analysis. The ROC analysis was performed using five-fold cross-validation and the mean AUC values were calculated and compared to evaluate the discriminative ability of each parameter or model.

RESULTS

Almost all quantitative diffusion parameters showed significant differences in distinguishing malignant breast lesions from both benign lesions (other than C2) and normal fibroglandular tissue (all parameters) (all P < 0.0167). In terms of the comparisons of benign lesions and normal fibroglandular tissues, the parameters derived from IVIM (Dp, f) and RSI (C1, C2, C1C2, F1, F2, F3) showed significant differences (all P < 0.005). When using individual parameters, RSI-derived parameters-F1, C1C2, and C2 values yielded the highest AUCs for the comparisons of malignant vs. benign, malignant vs. normal tissue and benign vs. normal tissue (AUCs = 0.871, 0.982, and 0.863, respectively). Furthermore, the combined diagnostic model (IVIM + RSI) exhibited the highest diagnostic efficacy for the pairwise discriminations (AUCs = 0.893, 0.991, and 0.928, respectively).

CONCLUSIONS

Quantitative parameters derived from the three-compartment RSI model have great promise as imaging indicators for the differential diagnosis of breast lesions compared with the bi-exponential IVIM model. Additionally, the combined model of IVIM and RSI achieves superior diagnostic performance in characterizing breast lesions.

Breast Imaging Reporting and Data System evaluation of breast lesions improved with virtual touch tissue imaging average grayscale values

BACKGROUND

Early breast cancer diagnosis is of great clinical importance for selecting treatment options, improving prognosis, and enhancing the quality of patients' survival.

OBJECTIVE

We investigated the value of virtual touch tissue imaging average grayscale values (VAGV) helper Breast Imaging Reporting and Data System (BI-RADS) in diagnosing breast malignancy.

METHODS

We retrospectively analyzed 141 breast tumors in 134 patients. All breast lesions were diagnosed pathologically by biopsy or surgical excision. All patients first underwent conventional ultrasound (US) followed by virtual touch tissue imaging (VTI). The measurement of the VAGV of the lesion was performed by Image J software. BI-RADS classification was performed for each lesion according to the US. We performed a two-by-two comparison of the diagnostic values of VAGV, BI-RADS, and BI-RADS+VAGV.

RESULTS

VAGV was lower in malignant tumors than in benign ones (35.82 ± 13.39 versus 73.58 ± 42.69, P< 0.001). The area under the receiver operating characteristic curve (AUC) value, sensitivity, and specificity of VAGV was 0.834, 84.09%, and 69.07%, respectively. Among BI-RADS, VAGV, and BI-RADS+VAGV, BI-RADS+VAGV had the highest AUC (0.926 versus 0.882, P= 0.0066; 0.926 versus 0.834, P= 0.0012). There was perfect agreement between the two radiologists using VAGV (ICC= 0.9796) and substantial agreement using BI-RADS (Kappa= 0.725).

CONCLUSION

Our study shows that VAGV can accurately diagnose breast cancer. VAGV effectively improves the diagnostic performance of BI-RADS.

Ceusia-Breast: computer-aided diagnosis with contrast enhanced ultrasound image analysis for breast lesions

BACKGROUND

In recent years, contrast-enhanced ultrasonography (CEUS) has been used for various applications in breast diagnosis. The superiority of CEUS over conventional B-mode imaging in the ultrasound diagnosis of the breast lesions in clinical practice has been widely confirmed. On the other hand, there have been many proposals for computer-aided diagnosis of breast lesions on B-mode ultrasound images, but few for CEUS. We propose a semi-automatic classification method based on machine learning in CEUS of breast lesions.

METHODS

The proposed method extracts spatial and temporal features from CEUS videos and breast tumors are classified as benign or malignant using linear support vector machines (SVM) with combination of selected optimal features. In the proposed method, tumor regions are extracted using the guidance information specified by the examiners, then morphological and texture features of tumor regions obtained from B-mode and CEUS images and TIC features obtained from CEUS video are extracted. Then, our method uses SVM classifiers to classify breast tumors as benign or malignant. During SVM training, many features are prepared, and useful features are selected. We name our proposed method "Ceucia-Breast" (Contrast Enhanced UltraSound Image Analysis for BREAST lesions).

RESULTS

The experimental results on 119 subjects show that the area under the receiver operating curve, accuracy, precision, and recall are 0.893, 0.816, 0.841 and 0.920, respectively. The classification performance is improved by our method over conventional methods using only B-mode images. In addition, we confirm that the selected features are consistent with the CEUS guidelines for breast tumor diagnosis. Furthermore, we conduct an experiment on the operator dependency of specifying guidance information and find that the intra-operator and inter-operator kappa coefficients are 1.0 and 0.798, respectively.

CONCLUSION

The experimental results show a significant improvement in classification performance compared to conventional classification methods using only B-mode images. We also confirm that the selected features are related to the findings that are considered important in clinical practice. Furthermore, we verify the intra- and inter-examiner correlation in the guidance input for region extraction and confirm that both correlations are in strong agreement.

Lupus mastitis and antiphospholipid syndrome treated with anticoagulation and immunosuppression: a case report

BACKGROUND

Systemic lupus erythematosus is an autoimmune disease that can have cutaneous and systemic manifestations. Lupus panniculitis, also known as lupus mastitis, is a subset of chronic cutaneous lupus erythematosus that involves inflammation of the subcutaneous fat. The pathogenesis of lupus mastitis is not fully understood. Diagnosis involves a combination of skin manifestations, imaging, and pathologic confirmation. Treatment typically includes steroids and antimalarials, with more severe disease requiring additional immunosuppressive medications. This report highlights a case of lupus mastitis treated with rituximab and a possible relationship between this disease process and thrombotic disease.

