Breast imaging: A survey

Interactively Fusing Global and Local Features for Benign and Malignant Classification of Breast Ultrasound Images

OBJECTIVE

Breast ultrasound (BUS) is used to classify benign and malignant breast tumors, and its automatic classification can reduce subjectivity. However, current convolutional neural networks (CNNs) face challenges in capturing global features, while vision transformer (ViT) networks have limitations in effectively extracting local features. Therefore, this study aimed to develop a deep learning method that enables the interaction and updating of intermediate features between CNN and ViT to achieve high-accuracy BUS image classification.

METHODS

This study introduced the CNN and transformer multi-stage fusion network (CTMF-Net) consisting of two branches: a CNN branch and a transformer branch. The CNN branch employs visual geometry group as its backbone, while the transformer branch utilizes ViT as its base network. Both branches were divided into four stages. At the end of each stage, a proposed feature interaction module facilitated feature interaction and fusion between the two branches. Additionally, the convolutional block attention module was employed to enhance relevant features after each stage of the CNN branch. Extensive experiments were conducted using various state-of-the-art deep-learning classification methods on three public breast ultrasound datasets (SYSU, UDIAT and BUSI).

RESULTS

For the internal validation on SYSU and UDIAT, our proposed method CTMF-Net achieved the highest accuracy of 90.14 ± 0.58% on SYSU and 92.04 ± 4.90% on UDIAT, which showed superior classification performance over other state-of-art networks (p < 0.05). Additionally, for external validation on BUSI, CTMF-Net showed outstanding performance, achieving the highest area under the curve score of 0.8704 when trained on SYSU, marking a 0.0126 improvement over the second-best visual geometry group attention ViT method. Similarly, when applied to UDIAT, CTMF-Net achieved an area under the curve score of 0.8505, surpassing the second-best global context ViT method by 0.0130.

CONCLUSION

Our proposed method, CTMF-Net, outperforms all existing methods and can effectively assist doctors in achieving more accurate classification performance of breast tumors.

Diagnostic Accuracy and Incremental Value of Contrast-Enhanced Mammography Compared With Full Field Digital Mammography in a Tertiary Cancer Care Center

OBJECTIVE

To assess the diagnostic accuracy and incremental value of contrast-enhanced mammography (CEM) compared with full-field digital mammography (FFDM).

METHODOLOGY

A retrospective analysis was performed with 150 consecutive patients who underwent CEM at our institute between November 2020 and February 2021, fulfilling the inclusion criteria. The first round of analysis included a review of FFDM with an interpretation of findings as per the Breast Imaging Reporting and Data System (BIRADS) lexicon and the assignment of the BIRADS category to the detected abnormalities. After this documentation, a second round of analysis included a review of recombined subtracted images of CEM. The diagnostic accuracy of FFDM and CEM was calculated with histopathology as the gold standard.

RESULTS

Among the 150 cases assessed, 202 lesions were detected with histopathological correlation, of which 42 were benign and 160 were malignant. The sensitivity of FFDM was 90.6% compared to 98.12% for CEM. The specificity of FFDM was 66.7% compared to 76.19% for CEM. The negative predictive value (NPV) of FFDM was low, at 65.12%; CEM showed a better NPV, at 91.43%. The positive predictive value (PPV) was almost the same, at 94.01% for CEM and 91.19% for FFDM. The area under the curve (AUC) was superior for CEM compared to that of FFDM, with a value of 0.87. FFDM had a low sensitivity, especially in dense breast parenchyma, at 88.79% and a specificity of 70%, whereas CEM showed a higher sensitivity, specificity, and NPV, measuring 99.14%, 76.67%, and 95.83%, respectively.

CONCLUSION

Superior sensitivity and high NPV for CEM make it a preferable modality compared with FFDM, especially in dense breast parenchyma, where CEM overcomes the limitations of FFDM. We conclude that CEM is superior to FFDM in evaluating the extent of disease, additional satellite lesion detection, and ruling out ambiguous findings.

IMAGGS: a radiogenomic framework for identifying multi-way associations in breast cancer subtypes

Investigating correlations between radiomic and genomic profiling in breast cancer (BC) molecular subtypes is crucial for understanding disease mechanisms and providing personalized treatment. We present a well-designed radiogenomic framework image-gene-gene set (IMAGGS), which detects multi-way associations in BC subtypes by integrating radiomic and genomic features. Our dataset consists of 721 patients, each of whom has 12 ultrasound (US) images captured from different angles and gene mutation data. To better characterize tumor traits, 12 multi-angle US images are fused using two distinct strategies. Then, we analyze complex many-to-many associations between phenotypic and genotypic features using a machine learning algorithm, deviating from the prevalent one-to-one relationship pattern observed in previous studies. Key radiomic and genomic features are screened using these associations. In addition, gene set enrichment analysis is performed to investigate the joint effects of gene sets and delve deeper into the biological functions of BC subtypes. We further validate the feasibility of IMAGGS in a glioblastoma multiforme dataset to demonstrate the scalability of IMAGGS across different modalities and diseases. Taken together, IMAGGS provides a comprehensive characterization for diseases by associating imaging, genes, and gene sets, paving the way for biological interpretation of radiomics and development of targeted therapy.

Association of hormone receptors and human epidermal growth factor receptor-2/neu expressions with clinicopathologic factors of breast carcinoma: a cross-sectional study in a tertiary care hospital, Kabul, Afghanistan

BACKGROUND

Breast cancer (BC) is one of the major causes of death worldwide. It is the most common cause of death before the age of 70 years. The incidence and mortality of BC are rapidly increasing, posing great challenges to the health system and economy of every nation.

METHODOLOGY

A cross-sectional analytical study was conducted at the Department of Pathology and Clinical Laboratory of the French Medical Institute for Mothers and Children (FMIC) to demonstrate the association of human epidermal growth factor receptor 2 (Her2/Neu) and estrogen receptor (ER)/ progesterone receptor (PR) with clinical as well as pathological parameters among women with BC. A consecutive nonprobability sampling method was used for this study over a span of one and a half years.

RESULTS

One hundred twenty participants diagnosed with breast cancer were included in the study. The mean age at diagnosis was 44.58 ± 11.16 years. Out of the total patients, 68 (56.7%) were above 40 years old, 108 (90%) were married, 94 (78.3%) were multiparous, and 88 (73.3%) had a history of breastfeeding. 33.3% of cases were within the age range of menopause (40-50 years). The positive expression rates of ER, PR, and Her2/neu were found to be 48.8%, 44.6%, and 44.6%, respectively, and Her2/neu overexpression was found to be higher among ER/PR-negative cases.

CONCLUSION

In our study, we demonstrated that among Afghan women, grade II invasive ductal carcinoma, not otherwise specified, was the most common type of BC and frequently affected women above the age of 40. We also revealed that the percentage of negative ER (50.4%), negative PR (54.4%), and concordant ER/PR-negative cases were high compared to other possibilities. Additionally, the study revealed that expression of Her2/neu was in contrast with the expression of ER and PR receptors. The findings of our study still support the importance of performing immunohistochemical stains for hormonal receptor classification in terms of better clinical outcomes and prognosis.

Assessment of average glandular dose in mammography practice of a teaching hospital in Ghana

Introduction

above the age of 40, women are advised to begin breast examinations and screenings for early detection of breast cancer. The average glandular dose (AGD) provides dosimetric information about the quantity of radiation received by the mammary glands during mammographic exposures. There is, therefore, the need to analyse the radiation dose received by patients presenting for mammography examinations.

Methods

a retrospective cross-sectional design was carried out on the data of 663 participants, conveniently sampled between the months of July 2021 and June 2022. Paired T-test was used to compare imaging parameters for cranio-caudal (CC), medio-lateral (ML), automatic exposure control (AEC), manual exposure control (MEC), and left and right breast. Pearson´s correlation was used to test for relationship between imaging parameters and AGD.

Results

the mean AGD per exposure was 1.9 ± 0.7 mGy for CC projections and 2.3 ± 1.2 mGy for ML projections. The mean AGD per examination for the study was 4.1 ± 1.4 mGy. A positive correlation was found between AGD per examination and exposure factors (tube loading and tube voltage), compressed breast thickness, and compression force. Patient age had no statistically significant relationship with the AGD per examination.

Conclusion

average glandular dose (AGD) was consistent with other findings in literature studies. It was also observed that MEC yielded lower AGD per exposure values than AEC. There was no significant difference in the mean AGD per exposure for left and right breasts.

A twin convolutional neural network with hybrid binary optimizer for multimodal breast cancer digital image classification

There is a wide application of deep learning technique to unimodal medical image analysis with significant classification accuracy performance observed. However, real-world diagnosis of some chronic diseases such as breast cancer often require multimodal data streams with different modalities of visual and textual content. Mammography, magnetic resonance imaging (MRI) and image-guided breast biopsy represent a few of multimodal visual streams considered by physicians in isolating cases of breast cancer. Unfortunately, most studies applying deep learning techniques to solving classification problems in digital breast images have often narrowed their study to unimodal samples. This is understood considering the challenging nature of multimodal image abnormality classification where the fusion of high dimension heterogeneous features learned needs to be projected into a common representation space. This paper presents a novel deep learning approach combining a dual/twin convolutional neural network (TwinCNN) framework to address the challenge of breast cancer image classification from multi-modalities. First, modality-based feature learning was achieved by extracting both low and high levels features using the networks embedded with TwinCNN. Secondly, to address the notorious problem of high dimensionality associated with the extracted features, binary optimization method is adapted to effectively eliminate non-discriminant features in the search space. Furthermore, a novel method for feature fusion is applied to computationally leverage the ground-truth and predicted labels for each sample to enable multimodality classification. To evaluate the proposed method, digital mammography images and digital histopathology breast biopsy samples from benchmark datasets namely MIAS and BreakHis respectively. Experimental results obtained showed that the classification accuracy and area under the curve (AUC) for the single modalities yielded 0.755 and 0.861871 for histology, and 0.791 and 0.638 for mammography. Furthermore, the study investigated classification accuracy resulting from the fused feature method, and the result obtained showed that 0.977, 0.913, and 0.667 for histology, mammography, and multimodality respectively. The findings from the study confirmed that multimodal image classification based on combination of image features and predicted label improves performance. In addition, the contribution of the study shows that feature dimensionality reduction based on binary optimizer supports the elimination of non-discriminant features capable of bottle-necking the classifier.

A Comprehensive Review of Computer-Aided Models for Breast Cancer Diagnosis Using Histopathology Images

Breast cancer is the second most common cancer in women who are mainly middle-aged and older. The American Cancer Society reported that the average risk of developing breast cancer sometime in their life is about 13%, and this incident rate has increased by 0.5% per year in recent years. A biopsy is done when screening tests and imaging results show suspicious breast changes. Advancements in computer-aided system capabilities and performance have fueled research using histopathology images in cancer diagnosis. Advances in machine learning and deep neural networks have tremendously increased the number of studies developing computerized detection and classification models. The dataset-dependent nature and trial-and-error approach of the deep networks' performance produced varying results in the literature. This work comprehensively reviews the studies published between 2010 and 2022 regarding commonly used public-domain datasets and methodologies used in preprocessing, segmentation, feature engineering, machine-learning approaches, classifiers, and performance metrics.

AI-Based Cancer Detection Model for Contrast-Enhanced Mammography

BACKGROUND

The recent development of deep neural network models for the analysis of breast images has been a breakthrough in computer-aided diagnostics (CAD). Contrast-enhanced mammography (CEM) is a recent mammography modality providing anatomical and functional imaging of the breast. Despite the clinical benefits it could bring, only a few research studies have been conducted around deep-learning (DL) based CAD for CEM, especially because the access to large databases is still limited. This study presents the development and evaluation of a CEM-CAD for enhancing lesion detection and breast classification.

MATERIALS & METHODS

A deep learning enhanced cancer detection model based on a YOLO architecture has been optimized and trained on a large CEM dataset of 1673 patients (7443 images) with biopsy-proven lesions from various hospitals and acquisition systems. The evaluation was conducted using metrics derived from the free receiver operating characteristic (FROC) for the lesion detection and the receiver operating characteristic (ROC) to evaluate the overall breast classification performance. The performances were evaluated for different types of image input and for each patient background parenchymal enhancement (BPE) level.

RESULTS

The optimized model achieved an area under the curve (AUROC) of 0.964 for breast classification. Using both low-energy and recombined image as inputs for the DL model shows greater performance than using only the recombined image. For the lesion detection, the model was able to detect 90% of all cancers with a false positive (non-cancer) rate of 0.128 per image. This study demonstrates a high impact of BPE on classification and detection performance.

CONCLUSION

The developed CEM CAD outperforms previously published papers and its performance is comparable to radiologist-reported classification and detection capability.

Ultrasound as a Method for Early Diagnosis of Breast Pathology

INTRODUCTION

Ultrasound is a non-invasive, low-cost technique that does not use ionising radiation and provides a "real-time" image, and for these reasons, this method is ideal in several situations.

PURPOSE

To demonstrate breast ultrasound evaluation as a first-line diagnostic method and to evaluate the variation of breast characteristics with age.

MATERIAL AND METHODS

A total of 105 women with a mean age of 30 years participated and were divided into three age groups: 18-39, 40-59, and 60-79 years, excluding participants subject to mastectomy. After completing the informed consent, all participants answered personal and sociodemographic questions, such as personal and family history, menstrual cycle, pregnancy, ultrasound, and mammography, among others. They were then submitted to a bilateral breast ultrasound examination. Subsequently, all the images and their data were analysed, and a technical report of the examination was given to all the participants.

RESULTS

A total of 105 women with a mean age of 30 years participated, 58 of whom underwent the examination for the first time. In 31, changes (of which only 7 were known) were diagnosed. It was verified that, according to age group, the density of the breast stroma varied; older women have less breast density.

CONCLUSIONS

Ultrasound is a good method for breast evaluation and can be considered important for the early evaluation of breast pathology and follow-up of the pathology.

Synchronous bilateral breast cancer with different histology

Synchronous bilateral breast cancer is a rare clinical entity. And due to both an improved prognosis and growing life expectancy on early detection, we have brought interest in case of patient with synchronous breast cancer. This study reports a case of synchronous bilateral breast cancer in an asymptomatic 70-year-old woman with a positive family history of breast cancer. This woman was diagnosed through radiological screenings, including mammograms, ultrasonography, and magnetic resonance imaging (MRI). On histopathologic examination of the core biopsy, the left breast mass was a Nottingham grade I invasive carcinoma of no particular type. The right breast mass was a Nottingham grade I invasive carcinoma with a mucinous component. After lumpectomies ultrasonography of the surgical specimens confirmed a small biopsy-proven invasive ductal cancer hypoechoic mass in the left breast, with an irregular margins and proven mucinous cancer mass in the right breast. The case was finally diagnosed as synchronous bilateral breast cancer of different pathologic types (ductal and mucinous).