CASE PRESENTATION

A 48-year-old African American female with systemic lupus erythematosus and antiphospholipid syndrome presented with new breast lesion. Mammography revealed calcifications and increased density with coarse trabecular pattern. Breast biopsy showed features of cutaneous lupus and occlusive vasculopathy. The patient was diagnosed with lupus mastitis and treated with anticoagulation, rituximab, mycophenolate mofetil, and quinacrine with resolution of her symptoms.

CONCLUSION

This patient experienced improvement in her breast symptoms with combination therapy including rituximab. There are only two other cases reported in literature of patients with lupus mastitis responding to rituximab, highlighting the possible role of B cell depleting therapy for those who have contraindications to standard treatments for lupus mastitis. While the pathophysiology of lupus mastitis is thought to be immune driven, some literature suggests that associated thrombosis commonly seen may be due to a physiologic overlap similar to antiphospholipid syndrome. The possible relationship between antiphospholipid syndrome and lupus mastitis and the use of antiplatelet and anticoagulation therapy is discussed and may warrant further investigation.

BUS-Net: Breast Tumour Detection Network for Ultrasound Images Using Bi-directional ConvLSTM and Dense Residual Connections

Breast ultrasound (BUS) imaging has become one of the key imaging modalities for medical image diagnosis and prognosis. However, the manual process of lesion delineation from ultrasound images can incur various challenges concerning variable shape, size, intensity, curvature, or other medical priors of the lesion in the image. Therefore, computer-aided diagnostic (CADx) techniques incorporating deep learning-based neural networks are automatically used to segment the lesion from BUS images. This paper proposes an encoder-decoder-based architecture to recognize and accurately segment the lesion from two-dimensional BUS images. The architecture is utilized with the residual connection in both encoder and decoder paths; bi-directional ConvLSTM (BConvLSTM) units in the decoder extract the minute and detailed region of interest (ROI) information. BConvLSTM units and residual blocks help the network weigh ROI information more than the similar background region. Two public BUS image datasets, one with 163 images and the other with 42 images, are used. The proposed model is trained with the augmented images (ten forms) of dataset one (with 163 images), and test results are produced on the second dataset and the testing set of the first dataset-the segmentation performance yielding comparable results with the state-of-the-art segmentation methodologies. Similarly, the visual results show that the proposed approach for BUS image segmentation can accurately identify lesion contours and can potentially be applied for similar and larger datasets.

Pectoralis major muscle sarcoma masquerading breast lesion: A rare case report with review of literature

Pectoralis major muscle sarcomas are extremely rare and can mimic breast lesion clinically. We report a case of poorly differentiated sarcoma of the pectoralis major muscle in a 63-year-old woman of south east Asian ethnicity presenting with a progressively increasing right breast lump. Mammography, ultrasonography (US), contrast-enhanced computed tomography, and biopsy were done to make the final diagnosis. Complete surgical excision was planned but deferred due to pulmonary metastasis, and the patient was treated with palliative chemotherapy. Clinical examination may be confusing but radiological and pathological investigations provide detailed information about the location and the extent of the disease and a definitive tissue diagnosis can only be made on histopathology which will be helpful in preoperative planning and further treatment of the patient.

Diffusion kurtosis imaging and intravoxel incoherent motion imaging parameters in breast lesions: Effect of radiologists' experience and region-of-interest selection

PURPOSE

To investigate the influence of ROI placement methods and radiologists' experience on diffusion kurtosis imaging (DKI) and intravoxel incoherent motion (IVIM) parameters' diagnostic performance in differentiating benign and malignant lesions based on the mass and non-mass enhancement (NME).

METHODS

We evaluated 138 lesions in 131 patients retrospectively. The IVIM and DKI parameter values were measured by three radiologists with different experiences independently using two different ROI placement methods. IVIM parameters include diffusion coefficient (ADCstand), true diffusion coefficient (ADCslow), pseudo-diffusion coefficient (ADCfast) and perfusion fraction (f). DKI parameters include mean diffusivity (MD) and mean kurtosis (MK). Each radiologist measured the lesions twice with a 3-month interval. We utilized intra-class correlation (ICC) to determine the inter- and intra-reader agreement for mass and NME, respectively. ROC analysis compared the diagnostic performance of parameters between different radiologists, ROI methods, and between mass and NME.

RESULTS

In mass lesions, inter- and intra-observer agreement were perfect for all parameters (ICC: 0.800-989). In NME, the inter-observer agreement was substantial to perfect for all parameters(ICC: 0.703-877), the intra-observer agreement of the senior and intermediate radiologists was substantial to perfect(ICC: 0.748-931) and the intra-observer agreement of the junior radiologist was moderate to substantial(ICC: 0.569-784). The diagnostic performance of ADCslow (Z = 2.209, P = 0.023), MD (mean diffusivity) (Z = 2.887, P = 0.004), and MK (mean kurtosis) (Z = 2.080, P = 0.038) in the small ROI measured by the senior radiologist was better than that of the junior radiologist for NME. The diagnostic performance of ADCslow in the large ROI measured by the senior radiologist (Z = 2.281, P = 0.023) and intermediate radiologist (Z = 2.867, P = 0.0041) was better than the junior radiologist for mass lesions. The diagnostic performance of ADCslow, ADCstand, MD, and MK did not show a significant difference between the two ROI placement methods (P > 0.05).

CONCLUSION

The observers' experience can influence the ROI selection and the diagnostic performance of ADCslow, ADCstand, MD, and MK measured using different methods show equal diagnostic performance.

Different precompression does not reduce the diagnostic value of virtual touch tissue imaging and quantification (VTIQ) in breast lesions, especially for the ratio of the shear wave velocity between lesions and surrounding tissues

OBJECTIVE

To investigate the accuracy of virtual touch tissue imaging and quantification (VTIQ) in the diagnosis of benign and malignant breast lesions under four different precompression levels. The shear wave velocity (SWV) ratios of lesion to surrounding tissue were also added for diagnosis.