Enhancing Targeted Therapy in Breast Cancer by Ultrasound-Responsive Nanocarriers

Currently, the response to cancer treatments is highly variable, and severe side effects and toxicity are experienced by patients receiving high doses of chemotherapy, such as those diagnosed with triple-negative breast cancer. The main goal of researchers and clinicians is to develop new effective treatments that will be able to specifically target and kill tumor cells by employing the minimum doses of drugs exerting a therapeutic effect. Despite the development of new formulations that overall can increase the drugs’ pharmacokinetics, and that are specifically designed to bind overexpressed molecules on cancer cells and achieve active targeting of the tumor, the desired clinical outcome has not been reached yet. In this review, we will discuss the current classification and standard of care for breast cancer, the application of nanomedicine, and ultrasound-responsive biocompatible carriers (micro/nanobubbles, liposomes, micelles, polymeric nanoparticles, and nanodroplets/nanoemulsions) employed in preclinical studies to target and enhance the delivery of drugs and genes to breast cancer.

Fractal Dimension Analysis of Pixel Dynamic Contrast Enhanced-Magnetic Resonance Imaging Pharmacokinetic Parameters for Discrimination of Benign and Malignant Breast Lesions

PURPOSE

Breast cancer is the most frequent cancer in women worldwide. However, its diagnosis mostly depends on visual examination of radiologic images, leading to an overdiagnosis with substantial costs. Therefore, a quantitative approach such as dynamic contrast enhanced (DCE)-magnetic resonance imaging (MRI) through pharmacokinetic (PK) modeling is required for reliable analysis. As PK parameters lack information on parameter heterogeneity, texture-based analysis is required to quantify PK parameter heterogeneity. Therefore, this study focused on determining the usefulness of fractal dimension (FD) as a potential imaging biomarker of tumor heterogeneity for discriminating benign and malignant breast lesions.

METHODS

Parametric maps for PK parameters, extravasation rate of contrast agent from blood plasma to extravascular extracellular space (K^trans) and volume fraction of extravascular extracellular space (v_e), were generated for the regions of interest (ROIs) under the standard model using 18 lesions. Then, tumor ROI and pixel DCE-MRI time-course data were analyzed to extract pixel values of K^trans and v_e. For each ROI, FD values of K^trans and v_e were computed using the blanket method.

RESULTS

The FD values of K^trans for benign and malignant lesions varied from 2.96 to 3.49 and from 2.37 to 3.16, respectively, whereas FD values of v_e for benign and malignant lesions varied from 3.01 to 5.15 and 2.42 to 3.44, respectively. There were significant differences in FD values derived from K^trans parametric maps (P = .0053) and v_e parametric maps (P = .0271) between benign and malignant lesions according to the statistical analysis.

CONCLUSION

Incorporating texture heterogeneity changes in breast lesions captured by FD with quantitative DCE-MRI parameters generated under the standard model is a potential marker for prediction of malignant lesions.

The Impact of Data Management on the Achievable Dose and Efficiency of Mammography and Radiography During the COVID-19 Era: A Facility-Based Cohort Study

Purpose

To evaluate the impact of using computational data management resources and analytical software on radiation doses in mammography and radiography during the COVID-19 pandemic, develop departmental diagnostic reference levels (DRLs), and describe achievable doses (ADs) for mammography and radiography based on measured dose parameters.

Patients and Methods

This ambispective cohort study enrolled 795 and 12,115 patients who underwent mammography and radiography, respectively, at the King Fahd Hospital of the University, Al-Khobar City, Saudi Arabia between May 25 and November 4, 2021. Demographic data were acquired from patients' electronic medical charts. Data on mammographic and radiographic dose determinants were acquired from the data management software. Based on the time when the data management software was operational in the institute, the study was divided into the pre-implementation and post-implementation phases. Continuous and categorical variables were compared between the two phases using an unpaired t-test and the chi-square test.

Results

The median accumulated average glandular dose (AGD; a mammographic dose determinant) in the post-implementation phase was three-fold higher than that in the pre-implementation phase. The average mammographic exposure time in the post-implementation phase was 16.3 ms shorter than that in the pre-implementation phase. Furthermore, the median values of the dose area product ([DAP], a radiographic dose determinant) were 9.72 and 19.4 cGycm² in the pre-implementation and post-implementation phases, respectively.

Conclusion

Although the data management software used in this study helped reduce the radiation exposure time by 16.3 ms in mammography, its impact on the mean accumulated AGD was unfavorable. Similarly, radiographic exposure indices, including DAP, tube voltage, tube current, and exposure time, were not significantly different after the data management software was implemented. Close monitoring of patient radiation doses in mammography and radiography, and dose reduction will become possible if imaging facilities use DRLs and ADs via automated systems.

Obesity and breast cancer

The paper presents an analysis of the latest literature data on the problem of obesity and breast cancer (BC). This review presents modern approaches to the diagnosis of BC in obese patients, new molecular methods of breast imaging,  analyzes the features of the course of BC with obesity depending on menstrual status, molecular biological subtypes of the tumor, the mechanisms of the development of BC against the background of obesity.

Application of Artificial Intelligence Techniques to Predict Risk of Recurrence of Breast Cancer: A Systematic Review

Breast cancer is the most common disease among women, with over 2.1 million new diagnoses each year worldwide. About 30% of patients initially presenting with early stage disease have a recurrence of cancer within 10 years. Predicting who will have a recurrence and who will not remains challenging, with consequent implications for associated treatment. Artificial intelligence strategies that can predict the risk of recurrence of breast cancer could help breast cancer clinicians avoid ineffective overtreatment. Despite its significance, most breast cancer recurrence datasets are insufficiently large, not publicly available, or imbalanced, making these studies more difficult. This systematic review investigates the role of artificial intelligence in the prediction of breast cancer recurrence. We summarise common techniques, features, training and testing methodologies, metrics, and discuss current challenges relating to implementation in clinical practice. We systematically reviewed works published between 1 January 2011 and 1 November 2021 using the methodology of Kitchenham and Charter. We leveraged Springer, Google Scholar, PubMed, and IEEE search engines. This review found three areas that require further work. First, there is no agreement on artificial intelligence methodologies, feature predictors, or assessment metrics. Second, issues such as sampling strategies, missing data, and class imbalance problems are rarely addressed or discussed. Third, representative datasets for breast cancer recurrence are scarce, which hinders model validation and deployment. We conclude that predicting breast cancer recurrence remains an open problem despite the use of artificial intelligence.

A VGG attention vision transformer network for benign and malignant classification of breast ultrasound images

PURPOSE

Breast cancer is the most commonly occurring cancer worldwide. The ultrasound reflectivity imaging technique can be used to obtain breast ultrasound (BUS) images, which can be used to classify benign and malignant tumors. However, the classification is subjective and dependent on the experience and skill of operators and doctors. The automatic classification method can assist doctors and improve the objectivity, but current convolution neural network (CNN) is not good at learning global features and vision transform (ViT) is not good at extraction local features. In this study, we proposed an VGG attention vision transformer (VGGA-ViT) network to overcome their disadvantages.

METHODS

In the proposed method, we used a convolutional neural network (CNN) module to extract the local features and employed a vision transformer (ViT) module to learn the global relationship between different regions and enhance the relevant local features. The CNN module was named the VGG attention (VGGA) module. It was composed of a visual geometry group (VGG) backbone, a feature extraction fully connected layer, and a squeeze-and-excitation (SE) block. Both the VGG backbone and the ViT module were pre-trained on the ImageNet dataset and re-trained using BUS samples in this study. Two BUS datasets were employed for validation.

RESULTS

Cross-validation was conducted on two BUS datasets.

CONCLUSIONS

In this study, we proposed the VGGA-ViT for the BUS classification, which was good at learning both local and global features. The proposed network achieved higher accuracy than the compared previous methods. This article is protected by copyright. All rights reserved.

Volumetric tri-modal imaging with combined photoacoustic, ultrasound, and shear wave elastography

Photoacoustic imaging is a hybrid imaging approach that combines the advantages of optical and ultrasonic imaging in one modality. However, for comprehensive tissue characterization, optical contrast alone is not always sufficient. In this study, we combined photoacoustic imaging with high-resolution ultrasound and shear wave elastography. The multi-modal system can calculate optical absorption, acoustic reflection, and stiffness volumetrically. We constructed a multi-modal phantom with contrast for each imaging modality to test the system's performance. Experimental results indicate that the system successfully visualizes the embedded structures. We envision that the system will lead to more comprehensive tissue characterization for cancer screening and diagnosis.

A novel bi-modal extended Huber loss function based refined mask RCNN approach for automatic multi instance detection and localization of breast cancer

Breast cancer is an extremely aggressive cancer in women. Its abnormalities can be observed in the form of masses, calcification and lumps. In order to reduce the mortality rate of women its detection is needed at an early stage. The present paper proposes a novel bi-modal extended Huber loss function based refined mask regional convolutional neural network for automatic multi-instance detection and localization of breast cancer. To refine and increase the efficacy of the proposed method three changes are casted. First, a pre-processing step is performed for mammogram and ultrasound breast images. Second, the features of the region proposal network are separately mapped for accurate region of interest. Third, to reduce overfitting and fast convergence, an extended Huber loss function is used at the place of Smooth L1( x) in boundary loss. To extend the functionality of Huber loss, the delta parameter is automated by the aid of median absolute deviation with grid search algorithm. It provides the best optimum value of delta instead of user-based value. The proposed method is compared with pre-existing methods in terms of accuracy, true positive rate, true negative rate, precision, F-score, balanced classification rate, Youden’s index, Jaccard Index and dice coefficient on CBIS-DDSM and ultrasound database. The experimental result shows that the proposed method is a better suited approach for multi-instance detection, localization and classification of breast cancer. It can be used as a diagnostic medium that helps in clinical purposes and leads to a precise diagnosis of breast cancer abnormalities.

The Accuracy of Electrical Impedance Tomography for Breast Cancer Detection: A Systematic Review and Meta-Analysis

Introduction

Incidence of breast cancer (BC) in 2020 is about 2.26 million new cases. It is the first common cancer accounting for 11.7% of all cancer worldwide. Disease complications and the mortality rate of breast cancer are highly dependent on the early diagnosis. Therefore, novel human breast-imaging techniques play an important role in minimizing the breast cancer morbidity and mortality rate. Electrical impedance tomography (EIT) is a noninvasive technique to image the breast using the electrical impedance behavior of the body tissues.

Objectives

The aims of this manuscript are as follows: (1) a comprehensive investigation of the accuracy of EIT for breast cancer diagnosis through searching pieces of evidence in the valid databases and (2) meta-analyses of the results.

Methods

The systematic search was performed in the electronic databases including PubMed, Web of Science, EMBASE, Science Direct, ProQuest, Scopus, and Google Scholar without time and language limitation until January 2021. Search terms were “EIT” and “Breast Cancer” with their synonyms. Relevant studies were included based on PRISMA and study objectives. Quality of studies and risk of bias were performed by QUADAS-2 tools. Then, relevant data were extracted in Excel form. The hierarchical/bivariate meta-analysis was performed with “metandi” package for the ROC plot of sensitivity and specificity. Forest plot of the Accuracy index and double arcsine transformations was applied to stabilize the variance. The heterogeneity of the studies was evaluated by the forest plots, χ2 test (assuming a significance at the a-level of 10%), and the I² statistic for the Accuracy index.

Results

A total of 4027 articles were found. Finally, 12 of which met our criteria. Overall, these articles included studies of 5487 breast cancer patients. EIT had an overall pooled sensitivity and specificity of 75.88% (95% CI, 61.92% to 85.89%) and 82.04% (95% CI, 69.72% to 90.06%), respectively. The pooled diagnostic odds ratio was 14.37 (95% CI, 6.22% to 33.20%), and the pooled effect of accuracy was 0.79 with 95% CI (0.73, 0.83).

Conclusions

This study showed that EIT can be used as a useful method alongside mammography. EIT sensitivity could not be compared with the sensitivity of MRI, but in terms of specificity, it can be considered as a new method that probably can get more attention. Furthermore, large-scale studies will be needed to support the evidence.

Artificial intelligence for breast cancer analysis: Trends & directions

Breast cancer is one of the leading causes of death among women. Early detection of breast cancer can significantly improve the lives of millions of women across the globe. Given importance of finding solution/framework for early detection and diagnosis, recently many AI researchers are focusing to automate this task. The other reasons for surge in research activities in this direction are advent of robust AI algorithms (deep learning), availability of hardware that can run/train those robust and complex AI algorithms and accessibility of large enough dataset required for training AI algorithms. Different imaging modalities that have been exploited by researchers to automate the task of breast cancer detection are mammograms, ultrasound, magnetic resonance imaging, histopathological images or any combination of them. This article analyzes these imaging modalities and presents their strengths and limitations. It also enlists resources from where their datasets can be accessed for research purpose. This article then summarizes AI and computer vision based state-of-the-art methods proposed in the last decade to detect breast cancer using various imaging modalities. Primarily, in this article we have focused on reviewing frameworks that have reported results using mammograms as it is the most widely used breast imaging modality that serves as the first test that medical practitioners usually prescribe for the detection of breast cancer. Another reason for focusing on mammogram imaging modalities is the availability of its labelled datasets. Datasets availability is one of the most important aspects for the development of AI based frameworks as such algorithms are data hungry and generally quality of dataset affects performance of AI based algorithms. In a nutshell, this research article will act as a primary resource for the research community working in the field of automated breast imaging analysis.

A U-Net Ensemble for breast lesion segmentation in DCE MRI

Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) has been recognized as an effective tool for Breast Cancer (BC) diagnosis. Automatic BC analysis from DCE-MRI depends on features extracted particularly from lesions, hence, lesions need to be accurately segmented as a prior step. Due to the time and experience required to manually segment lesions in 4D DCE-MRI, automating this task is expected to reduce the workload, reduce observer variability and improve diagnostic accuracy. In this paper we propose an automated method for breast lesion segmentation from DCE-MRI based on a U-Net framework. The contributions of this work are the proposal of a modified U-Net architecture and the analysis of the input DCE information. In that sense, we propose the use of an ensemble method combining three U-Net models, each using a different input combination, outperforming all individual methods and other existing approaches. For evaluation, we use a subset of 46 cases from the TCGA-BRCA dataset, a challenging and publicly available dataset not reported to date for this task. Due to the incomplete annotations provided, we complement them with the help of a radiologist in order to include secondary lesions that were not originally segmented. The proposed ensemble method obtains a mean Dice Similarity Coefficient (DSC) of 0.680 (0.802 for main lesions) which outperforms state-of-the art methods using the same dataset, demonstrating the effectiveness of our method considering the complexity of the dataset.