METHODS

167 female patients with breast lesions were included in this single center prospective study. VTIQ was performed under four different precompression levels. The SWV of the lesion, surrounding fat, and gland tissue were measured at the same depth as much as feasible 7 times. The breast lesions studied were all histopathologically confirmed. The VTIQ parameters were compared between precompression levels. The area under the receiver operating characteristic curve (AUC) was calculated to evaluate the diagnostic performance of each parameter of the VTIQ.

RESULTS

The VTIQ parameters of the malignant lesions were significantly higher than those of benign lesions in all precompression levels (P < 0.001). SWV of the lesion, fat, and gland tissue increased significantly with increasing precompression. The VTIQ parameters had great diagnostic performance for breast lesions in all precompression levels (AUC = 0.765-0.911). There was no significant difference between the precompression levels of the lesion-to-fat SWV ratio and the lesion-to-gland SWV ratio in benign and malignant lesions, and the cut-off coefficients of variation were 7.42% and 8.55%, respectively.

CONCLUSIONS

Precompression can increase the stiffness of breast lesions, fat and gland tissues, but does not reduce diagnostic value of VTIQ parameters in the breast. Under different precompression levels, the diagnosis of breast lesions by the ratio of the SWV of the lesion to the surrounding tissues is more stable.

The role of mineralocorticoids and glucocorticoids under the impact of 11β-hydroxysteroid dehydrogenase in human breast lesions

We attempted to explore the possible involvement of the in situ availability of mineralocorticoids and mineralocorticoid receptor (MR) in the pathogenesis of mammary ductal carcinoma. We also explored their individual profiles among different subtypes of invasive ductal carcinomas of no special type (IDC-NST) by evaluating the status of MR, Glucocorticoid receptor (GR), and 11β hydroxysteroid dehydrogenase (HSD) 1/2 at each stage of the putative cascade of the mammary ductal proliferative disorders. In this study, IDC-NST, ductal carcinoma in situ (DCIS), atypical ductal hyperplasia (ADH), and non-pathological breast tissues were all evaluated by immunohistochemistry. MR was significantly lower in ADH than in DCIS or IDC-NST. 11βHSD2 was significantly lower in ADH than normal breast tissue and 11βHSD1 was significantly higher in DCIS than normal, ADH, or IDC-NST. MR in progesterone receptor (PR)-positive IDC-NST cases tended to be associated with the Ki-67 labeling index. Results of the present study demonstrated that the status of MR and GR in conjunction with the 11βHSDs was correlated with the development of low-grade proliferative disorders in mammary glands. In addition, the potential crosstalk between MR and PR could also influence cell proliferation of breast carcinoma cells but further investigations are required for clarification.

Gail Model Improves the Diagnostic Performance of the Fifth Edition of Ultrasound BI-RADS for Predicting Breast Cancer: A Multicenter Prospective Study

RATIONALE AND OBJECTIVES

The sonographic appearance of benign and malignant breast nodules overlaps to some extent, and we aimed to assess the performance of the Gail model as an adjunctive tool to ultrasound (US) Breast Imaging Reporting and Data System (BI-RADS) for predicting the malignancy of nodules.

MATERIALS AND METHODS

From 2018 to 2019, 2607 patients were prospectively enrolled by 35 health care facilities. An individual breast cancer risk was assessed by the Gail model. Based on B-mode US, color Doppler, and elastography, all nodules were evaluated according to the fifth edition of BI-RADS, and these nodules were all confirmed later by pathology.

RESULTS

We demonstrated that the Gail model, age, tumor size, tumor shape, growth orientation, margin, contour, acoustic shadowing, microcalcification, presence of duct ectasia, presence of architectural distortion, color Doppler flow, BI-RADS, and elastography score were significantly related to breast cancer (all p < 0.001). The sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and area under the curve (AUC) for combining the Gail model with the BI-RADS category were 95.6%, 91.3%, 85.0%, 97.6%, 92.8%, and 0.98, respectively. Combining the Gail model with the BI-RADS showed better diagnostic efficiency than the BI-RADS and Gail model alone (AUC 0.98 vs 0.80, p < 0.001; AUC 0.98 vs 0.55, p < 0.001) and demonstrated a higher specificity than the BI-RADS (91.3% vs 59.4%, p < 0.001).

CONCLUSION

The Gail model could be used to differentiate malignant and benign breast lesions. Combined with the BI-RADS category, the Gail model was adjunctive to US for predicting breast lesions for malignancy. For the diagnosis of malignancy, more attention should be paid to high-risk patients with breast lesions.

The value of color-map virtual touch tissue imaging (CMV) in the differential diagnosis of benign and malignant breast lesions

BACKGROUND

Researchers have evaluated the virtual touch tissue imaging (VTI) value in the diagnosis of breast lesions, mostly based on gray-scale.

PURPOSE

This study aimed to evaluate the value of color-map virtual touch tissue imaging (CMV) in the diagnosis of breast lesions.

METHODS

We retrospectively analyzed the VTI images of 55 breast lesions in 49 female patients who underwent an examination of breast lesions in our hospital from January 2019 to December 2019. The pathological results were taken as the gold standard. The receiver operating characteristic (ROC) curve of CMV was analyzed, and its diagnostic performance was evaluated. Weighted Kappa (k) statistics were used to assess the inter-observer agreement for CMV.

RESULTS

A total of 55 breast lesions were included, including 19 malignant lesions and 36 benign lesions. Multivariate analysis showed that patients with higher CMV scores (P = 0.014, odds ratio [OR] = 13.667, 95% confidence interval = 1.702-109.773) were independent predictors of breast cancer. The sensitivity, specificity, and the area under curve (AUC) of CMV were 94.47%, 72.22%, and 0.912. The CMV's inter-observer agreement was almost perfect among radiologists with different work experience (k = 0.854, standard error = 0.049, 95% CI = 0.758-0.950).

CONCLUSIOS

CMV has high accuracy and repeatability in the diagnosis of malignant breast lesions.

Multi-parameter Sub-Hertz Analysis of Viscoelasticity With a Quality Metric for Differentiation of Breast Masses

We applied sub-Hertz analysis of viscoelasticity (SAVE) to differentiate breast masses in pre-biopsy patients. Tissue response during external ramp-and-hold stress was ultrasonically detected. Displacements were used to acquire tissue viscoelastic parameters. The fast instantaneous response and slow creep-like deformations were modeled as the response of a linear standard solid from which viscoelastic parameters were estimated. These parameters were used in a multi-variable classification framework to differentiate malignant from benign masses identified by pathology. When employing all viscoelasticity parameters, SAVE resulted in 71.43% accuracy in differentiating lesions. When combined with ultrasound features and lesion size, accuracy was 82.24%. Adding a quality metric based on uniaxial motion increased the accuracy to 81.25%. When all three were combined with SAVE, accuracy was 91.3%. These results confirm the utility of SAVE as a robust ultrasound-based diagnostic tool for non-invasive differentiation of breast masses when used as stand-alone biomarkers or in conjunction with ultrasonic features.

Intrathoracic hooked wire migration managed by minimally invasive thoracoscopic surgery

Background: Needle-guided localization wire is widely used to locate non-palpable breast lesion before surgery. A rare complication of this technique is wire migration. We report a case of an intrathoracic hooked wire migration in a 41-year-old female treated by video-assisted thoracoscopic surgery (VATS).Methods: We report a recent patient history and we review the cases reported in the literature.Results: Hook removal by thoracoscopy seemed to be the less invasive and most effective approach for this stable case. Even asymptomatic migration should be treated, and the device removed. The less invasive approach can be considered after estimating the risk and best possible timing.Conclusion: The loss of a hooked wire can lead to dramatic lesions. In every case, the device must be found or, if not, migration ruled out. The hooked wire must be removed, and the timing and the approach must be adapted to each case. VATS should be considered, in a stable patient to assess the lesions, to treat them and to remove the device.

Deep feature-based automatic classification of mammograms

Breast cancer has the second highest frequency of death rate among women worldwide. Early-stage prevention becomes complex due to reasons unknown. However, some typical signatures like masses and micro-calcifications upon investigating mammograms can help diagnose women better. Manual diagnosis is a hard task the radiologists carry out frequently. For their assistance, many computer-aided diagnosis (CADx) approaches have been developed. To improve upon the state of the art, we proposed a deep ensemble transfer learning and neural network classifier for automatic feature extraction and classification. In computer-assisted mammography, deep learning-based architectures are generally not trained on mammogram images directly. Instead, the images are pre-processed beforehand, and then they are adopted to be given as input to the ensemble model proposed. The robust features extracted from the ensemble model are optimized into a feature vector which are further classified using the neural network (nntraintool). The network was trained and tested to separate out benign and malignant tumors, thus achieving an accuracy of 0.88 with an area under curve (AUC) of 0.88. The attained results show that the proposed methodology is a promising and robust CADx system for breast cancer classification. Graphical Abstract Flow diagram of the proposed approach. Figure depicts the deep ensemble extracting the robust features with the final classification using neural networks.

Deep Learning Computer-Aided Diagnosis for Breast Lesion in Digital Mammogram

For computer-aided diagnosis (CAD), detection, segmentation, and classification from medical imagery are three key components to efficiently assist physicians for accurate diagnosis. In this chapter, a completely integrated CAD system based on deep learning is presented to diagnose breast lesions from digital X-ray mammograms involving detection, segmentation, and classification. To automatically detect breast lesions from mammograms, a regional deep learning approach called You-Only-Look-Once (YOLO) is used. To segment breast lesions, full resolution convolutional network (FrCN), a novel segmentation model of deep network, is implemented and used. Finally, three conventional deep learning models including regular feedforward CNN, ResNet-50, and InceptionResNet-V2 are separately adopted and used to classify or recognize the detected and segmented breast lesion as either benign or malignant. To evaluate the integrated CAD system for detection, segmentation, and classification, the publicly available and annotated INbreast database is used over fivefold cross-validation tests. The evaluation results of the YOLO-based detection achieved detection accuracy of 97.27%, Matthews's correlation coefficient (MCC) of 93.93%, and F1-score of 98.02%. Moreover, the results of the breast lesion segmentation via FrCN achieved an overall accuracy of 92.97%, MCC of 85.93%, Dice (F1-score) of 92.69%, and Jaccard similarity coefficient of 86.37%. The detected and segmented breast lesions are classified via CNN, ResNet-50, and InceptionResNet-V2 achieving an average overall accuracies of 88.74%, 92.56%, and 95.32%, respectively. The performance evaluation results through all stages of detection, segmentation, and classification show that the integrated CAD system outperforms the latest conventional deep learning methodologies. We conclude that our CAD system could be used to assist radiologists over all stages of detection, segmentation, and classification for diagnosis of breast lesions.

Semi-supervised segmentation of lesion from breast ultrasound images with attentional generative adversarial network. Comput Methods Programs Biomed. 2020;189:105275. [PMID: 31978805 DOI: 10.1016/j.cmpb.2019.105275]

BACKGROUND AND OBJECTIVE

Automatic segmentation of breast lesion from ultrasound images is a crucial module for the computer aided diagnostic systems in clinical practice. Large-scale breast ultrasound (BUS) images remain unannotated and need to be effectively explored to improve the segmentation quality. To address this, a semi-supervised segmentation network is proposed based on generative adversarial networks (GAN).

METHODS

In this paper, a semi-supervised learning model, denoted as BUS-GAN, consisting of a segmentation base network-BUS-S and an evaluation base network-BUS-E, is proposed. The BUS-S network can densely extract multi-scale features in order to accommodate the individual variance of breast lesion, thereby enhancing the robustness of segmentation. Besides, the BUS-E network adopts a dual-attentive-fusion block having two independent spatial attention paths on the predicted segmentation map and leverages the corresponding original image to distill geometrical-level and intensity-level information, respectively, so that to enlarge the difference between lesion region and background, thus improving the discriminative ability of the BUS-E network. Then, through adversarial training, the BUS-GAN model can achieve higher segmentation quality because the BUS-E network guides the BUS-S network to generate more accurate segmentation maps with more similar distribution as ground truth.