An Early Breast Cancer Detection System Using Stacked Auto Encoder Deep Neural Network with Particle Swarm Optimization Based Classification Method

The demand in breast cancer’s early detection and diagnosis over the last few decade has given a new research avenues. For an individual who is suffered from breast cancer, a successful treatment plan can be specified if early stage diagnosis of non-communicable disease is done as stated by world health organization (WHO). Around the world, mortality can be reduced by cure disease’s early diagnosis. For breast cancer’s early detection and to detect other abnormalities of human breast tissue, digital mammogram is used as a most popular screening method. Early detection is assisted by periodic clinical check-ups and self-tests and survival chance is significantly enhanced by it. For mammograms (MGs), deep learning (DL) methods are investigated by researchers due to traditional computer-aided detection (CAD) systems limitations and breast cancer’s early detection’s extreme importance and patients false diagnosis high impact. So, there is need to have a noninvasive cancer detection system which is efficient, accurate, fast and robust. There are two process in proposed work, Histogram Rehabilitated Local Contrast Enhancement (HRLCE) technique is used in initial process for contrast enhancement with two processing stages. Contrast enhancements potentiality is enhanced while preserving image’s local details by this technique. So, for cancer classification, Particle Swarm Optimization (PSO) and stacked auto encoders (SAE) combined with framework based on DNN called SAE-PSO-DNN Model is used. The SAE-DNN parameters with two hidden layers are tuned using PSO and Limited-memory BFGS (LBFGS) is used as a technique for reducing features. Specificity, sensitivity, normalized root mean square erro (NRMSE), accuracy parameters are used for evaluating SAE-PSO-DNN models results. Around 92% of accurate results are produced by SAE-PSO-DNN model as shown in experimentation results, which is far better than Convolutional Neural Network (CNN) as well as Support Vector Machine (SVM) techniques.

The importance of preoperative imaging methods in reduction mammoplasty

Breast reduction surgery is a common procedure in plastic surgery clinics, and there are varying practices regarding preoperative mammography and breast ultrasound in patients who have no history of cancer and no symptoms other than those caused by macromastia. In this study, we retrospectively analyzed the imaging findings of patients who presented to the plastic surgery outpatient clinic due to macromastia between January 1, 2006 and June 1, 2020 and underwent mammography and/or breast ultrasound for preoperative screening and breast magnetic resonance imaging for further examination. Patients with a history of breast cancer diagnosed prior to preoperative screening or any other breast disease were excluded. Radiologically suspicious findings were significantly more common in patients over 40 years of age and significantly less frequent in the group under 50 years of age. When the patients were grouped by the decade of life, the frequency of radiologically suspicious findings was significantly lower in the 20-29 group and significantly higher in the 40-49 and 50-59 groups. Malignancy was not detected by histopathological examination in any patient. Proliferative lesions were detected in 10 reduction mammoplasty specimens (2.4%) of 7 patients (3.1%). The correlation between radiological and histopathological findings may be weak in macromastia patients. Most suspicious radiological findings appear to be the population between 40 and 59 years of age .

Optimization of Local Contrast Factor with Adaptive Brightness Improvement: Impact on Mammogram Image Analysis

Mammography has been known worldwide as the most common imaging modalities utilized for early detection of breast cancer. The mammographic images produced are in greyscale, however they often produced low contrast images, contain artefacts and noise, as well as non-uniform illumination. These limitations can be overcame in the pre-processing stage with the image enhancement process. Therefore, in this research we developed an optimized enhancement framework where the local contrast factor is manipulated to preserve details of the image. This method aims to improve the overall image visibility without altering histogram of the original image, which will affect the segmentation and classification processes. We performed dark background removal in the image histogram at early stage to increase the efficiency of new mean histogram calculation. Then, the histogram is separated into two partitions to allow histogram clipping process to be conducted individually for underexposed and overexposed areas. Consequently, the local contrast factor optimization is conducted to preserve the image details. The results from our proposed method are compared with other methods by the measurement of peak signal-to-noise ratio, structural similarity index, average contrast, and average entropy difference. The results portrayed that our proposed method yield better quality over the others with highest peak signal-to-noise ratio of 32.676. In addition, in terms of qualitative analysis, our proposed method depicted betterlesion segmentation with smoother shape of the lesion.

Evaluation of the Clinical and Imaging Findings of Breast Examinations in a Tertiary Facility in Ghana

Breast diseases have been one of the major battles the world has been fighting. In winning this fight, the role of medical imaging cannot be overlooked. Breast imaging reveals hidden lesions which aid physicians to give the appropriate diagnosis and definitive treatment, hence this study, to determine the clinical and imaging findings of breast examinations to document the radiologic features in our setting. This cross-sectional retrospective study reviewed the sociodemographics, imaging reports (mammography and ultrasonography with BI-RADS scores and their features), and the clinical data of 425 patients from September 2017 to September 2020 in the Cape Coast Teaching Hospital. 72 solid lesions with their histology reports were also reviewed. Data obtained were organized, coded, and analyzed using Statistical Package for Social Sciences (SPSS Inc., Chicago, IL, USA) version 20.0. The results obtained were presented in appropriate tables and charts. A chi-squared test was employed for associations and statistical significance was specified at p ≤ 0.05. 63.29% of the patients were married, but only 18.59% had a positive family history of breast cancer. BI-RADS scores 1(57.46%) and 2(27.99%) were the most recurrent findings. The most common BI-RADS 2, 3, 4, and 5 imaging features were benign-looking axillary lymph nodes (66.33%), well-defined solid masses (61.54%), ill-defined solid masses (42.86%), and ill-defined solid masses with suspicious-looking axillary lymph nodes (100.00%), respectively. The most frequent indications were routine screening (49.18%), mastalgia (26.59%), and painless breast masses (19.77%). There was significant association between duration of symptoms and breast cancer (p value = 0.007). In conclusion, routine breast screening and mastalgia were the topmost indications for breast imaging. BI-RADS 1 and 2 were the commonest BI-RADS scores, and benign-looking axillary lymph nodes and simple cysts were the most frequent imaging features for BI-RADS 2 and ill-defined solid masses and suspicious-looking axillary lymph nodes for BI-RADS 4 and 5.

Diabetes, Obesity, and Inflammation: Impact on Clinical and Radiographic Features of Breast Cancer

Obesity, diabetes, and inflammation increase the risk of breast cancer, the most common malignancy in women. One of the mainstays of breast cancer treatment and improving outcomes is early detection through imaging-based screening. There may be a role for individualized imaging strategies for patients with certain co-morbidities. Herein, we review the literature regarding the accuracy of conventional imaging modalities in obese and diabetic women, the potential role of anti-inflammatory agents to improve detection, and the novel molecular imaging techniques that may have a role for breast cancer screening in these patients. We demonstrate that with conventional imaging modalities, increased sensitivity often comes with a loss of specificity, resulting in unnecessary biopsies and overtreatment. Obese women have body size limitations that impair image quality, and diabetes increases the risk for dense breast tis-sue. Increased density is known to obscure the diagnosis of cancer on routine screening mammography. Novel molecu-lar imaging agents with targets such as estrogen receptor, human epidermal growth factor receptor 2 (HER2), pyrimi-dine analogues, and ligand-targeted receptor probes, among others, have potential to reduce false positive results. They can also improve detection rates with increased resolution and inform therapeutic decision making. These emerg-ing imaging techniques promise to improve breast cancer diagnosis in obese patients with diabetes who have dense breasts, but more work is needed to validate their clinical application.

Breast Cancer Image Classification: A Review

BACKGROUND

Breast cancer represents uncontrolled breast cell growth. Breast cancer is the most diagnosed cancer in women worldwide. Early detection of breast cancer improves the chances of survival and increases treatment options. There are various methods for screening breast cancer, such as mammogram, ultrasound, computed tomography and Magnetic Resonance Imaging (MRI). MRI is gaining prominence as an alternative screening tool for early detection and breast cancer diagnosis. Nevertheless, MRI can hardly be examined without the use of a Computer-Aided Diagnosis (CAD) framework, due to the vast amount of data.

OBJECTIVE

This paper aims to cover the approaches used in the CAD system for the detection of breast cancer.

METHODS

In this paper, the methods used in CAD systems are categories into two classes: the conventional approach and artificial intelligence (AI) approach.