RESULTS

The counterpart semi-supervised segmentation methods and the proposed BUS-GAN model were trained with 2000 in-house images, including 100 annotated images and 1900 unannotated images, and tested on two different sites, including 800 in-house images and 163 public images. The results validate that the proposed BUS-GAN model can achieve higher segmentation accuracy on both the in-house testing dataset and the public dataset than state-of-the-art semi-supervised segmentation methods.

CONCLUSIONS

The developed BUS-GAN model can effectively utilize the unannotated breast ultrasound images to improve the segmentation quality. In the future, the proposed segmentation method can be a potential module for the automatic breast ultrasound diagnose system, thus relieving the burden of a tedious image annotation process and alleviating the subjective influence of physicians' experiences in clinical practice. Our code will be made available on https://github.com/fiy2W/BUS-GAN.

Evaluating the performance of convolutional neural networks with direct acyclic graph architectures in automatic segmentation of breast lesion in US images

Background

Outlining lesion contours in Ultra Sound (US) breast images is an important step in breast cancer diagnosis. Malignant lesions infiltrate the surrounding tissue, generating irregular contours, with spiculation and angulated margins, whereas benign lesions produce contours with a smooth outline and elliptical shape. In breast imaging, the majority of the existing publications in the literature focus on using Convolutional Neural Networks (CNNs) for segmentation and classification of lesions in mammographic images. In this study our main objective is to assess the ability of CNNs in detecting contour irregularities in breast lesions in US images.

Methods

In this study we compare the performance of two CNNs with Direct Acyclic Graph (DAG) architecture and one CNN with a series architecture for breast lesion segmentation in US images. DAG and series architectures are both feedforward networks. The difference is that a DAG architecture could have more than one path between the first layer and end layer, whereas a series architecture has only one path from the beginning layer to the end layer. The CNN architectures were evaluated with two datasets.

Results

With the more complex DAG architecture, the following mean values were obtained for the metrics used to evaluate the segmented contours: global accuracy: 0.956; IOU: 0.876; F measure: 68.77%; Dice coefficient: 0.892.

Conclusion

The CNN DAG architecture shows the best metric values used for quantitatively evaluating the segmented contours compared with the gold-standard contours. The segmented contours obtained with this architecture also have more details and irregularities, like the gold-standard contours.

Reducing Unnecessary Biopsy of Breast Lesions: Preliminary Results with Combination of Strain and Shear-Wave Elastography

The aim of our study was to compare strain elastography (SE), acoustic radiation force impulse-inducing Virtual Touch Imaging ([VTI] Siemens Medical Solutions, Mountain View, CA, USA), Virtual Touch Imaging Quantification ([VTIQ] Siemens Medical Solutions) and combined methods in the evaluation of ultrasound (US) Breast Imaging-Reporting and Data System (BI-RADS) category 4 lesions to explore an applicable way to reduce unnecessary biopsy by reducing false positives of conventional US without yielding false-negative cases. A total of 267 patients with 278 BI-RADS category 4 lesions (151 benign and 127 malignant) were evaluated with conventional B-mode US, SE, VTI and VTIQ implemented on a Siemens Acuson S2000 US system. Diagnostic performance, including area under the receiver operating characteristic curve, sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) were evaluated. Overall, VTI alone exhibited the highest NPV (91.74%), although combined elastic methods exhibited higher NPV than single methods, with the highest NPV at 100% when the VTI, SE and VTIQ methods were combined. Compared with conventional US, PPV increased from 45.7% (127 of 278) to 63.18% (127 of 201) when adding combined elastography (VTI + SE +VTIQ). In addition, 52.5% (63/120) and 50.8% (61/120) of BI-RADS 4 A lesions were downgraded when using combined methods (VTI + SE and VTI + SE + VTIQ, respectively) without missing any cancer. However, 2 intraductal papillomas and 1 phyllodes tumor were not identified. In conclusion, the combination of different elastic methods have the potential to downgrade BI-RADS 4A lesions to reduce false-positive biopsies without increasing the risk of missing cancers.

Comparison of the diagnostic performance of synthesized two-dimensional mammography and full-field digital mammography alone or in combination with digital breast tomosynthesis

PURPOSE

To investigate whether digital breast tomosynthesis (DBT) and subsequently generated synthesized mammography (SM) images show a better performance than full-field digital mammography (FFDM) images for diagnosing malignant breast lesions. In addition, the radiation doses for SM using different procedures were compared.

MATERIALS AND METHODS

This prospective study enrolled 212 women (age ≥ 25 years) with clinically suspicious breast lesions. All participants underwent FFDM and DBT with the same breast compression. Finally, 222 lesions were confirmed by pathological analysis. The mammogram results were evaluated according to the BI-RADS criteria and compared with the pathological results. The diagnostic performances, morphological features and average glandular doses (AGDs) were compared.

RESULTS

In total, 141 malignant lesions and 81 benign lesions were confirmed by pathological analysis. The overall AGD showed no significant difference between FFDM and DBT. Compared with 2D imaging, the AUC values of FFDM plus DBT and SM plus DBT were both significantly different overall (P = 0.0002) and remained significantly different in dense breasts (P < 0.0001). In terms of morphologic characteristics, lesions showed similar morphology between FFDM and SM, while the lesion characteristics were discordant from 2D imaging to DBT in 33 lesions in dense breasts.

CONCLUSIONS

Compared to FFDM, 2D SM images generated from DBT had significantly improved diagnostic efficacy for detecting malignant breast lesions without increasing radiation doses. This new procedure is useful for characterizing breast lesions, particularly in dense breasts.