RESULTS

The conventional approach covers the basic steps of image processing, such as preprocessing, segmentation, feature extraction and classification. The AI approach covers the various convolutional and deep learning networks used for diagnosis.

CONCLUSION

This review discusses some of the core concepts used in breast cancer and presents a comprehensive review of efforts in the past to address this problem.

Mammography Diagnostic Reference Levels (DRLs) in Ghana

INTRODUCTION

Diagnostic Reference Levels (DRLs) are essential for optimisation in mammography. A local DRL for screen-film mammography has been established in Ghana but none exists for the digital mammography systems. Furthermore, technological advancement is phasing out the use of screen-film mammography and replacing it with digital mammography systems. This study aims to establish the local DRLs used in digital mammography across three institutions in Ghana to guide mammography practice.

METHODS

Average glandular dose (AGD), compressed breast thickness (CBT), age of patients, entrance surface exposure (ESE), kVp, and mAs were retrospectively extracted from three digital mammography systems. The 75th and 95th percentile values were obtained for the AGD of each mammography projection and at CBT of 60 ± 5 mm. The correlation between the AGD and CBT, kVp, mAs, and ESE were investigated.

RESULTS

The 75th percentile for the AGD at CBT of 60 ± 5 mm for Centres 1, 2, 3, and all centres were 2.3, 1.8, 2.1, and 2.0 mGy respectively. The DRLs obtained were comparably higher than international studies except those of the United Kingdom. The AGD showed a strong positive correlation with the CBT, kVp, mAs, and ESE. There was variability in the AGD applied across the three centres for the craniocaudal (CC) and mediolateral oblique (MLO) projections. The mean AGD, mAs, and ESE for all the three centres and per centre recorded were higher than previous studies, but the mean kVp and CBT were lower than previous studies.

CONCLUSION

The higher DRLs estimated in this preliminary study indicates that there is a need for dose optimisation in digital mammography practice in Ghana to improve radiation protection.

IMPLICATIONS FOR PRACTICE

The findings will guide the process of optimisation and limit the variations in the radiation dose during mammography practice.

An introduction to the Cyrcadia Breast Monitor: A wearable breast health monitoring device

BACKGROUND

The most common breast cancer detection modalities are generally limited by radiation exposure, discomfort, high costs, inter-observer variabilities in image interpretation, and low sensitivity in detecting cancer in dense breast tissue. Therefore, there is a clear need for an affordable and effective adjunct modality that can address these limitations. The Cyrcadia Breast Monitor (CBM) is a non-invasive, non-compressive, and non-radiogenic wearable device developed as an adjunct to current modalities to assist in the detection of breast tissue abnormalities in any type of breast tissue.

METHODS

The CBM records thermodynamic metabolic data from the breast skin surface over a period of time using two wearable biometric patches consisting of eight sensors each and a data recording device. The acquired multi-dimensional temperature time series data are analyzed to determine the presence of breast tissue abnormalities. The objective of this paper is to present the scientific background of CBM and also to describe the history around the design and development of the technology.

RESULTS

The results of using the CBM device in the initial clinical studies are also presented. Twenty four-hour long breast skin temperature circadian rhythm data was collected from 93 benign and 108 malignant female study subjects in the initial clinical studies. The predictive model developed using these datasets could differentiate benign and malignant lesions with 78% accuracy, 83.6% sensitivity and 71.5% specificity. A pilot study of 173 female study subjects is underway, in order to validate this predictive model in an independent test population.

CONCLUSIONS

The results from the initial studies indicate that the CBM may be valuable for breast health monitoring under physician supervision for confirmation of any abnormal changes, potentially prior to other methods, such as, biopsies. Studies are being conducted and planned to validate the technology and also to evaluate its ability as an adjunct breast health monitoring device for identifying abnormalities in difficult-to-diagnose dense breast tissue.

An attention-supervised full-resolution residual network for the segmentation of breast ultrasound images

PURPOSE

Breast cancer is the most common cancer among women worldwide. Medical ultrasound imaging is one of the widely applied breast imaging methods for breast tumors. Automatic breast ultrasound (BUS) image segmentation can measure the size of tumors objectively. However, various ultrasound artifacts hinder segmentation. We proposed an attention-supervised full-resolution residual network (ASFRRN) to segment tumors from BUS images.

METHODS

In the proposed method, Global Attention Upsample (GAU) and deep supervision were introduced into a full-resolution residual network (FRRN), where GAU learns to merge features at different levels with attention for deep supervision. Two datasets were employed for evaluation. One (Dataset A) consisted of 163 BUS images with tumors (53 malignant and 110 benign) from UDIAT Centre Diagnostic, and the other (Dataset B) included 980 BUS images with tumors (595 malignant and 385 benign) from the Sun Yat-sen University Cancer Center. The tumors from both datasets were manually segmented by medical doctors. For evaluation, the Dice coefficient (Dice), Jaccard similarity coefficient (JSC), and F1 score were calculated.

RESULTS

For Dataset A, the proposed method achieved higher Dice (84.3  ±  10.0%), JSC (75.2  ±  10.7%), and F1 score (84.3  ±  10.0%) than the previous best method: FRRN. For Dataset B, the proposed method also achieved higher Dice (90.7  ±  13.0%), JSC (83.7  ±  14.8%), and F1 score (90.7  ±  13.0%) than the previous best methods: DeepLabv3 and dual attention network (DANet). For Dataset A + B, the proposed method achieved higher Dice (90.5  ±  13.1%), JSC (83.3  ±  14.8%), and F1 score (90.5  ±  13.1%) than the previous best method: DeepLabv3. Additionally, the parameter number of ASFRRN was only 10.6 M, which is less than those of DANet (71.4 M) and DeepLabv3 (41.3 M).

CONCLUSIONS

We proposed ASFRRN, which combined with FRRN, attention mechanism, and deep supervision to segment tumors from BUS images. It achieved high segmentation accuracy with a reduced parameter number.

Compression force variability in mammography in Ghana - A baseline study

INTRODUCTION

Breast compression during mammographic examinations improves image quality and patient management. Several studies have been conducted to assess compression force variability among practitioners in order to establish compression guidelines. However, no such study has been conducted in Ghana. This study aims to investigate the compression force variability in mammography in Ghana.

METHODS

This retrospective study used data gathered from 1071 screening and diagnostic mammography patients from January, 2018-December, 2019. Data were gathered by seven radiographers at three centers. Compression force, breast thickness and practitioners' years of work experience were recorded. Compression force variability among practitioners and the correlation between compression force and breast thickness were investigated.

RESULTS

Mean compression force values recorded for craniocaudal (CC) (17.2 daN) and mediolateral oblique (MLO) (18.2 daN), were within the recommended values used by western countries. Most of the mammograms performed - 80% - were within the National Health Service Breast Screening Programme (NHSBSP) range. However, 65% were above the Norwegian Breast Cancer Screening Programme (NBCSP) range. Compression forces varied significantly (p = 0.0001) among practitioners. Compression forces increased significantly (p = 0.0001) with the years of work experience. A weak negative correlation (r = -0.144) and a weak positive correlation (r = 0.142) were established between compression force and breast thickness for CC and MLO projections respectively.

CONCLUSION

This initial study confirmed that although wide variations in compression force exist among practitioners in Ghana, most practitioners used compression forces broadly within the range set by the NHSBSP. As no national guidelines for compression force currently exist in Ghana, provision of these may help to reduce the range of variations recorded.

IMPLICATIONS FOR PRACTICE

Confirmation of variations in compression will guide future practice to minimize image quality disparities and improve quality of care.

Surface-Enhanced Raman Spectroscopy Characterization of Breast Cell Phenotypes: Effect of Nanoparticle Geometry

The use of surface-enhanced Raman spectroscopy (SERS) to delineate between the breast epithelial cell lines MCF10A, SK-BR-3, and MDA-MB-231 is explored utilizing varied morphologies of gold nanoparticles. The nanoparticles studied had spherical, star-like, and quasi-fractal (nanocaltrop) morphologies and possessed varying degrees of surface inhomogeneity and complexity. The efficacy of Raman enhancement of these nanoparticles was a function of their size, their surface morphology, and the associated density of "hot spots," as well as their cellular uptake. The spherical and star-like nanoparticles provided strong signal enhancement that allowed for the discernment among the three cell phenotypes based solely on the acquired Raman spectra. The presence of overlapping Raman band spectral regions, as well as unique spectral bands, suggests that the underlying biological differences between these cells can be accessed without the need for tagging the nanoparticles or for specific cell targeting, demonstrating the potential ubiquity of this technique in imaging any cancer. This work provides clear evidence for the potential application of SERS as a tool for mapping cancerous lesions, possibly during surgery and under histopathological analysis.

Estimation of cancer risks during mammography procedure in Saudi Arabia

The aims of the present work were to quantify radiation doses arises from patients' exposure in mammographic X-ray imaging procedures and to estimate the radiation induced cancer risk. Sixty patients were evaluated using a calibrated digital mammography unit at King Khaled Hospital and Prince Sultan Center, Alkharj, Saudi Arabia. The average patient age (years) was 44.4 ± 10 (26-69). The average and range of exposure parameters were 29.1 ± 1.9 (24.0-33.0) and 78.4 ± 17.5 (28.0-173.0) for X-ray tube potential (kVp) and current multiplied by the exposure time (s) (mAs), respectively. The MGD (mGy) per single projection for craniocaudal (CC), Medio lateral oblique (MLO) and lateromedial (LM) was 1.02 ± 0.2 (0.4-1.8), 1.1 ± 0.3 (0.5-1.8), 1.1 ± 0.3 (0.5-1.9) per procedure, in that order. The average cancer risk per projection is 177 per million procedures. The cancer risk is significant during multiple image acquisition. The study revealed that 80% of the procedures with normal findings. However, precise justification is required especially for young patients.

Review and Comparison of Cancer Biomarker Trends in Urine as a Basis for New Diagnostic Pathways

Cancer is one of the major causes of mortality worldwide and its already large burden is projected to increase significantly in the near future with a predicted 22 million new cancer cases and 13 million cancer-related deaths occurring annually by 2030. Unfortunately, current procedures for diagnosis are characterized by low diagnostic accuracies. Given the proved correlation between cancer presence and alterations of biological fluid composition, many researchers suggested their characterization to improve cancer detection at early stages. This paper reviews the information that can be found in the scientific literature, regarding the correlation of different cancer forms with the presence of specific metabolites in human urine, in a schematic and easily interpretable form, because of the huge amount of relevant literature. The originality of this paper relies on the attempt to point out the odor properties of such metabolites, and thus to highlight the correlation between urine odor alterations and cancer presence, which is proven by recent literature suggesting the analysis of urine odor for diagnostic purposes. This investigation aims to evaluate the possibility to compare the results of studies based on different approaches to be able in the future to identify those compounds responsible for urine odor alteration.

Reviewing ensemble classification methods in breast cancer

CONTEXT

Ensemble methods consist of combining more than one single technique to solve the same task. This approach was designed to overcome the weaknesses of single techniques and consolidate their strengths. Ensemble methods are now widely used to carry out prediction tasks (e.g. classification and regression) in several fields, including that of bioinformatics. Researchers have particularly begun to employ ensemble techniques to improve research into breast cancer, as this is the most frequent type of cancer and accounts for most of the deaths among women.

OBJECTIVE AND METHOD

The goal of this study is to analyse the state of the art in ensemble classification methods when applied to breast cancer as regards 9 aspects: publication venues, medical tasks tackled, empirical and research types adopted, types of ensembles proposed, single techniques used to construct the ensembles, validation framework adopted to evaluate the proposed ensembles, tools used to build the ensembles, and optimization methods used for the single techniques. This paper was undertaken as a systematic mapping study.

RESULTS

A total of 193 papers that were published from the year 2000 onwards, were selected from four online databases: IEEE Xplore, ACM digital library, Scopus and PubMed. This study found that of the six medical tasks that exist, the diagnosis medical task was that most frequently researched, and that the experiment-based empirical type and evaluation-based research type were the most dominant approaches adopted in the selected studies. The homogeneous type was that most widely used to perform the classification task. With regard to single techniques, this mapping study found that decision trees, support vector machines and artificial neural networks were those most frequently adopted to build ensemble classifiers. In the case of the evaluation framework, the Wisconsin Breast Cancer dataset was the most frequently used by researchers to perform their experiments, while the most noticeable validation method was k-fold cross-validation. Several tools are available to perform experiments related to ensemble classification methods, such as Weka and R Software. Few researchers took into account the optimisation of the single technique of which their proposed ensemble was composed, while the grid search method was that most frequently adopted to tune the parameter settings of a single classifier.

CONCLUSION

This paper reports an in-depth study of the application of ensemble methods as regards breast cancer. Our results show that there are several gaps and issues and we, therefore, provide researchers in the field of breast cancer research with recommendations. Moreover, after analysing the papers found in this systematic mapping study, we discovered that the majority report positive results concerning the accuracy of ensemble classifiers when compared to the single classifiers. In order to aggregate the evidence reported in literature, it will, therefore, be necessary to perform a systematic literature review and meta-analysis in which an in-depth analysis could be conducted so as to confirm the superiority of ensemble classifiers over the classical techniques.

Visual search in breast imaging

Breast cancer is the most common cancer among females worldwide and large volumes of breast images are produced and interpreted annually. As long as radiologists interpret these images, the diagnostic accuracy will be limited by human factors and both false-positive and false-negative errors might occur. By understanding visual search in breast images, we may be able to identify causes of diagnostic errors, find ways to reduce them, and also provide a better education to radiology residents. Many visual search studies in breast radiology have been devoted to mammography. These studies showed that 70% of missed lesions on mammograms attract radiologists' visual attention and that a plethora of different reasons, such as satisfaction of search, incorrect background sampling, and incorrect first impression can cause diagnostic errors in the interpretation of mammograms. Recently, highly accurate tools, which rely on both eye-tracking data and the content of the mammogram, have been proposed to provide feedback to the radiologists. Improving these tools and determining the optimal pathway to integrate them in the radiology workflow could be a possible line of future research. Moreover, in the past few years deep learning has led to improving diagnostic accuracy of computerized diagnostic tools and visual search studies will be required to understand how radiologists interact with the prompts from these tools, and to identify the best way to utilize them. Visual search in other breast imaging modalities, such as breast ultrasound and digital breast tomosynthesis, have so far received less attention, probably due to associated complexities of eye-tracking monitoring and analysing the data. For example, in digital breast tomosynthesis, scrolling through the image results in longer trials, adds a new factor to the study's complexity and makes calculation of gaze parameters more difficult. However, considering the wide utilization of three-dimensional imaging modalities, more visual search studies involving reading stack-view examinations are required in the future. To conclude, in the past few decades visual search studies provided extensive understanding about underlying reasons for diagnostic errors in breast radiology and characterized differences between experts' and novices' visual search patterns. Further visual search studies are required to investigate radiologists' interaction with relatively newer imaging modalities and artificial intelligence tools.

Optoacoustic imaging of the breast: correlation with histopathology and histopathologic biomarkers

Aim

This study was conducted in order to investigate the role of gray-scale ultrasound (US) and optoacoustic imaging combined with gray-scale ultrasound (OA/US) to better differentiate between breast cancer molecular subtypes.