Virtual touch tissue imaging and quantification (VTIQ) in the evaluation of breast lesions: The associated factors leading to misdiagnosis

PURPOSE

To investigate the factors that could cause a misdiagnosis in virtual touch tissue imaging and quantification (VTIQ) when differentiating benign and malignant breast lesions, and to analyze the imaging characteristics of those lesions with incorrect findings.

METHODS

The conventional ultrasound (CUS) features and the VTIQ parameters of 153 benign lesions and 99 malignant lesions were retrospectively analyzed and compared with histopathological and/or core-needle biopsy (CNB)-proven results. Independent variables that led to inaccurate VTIQ results were selected by binary logistic regression analysis.

RESULTS

The maximum shear wave speed (SWS-max), the mean SWS (SWS-mean), the minimum SWS (SWS-min), the lesion-to-fat SWS ratio (SWS-L/F), and the lesion-to-gland SWS ratio (SWS-L/G) in malignant lesions were significantly higher than those in benign lesions (all P < 0.001). The false-positive rate (FPR) of benign lesions and the false-negative rate (FNR) of malignant lesions were 9.8% and 19.2%, respectively, using an SWS-max cut-off value of 4.46 m/s. Diameter, depth, and posterior acoustic features were independent variables related to false-positive VTIQ findings (P: 0.049, 0.010 and 0.032, respectively). The invasive status and the histologic grade of infiltrating carcinoma were significantly associated with false-negative VTIQ findings (P: 0.026 and 0.015).

CONCLUSION

Diameter, depth, posterior acoustic features, invasive status, and histologic grade have a significant influence on the accuracy of VTIQ results, and these characteristics of breast lesions should be taken into account when interpreting the results of VTIQ examinations.

Breast lesion shape and margin evaluation: BI-RADS based metrics understate radiologists' actual levels of agreement

BACKGROUND

While there is much literature describing the radiologic detection of breast cancer, there are limited data available on the agreement between experts when delineating and classifying breast lesions. The aim of this work is to measure the level of agreement between expert radiologists when delineating and classifying breast lesions as demonstrated through Breast Imaging Reporting and Data System (BI-RADS) and quantitative shape metrics.

METHODS

Forty mammographic images, each containing a single lesion, were presented to nine expert breast radiologists using a high specification interactive digital drawing tablet with stylus. Each reader was asked to manually delineate the breast masses using the tablet and stylus and then visually classify the lesion according to the American College of Radiology (ACR) BI-RADS lexicon. The delineated lesion compactness and elongation were computed using Matlab software. Intraclass Correlation Coefficient (ICC) and Cohen's kappa were used to assess inter-observer agreement for delineation and classification outcomes, respectively.

RESULTS

Inter-observer agreement was fair for BI-RADS shape (kappa = 0.37) and moderate for margin (kappa = 0.58) assessments. Agreement for quantitative shape metrics was good for lesion elongation (ICC = 0.82) and excellent for compactness (ICC = 0.93).

CONCLUSIONS

Fair to moderate levels of agreement was shown by radiologists for shape and margin classifications of cancers using the BI-RADS lexicon. When quantitative shape metrics were used to evaluate radiologists' delineation of lesions, good to excellent inter-observer agreement was found. The results suggest that qualitative descriptors such as BI-RADS lesion shape and margin understate the actual level of expert radiologist agreement.

Ultrasound shear wave elastography of breast lesions: correlation of anisotropy with clinical and histopathological findings

Background

Ultrasound shear-wave elastography (SWE) may increase specificity of breast lesion assessment with ultrasound, but elasticity measurements may change with transducer orientation, defined as anisotropy. In this study, we aimed to observe the anisotropy of SWE of breast lesions, and its correlation with clinical and histopathological findings.

Methods

This retrospective study was approved by institutional review board. From June 2014 to June 2015, a total of 276 women (mean age, 48.75 ± 12.12 years) with 276 breast lesions (174 malignant, 102 benign) were enrolled for conventional ultrasound and SWE before surgical excision. Elasticity modulus in the longest diameter and orthogonal diameter were recorded, including maximum elasticity (Emax), mean elasticity (Emean), standard deviation (Esd) and ratio between mean elasticity of lesion and normal fatty tissue (Eratio). Anisotropy coefficients including anisotropic difference (AD) and anisotropy factors (AF) were calculated, and correlations with malignancy, tumor size, palpability, movability, lesion location and histopathology were analyzed.

Results

The average Emax, Emean, Esd and Eratio of the longest diameter were significantly higher than orthogonal diameter (P < 0.05). AUCs of ADs and AFs were inferior to quantitative parameters (P < 0.001), with AUCs of AFs superior to ADs (P < 0.001). ADs showed no significant correlation with malignancy, palpability, movability, distance from nipple and skin, and histopathological patterns. ADmean was significantly higher in inner half than outer half of the breast (P = 0.034). Higher AFs were significantly correlated with larger lesion size (P = 0.042), palpability (P < 0.05), shorter distance from nipple and skin (P < 0.05) and higher suspicion for malignancy (P < 0.001). AFs were significantly higher in IDC than DCIS (P < 0.05), higher in Grade II/III than Grade I IDC (P < 0.001), and correlated with ER/PR(+) (P < 0.05).

Conclusions

AF of SWE was an indicator for malignancy and more aggressive breast cancer.

Spatial Compounding Technique to Obtain Rotation Elastogram: A Feasibility Study

The perception of stiffness and slipperiness of a breast mass on palpation is used by physicians to assess the level of suspicion of a lesion as being malignant or benign. However, most current ultrasound elastography imaging methods provide only stiffness-related information. There is no existing approach that provides information about the local rigid body rotation undergone by only a loosely bonded, asymmetrically oriented lesion subjected to a small quasi-static compression. The inherent poor lateral resolution in ultrasound imaging poses a limitation in estimating the local rigid body rotation. Several techniques have been reported in the literature to improve the lateral resolution in ultrasound imaging, and among them is spatial compounding. In this study, we explore the feasibility of obtaining better-quality rotation elastograms with spatial compounding through simulations using Field II and experiments on tissue-mimicking phantoms. The phantom was subjected to axial compression (∼1%-2%) from the top, and the angular axial and lateral displacement estimates were obtained using a multilevel 2-D displacement tracking algorithm at different insonification angles. A rotation elastogram (RE) was obtained by taking half of the difference between the lateral gradient of the axial displacement estimates and the axial gradient of the lateral displacement estimates. Contrast-to-noise ratio was used to quantify the improvements in quality of RE. Contrast-to-noise ratio values were calculated by varying the maximum steering angle and the incremental angle, and its effects on RE quality were evaluated. Both simulation and experimental results corroborated and indicated a significant improvement in the quality of RE using compounding technique.