Materials and methods

All 67 malignant masses included in the Maestro trial were retrospectively reviewed to compare US and OA/US feature scores and histopathological findings. Kruskal–Wallis tests were used to analyze the relationship between US and OA/US features and molecular subtypes of breast cancer. If a significant relationship was found, additional Wilcoxon–Mann–Whitney tests were used to identify the differences between molecular subtype groups.

Results

US sound transmission helped to differentiate between LUMA and LUMB, LUMB and TNBC, and LUMB and all other molecular subtypes combined (p values < 0.05). Regarding OA/US features, the sum of internal features helped to differentiate between TNBC and HER2-enriched subtypes (p = 0.049). Internal vessels (p = 0.025), sum of all internal features (p = 0.019), and sum of internal and external features (p = 0.028) helped to differentiate between LUMA and LUMB. All internal features, the sum of all internal features, the sum of all internal and external features, and the ratio of internal and external features helped to differentiate between LUMA and TNBC. The same features also helped to differentiate between LUMA and TNBC from other molecular subtypes (p values < 0.05).

Conclusions

The use of OA/US might help radiologists to better differentiate between breast cancer molecular subtypes. Further studies need to be carried out in order to validate these results.

Key Points

• The combination of functional and morphologic information provided by optoacoustic imaging (OA) combined with gray-scale US helped to differentiate between breast cancer molecular subtypes.

Electronic supplementary material

The online version of this article (10.1007/s00330-019-06262-0) contains supplementary material, which is available to authorized users.

Attributes, Performance, and Gaps in Current & Emerging Breast Cancer Screening Technologies

Background:

Early detection of breast cancer, combined with effective treatment, can reduce mortality. Millions of women are diagnosed with breast cancer and many die every year globally. Numerous early detection screening tests have been employed. A wide range of current breast cancer screening methods are reviewed based on a series of searchers focused on clinical testing and performance. </P><P> Discussion: The key factors evaluated centre around the trade-offs between accuracy (sensitivity and specificity), operator dependence of results, invasiveness, comfort, time required, and cost. All of these factors affect the quality of the screen, access/eligibility, and/or compliance to screening programs by eligible women. This survey article provides an overview of the working principles, benefits, limitations, performance, and cost of current breast cancer detection techniques. It is based on an extensive literature review focusing on published works reporting the main performance, cost, and comfort/compliance metrics considered.

Conclusion:

Due to limitations and drawbacks of existing breast cancer screening methods there is a need for better screening methods. Emerging, non-invasive methods offer promise to mitigate the issues particularly around comfort/pain and radiation dose, which would improve compliance and enable all ages to be screened regularly. However, these methods must still undergo significant validation testing to prove they can provide realistic screening alternatives to the current accepted standards.

Iterative simulations to estimate the elastic properties from a series of MRI images followed by MRI-US validation

The modeling of breast deformations is of interest in medical applications such as image-guided biopsy, or image registration for diagnostic purposes. In order to have such information, it is needed to extract the mechanical properties of the tissues. In this work, we propose an iterative technique based on finite element analysis that estimates the elastic modulus of realistic breast phantoms, starting from MRI images acquired in different positions (prone and supine), when deformed only by the gravity force. We validated the method using both a single-modality evaluation in which we simulated the effect of the gravity force to generate four different configurations (prone, supine, lateral, and vertical) and a multi-modality evaluation in which we simulated a series of changes in orientation (prone to supine). Validation is performed, respectively, on surface points and lesions using as ground-truth data from MRI images, and on target lesions inside the breast phantom compared with the actual target segmented from the US image. The use of pre-operative images is limited at the moment to diagnostic purposes. By using our method we can compute patient-specific mechanical properties that allow compensating deformations. Graphical Abstract Workflow of the proposed method and comparative results of the prone-to-supine simulation (red volumes) validated using MRI data (blue volumes).

Downgrading of Breast Masses Suspicious for Cancer by Using Optoacoustic Breast Imaging

Purpose To assess the ability of optoacoustic (OA) ultrasonography (US) to help correctly downgrade benign masses classified as Breast Imaging Reporting and Data System (BI-RADS) 4a and 4b to BI-RADS 3 or 2. Materials and Methods OA/US technology uses laser light to detect relative amounts of oxygenated and deoxygenated hemoglobin in and around suspicious breast masses. In this prospective, multicenter study, results of 209 patients with 215 breast masses classified as BI-RADS 4a or 4b at US are reported. Patients were enrolled between 2015 and 2016. Masses were first evaluated with US with knowledge of previous clinical information and imaging results, and from this information a US imaging-based probability of malignancy (POM) and BI-RADS category were assigned to each mass. The same masses were then re-evaluated at OA/US. During the OA/US evaluation, radiologists scored five OA/US features, and then reassigned an OA/US-based POM and BI-RADS category for each mass. BI-RADS downgrade and upgrade percentages at OA/US were assessed by using a weighted sum of the five OA feature scores. Results At OA/US, 47.9% (57 of 119; 95% CI: 0.39, 0.57) of benign masses classified as BI-RADS 4a and 11.1% (three of 27; 95% CI: 0.03, 0.28) of masses classified as BI-RADS 4b were correctly downgraded to BI-RADS 3 or 2. Two of seven malignant masses classified as BI-RADS 4a at US were incorrectly downgraded, and one of 60 malignant masses classified as BI-RADS 4b at US was incorrectly downgraded for a total of 4.5% (three of 67; 95% CI: 0.01, 0.13) false-negative findings. Conclusion At OA/US, benign masses classified as BI-RADS 4a could be downgraded in BI-RADS category, which would potentially decrease biopsies negative for cancer and short-interval follow-up examinations, with the limitation that a few masses may be inappropriately downgraded.

Whole blood microRNAs as potential biomarkers in post-operative early breast cancer patients

BACKGROUND

microRNAs (miRNAs) are considered promising cancer biomarkers, showing high reliability, sensitivity and stability. Our study aimed to identify associations between whole blood miRNA profiles, presence of circulating tumor cells (CTCs) and clinical outcome in post-operative early breast cancer patients (EBC) to assess the utility of miRNAs as prognostic markers in this setting.

METHOD

A total of 48 post-operative patients, recruited in frame of the SUCCESS A trial, were included in this retrospective study and tested with a panel of 8 miRNAs (miR-10b, -19a, - 21, - 22, -20a, - 127, - 155, -200b). Additional 17 female healthy donors with no previous history of cancer were included in the study as negative controls. Blood samples were collected at different time points (pre-adjuvant therapy, post-adjuvant therapy, 2 years follow up), total RNA was extracted and the relative concentration of each miRNA was measured by quantitative PCR and compared in patients stratified on blood collection time or CTC detection. Furthermore, we compared miRNA profiles of patients, for each time point separately, and healthy donors. CTCs were visualized and quantified with immunocytochemistry analysis. Data were analyzed using non-parametric statistical tests.

RESULTS

In our experimental system, miR-19a, miR-22 and miR-127 showed the most promising results, differentiating patients at different time points and from healthy controls, while miR-20a, miR-21 and miR-200b did not show any difference among the different groups. miR-10b and miR-155 were never detectable in our experimental system. With respect to patients' clinical characteristics, we found a significant correlation between miR-200b and lymph node status and between miR-20a and tumor type. Furthermore, miR-127 correlated with the presence of CTCs. Finally, we found a borderline significance between Progression Free Survival and miR-19a levels.

CONCLUSIONS

This pilot study suggests that profiling whole blood miRNAs could help to better stratify post-operative EBC patients without any sign of metastasis to prevent later relapse or metastatic events.