Value of Virtual Touch Tissue Imaging Quantification for Evaluation of Ultrasound Breast Imaging-Reporting and Data System Category 4 Lesions

The purpose of the study was to evaluate the value of 2-D shear wave elastography (SWE) of virtual touch tissue imaging quantification (VTIQ) for ultrasound (US) Breast Imaging-Reporting and Data System (BI-RADS) category 4 lesions. One hundred sixteen lesions were subject to conventional US, conventional strain elastography (SE) of elasticity imaging (EI), acoustic radiation force impulse (ARFI)-induced SE of virtual touch tissue imaging (VTI) and VTIQ before biopsies. Of the 116 lesions, 69 (59.5%) were benign and 47 (40.5%) were malignant. Significant differences were found between benign and malignant lesions in EI score, VTI score and shear wave speed (SWS) on VTIQ (both p < 0.05). The cut-off values were EI score ≥4, VTI score ≥4 and SWS ≥3.49 m/s, respectively. The diagnostic performance of VTIQ in terms of area under receiver operating characteristic curve (AUROC) were the highest (i.e., AUROC = 0.907), in comparison with EI, VTI alone or a combination of both. The associated sensitivity, specificity and accuracy were 87.2%, 82.6% and 84.5%, respectively. The combination of VTI and VTIQ, however, was similar with US BI-RADS (p = 0.475) in sensitivity in that only two (4.3%) of 47 malignant lesions were misdiagnosed as benign that were BI-RADS category 4b on US. VTIQ is valuable to differentiate benign from malignant BI-RADS category 4 lesions, and the combination of VTI and VTIQ might be useful for patient selection before biopsy.

[Breast lesions as the presenting feature of giant cell arteritis]

INTRODUCTION

Giant cell arteritis most commonly involves the external carotid branches. Although they are less typical, extra-cephalic forms have also been reported.

CASE REPORT

We report the case of a 59-year-old female patient who developed bilateral, painful breast nodules with fever and altered general status since two months. Two weeks later, she presented frontal headache and scalp tenderness. A colour duplex ultrasound of the temporal artery showed a halo sign. The results of a breast needle biopsy were inconclusive but the temporal artery biopsy confirmed the diagnosis of giant cell arteritis. The disease course was rapidly favourable after institution of corticosteroids.

INTRODUCTION

Breast involvement is rare but could be the first sign of giant cell arteritis. The internal mammary artery, which is a branch of the subclavian artery, can be affected and responsible for breast nodules.

Image manifold revealing for breast lesion segmentation in DCE-MRI

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is widely used for breast lesion differentiation. Manual segmentation in DCE-MRI is difficult and open to viewer interpretation. In this paper, an automatic segmentation method based on image manifold revealing was introduced to overcome the problems of the currently used method. First, high dimensional datasets were constructed from a dynamic image series. Next, an embedded image manifold was revealed in the feature image by nonlinear dimensionality reduction technique. In the last stage, k-means clustering was performed to obtain final segmentation results. The proposed method was applied in actual clinical cases and compared with the gold standard. Statistical analysis showed that the proposed method achieved an acceptable accuracy, sensitivity, and specificity rates.

A novel two-dimensional quantitative shear wave elastography for differentiating malignant from benign breast lesions

OBJECTIVE

The purpose of this study was to evaluate the diagnostic performance of a novel quantitative shear wave elastography (SWE) of virtual touch tissue imaging quantification (VTIQ) in diagnosis of breast lesions.

METHODS

The conventional ultrasound (US) and VTIQ images of 133 pathologically proven breast lesions in 98 patients were assessed. The breast lesions were classified by US breast imaging reporting and data system (BI-RADS) category. The maximum, minimum, mean and median shear wave velocity (SWV) values on VTIQ in the lesions were obtained. The area under the receiver operating curve (AUC) was computed.

RESULTS

Twenty-six of 133 lesions were malignant and 107 were benign. The sensitivity and specificity for US BI-RADS assessment were 96.2% and 62.6% respectively. The SWVs in malignant lesions were all significantly higher than those in benign ones (all P < 0.001). The AUC for mean SWV value was slightly higher than AUC for maximum, minimum and median SWV values, whereas no significant differences among them were found (all P > 0.05). The cut-off value of mean SWV was 3.68 m/s, with associated sensitivity and specificity of 93.3% and 79.4% respectively.

CONCLUSION

The novel quantitative SWE of VTIQ is helpful in differentiating breast lesions. Adding the quantitative SWE of VTIQ to the US BI-RADS assessment improves the specificity in diagnosing breast lesions without loss of sensitivity.

Acoustic radiation force impulse elastography for differentiation of malignant and benign breast lesions: a meta-analysis

This meta-analysis was aimed to assess the diagnostic performance of acoustic radiation force impulse (ARFI) elastography for the differentiation of malignant and benign breast lesions. The databases of PubMed, Web of Science(TM), WanFang, Vip, SinoMed and China National Knowledge Infrastructure were searched for all studies that evaluated the diagnostic performance of ARFI including virtual touch tissue quantification (VTQ) and virtual touch tissue imaging (VTI). All the studies were published prior to Mar. 21, 2014. The studies published in English or Chinese were collected. A total of 11 studies, including 1,408 breast lesions from 1,245 women, were analyzed. The values of summary sensitivity and summary specificity were 0.843 (95% confidence interval [CI]: 0.811-0.872) and 0.932 (95% CI: 0.913-0.948) for VTQ of ARFI, and 0.864 (95% CI: 0.799-0.914) and 0.882 (95% CI: 0.832-0.922) for VTI of ARFI, respectively. Subgroup analysis excluding mucinous carcinoma and carcinoma in situ showed higher summary sensitivity (0.877 95% CI: 0.835-0.911), higher summary specificity (0.943 95% CI: 0.921-0.960) and lower heterogeneity (I(2)=23.5%). The cut-off values for shear wave velocity of VTQ ranged widely from 2.89 to 6.71 m/s, while the VTI ranged narrowly from 1.37 to 1.66. In general, ARFI elastography seems to be a good method for differentiation between benign and malignant breast lesions. However, its usefulness for identifying breast mucinous carcinoma and breast carcinoma in situ is limited. VTI seems to be more reliable and repeatable than VTQ.

Needle core biopsy for breast lesions: An audit of 467 needle core biopsies

Background:

Breast cancer is the commonest cancer among women in urban India. Triple assessment includes clinical, radiological and cytological assessment of breast lesions. Guided core needle biopsy has replaced fine needle aspiration cytology in most of the western countries. In resource poor countries FNAC is still a very valuable and cost effective method to diagnose breast lesions. Pitfalls include increased rates of non diagnostic smears, and inadequate smears. Further procedures may be required and this increases the cost, anxiety and delay in diagnosis.

Aims:

The aim of this study is to analyze the concordance of radiological and histopathology findings in BI-RADS category 3,4,5 lesions following a core biopsy.

Materials and Methods:

Data was retrospectively collected from consecutive symptomatic and opportunistic screen detected patients with abnormalities who underwent ultrasound guided interventional procedures from Jan 2010 to Aug 2011. Symptomatic patients underwent clinical examination, mammogram and breast ultrasound. Women under 35 years of age had only breast ultrasound. Core biopsy was performed under ultrasound guidance or clinically by a breast surgeon/ radiologist for BI-RADS category 3,4,5 lesions.

Statistical Methods:

Chi square test was done to show the strength of association of imaging findings and histopathology results of core biopsy.

Results:

437 patients were symptomatic and 30 patients had screen detected abnormalities. The positive predictive value for BI-RADS 5 lesions for malignancy is 93.25% and the negative predictive value of BI-RADS category 3 lesions for cancer is 98.4%. False negative diagnosis on core biopsy was 0.85%. We were able to defer surgery in 60% of the patients with a clear radiological and pathological benign diagnosis.

Conclusion:

The PPV and NPV for cancer is high with needle core biopsy in BI-RADS 3,4,5 lesions. Where there is no discordance between clinical, radiology and pathology findings, surgery can be avoided in benign lesions. While in resource poor countries FNAC continues to be a valuable method in the diagnosis of palpable and non palpable breast lesions, the practice of needle core biopsy provides the most accurate and optimal diagnostic information.

Quantification of acoustic radiation force impulse in differentiating between malignant and benign breast lesions

The aim of this study was to evaluate the use of gray-level quantification (GLQ) in virtual touch tissue imaging (VTI) in the differential diagnosis of breast lesions. GLQ values of 153 lesions (101 benign, 52 malignant) were analyzed with matrix laboratory software (MATLAB, The MathWorks, Natick, MA, USA), with gray levels ranging from 0 (pure black) to 255 (pure white). The diagnostic performance of GLQ was also evaluated using receiver operating characteristic curve analysis. The mean GLQ value for benign lesions (103.27 ± 39.44) differed significantly from that for malignant lesions (44.57 ± 13.61) (p < 0.001). At a cutoff value of 52.31, the sensitivity, specificity, accuracy, positive predictive value and negative predictive value were 86.5%, 93.1%, 90.8%, 86.5% and 93.1%, respectively. In conclusion, we have proposed a method for quantification of gray levels in VTI for the differential diagnosis of breast lesions. Our results indicate that this method has the potential to aid in the classification of benign and malignant breast masses.

Diagnostic utility of combined ultrasonography and mammography in the evaluation of women with mammographically dense breasts

PURPOSE

To assess the diagnostic utility and additional cost of combined breast ultrasonography and mammography in the evaluation of asymptomatic women with mammographically dense breasts.

MATERIALS AND METHODS

Of 5108 asymptomatic women, who underwent mammography, 1754 had dense breasts (BI-RADS 3 or 4) and negative mammographic outcome. They were divided in 4 subgroups according to their age (<40 yrs; 40-49 yrs; 50-59 yrs; >59 yrs). Breast ultrasonography was performed immediately after mammography. Lesions detected at ultrasonography were examined cytologically/histologically. Mammograms from women, who were diagnosed carcinoma at ultrasonography, were reviewed by an external radiologist. Costs per diagnosed carcinoma and per examined woman were calculated on the basis of current regional charges.

RESULTS

Mammographies (5108) were performed, 67 cancers were detected (cancer detection rate 13.1‰): mammography identified 55 carcinomas and ultrasonography performed in women with dense breasts identified 12 cancers (17.9% of all cancers detected, overall cancer detection rate 6.8‰). Ultrasonography identified a benign condition in 1567 out of 1754 women (89.3%) (in 925 absence of focal lesions; 438 simple cysts; 56 ductal ectasia; 148 benign solid lesions); 97 complex cysts, 52 lesions that could not be differentiated as liquid or solid lesions, and 38 solid lesions suspicious for malignancy in the remaining 187 out of 1754 patients (10.7%). Cytology/histology confirmed carcinoma in 12 women (overall biopsy rate 26.2‰, benign biopsy rate 19.4‰). The additional costs were: € 6,123.45 per detected cancer, € 41.89 per examined woman.

CONCLUSION

Breast ultrasonography immediately after mammography in women with dense breasts is useful to avoid diagnostic delays and inconvenient medico-legal implications even though this procedure involves increased costs.