Breast Boundary Research Articles

Prepectoral Breast Reconstruction Prompts Revisiting the Anatomic Boundaries of the Breast: A Radiographic and Cadaveric study.

This study aimed to determine the location of superficial fascial system (SFS) condensations in relation to classic anatomic breast boundaries. Cadaveric studies have provided some understanding, but knowledge about the precise location of these condensations remains limited. Preoperative breast MRI was conducted to assess the fascial condensations defining the breast footprint relative to landmarks like the latissimus, clavicle, sternal border, and inframammary fold (IMF). Concurrently, cadaveric mastectomies were performed to measure the breast borders in relation to these landmarks for comparison. 290 breasts underwent preoperative MRI. Eight cadaveric breast dissections were completed. Radiographically, the lateral breast fascial condensation was 3.9 cm medial to the latissimus dorsi, while cadaveric measurements were 4.5 cm. The medial condensation was 2.1 cm lateral to the sternal border radiographically and 4.8 cm to midline in cadavers. The superior fascial condensation was 2.3 cm inferior to the clavicle radiographically and 5.5 cm by dissection. The inferior condensation was above the IMF in 82.7% of breasts by MRI and 100% of cadaveric breasts. MRI and cadaveric investigation showed similar patterns of breast tissue fascial condensations relative to standard breast boundaries. Breast skin flap thickness was greater peripherally (MRI: 11.5 mm, cadaver: 11.1 mm) than centrally (MRI: 6.6 mm, cadaver: 5.5 mm). MRI and cadaveric analysis demonstrated a close correlation between SFS and standard breast boundaries, although variations existed among subjects. Collaborating with breast oncologists and utilizing preoperative imaging to identify individual fascial condensations may enhance the camouflage of prepectoral implants after mastectomy.

Using Photogrammetry to Analyze Anatomical Changes in the Nipple in Juvenile and Adolescent Scoliosis Patients.

The need for an objective set of anterior trunk measurements, such as nipple and clavicular shoulder joints, is essential to quantify the anterior asymmetry present in scoliosis. This study aims to characterize breast asymmetry (BA) in young individuals with scoliosis using photogrammetry. Digital photographs of the anterior trunk of the 51 scoliosis patients aged 4-20years were taken from an anterior perspective. These images were then transferred to a computer. Ten parameters were measured using the ImageJ software. The positions of patients' nipples were classified into 6 types based on ratios on the x-axis. The majority of patients had a right breast that was larger, intensifying the apparent BA due to trunk rotation. The apical vertebra level in patients was found at T8 in 23.6% and T9 in 45.1%. In 92.5% of the patients, the right breast was consistently larger. The lengths between the lateral boundaries and nipples of the right and left breasts and between the medial boundary and nipple of the right breast were statistically significantly higher in males than in females (p < 0.05). Significant differences were found when comparing the values of the lengths between the medial boundaries and nipples of the right and left breasts, the difference in length between the right and left acromioclavicular joint lines, and the angles of the nipple and acromioclavicular joint with the degrees of scoliosis in juvenile and adolescents (p < 0.05). Pearson regression analysis revealed a significant correlation between BA differences and the Cobb angle with a correlation coefficient of 0.901. Factors related to breast aesthetics, like differences in the height of nipples and the distance from the sternal notch to the nipple, represent 30% of the overall score. The study concluded that there is a significant correlation between the severity of scoliosis and BA differences. Augmentation mammaplasty for BA not only decreased the breast difference but also leveled the nipple disparities. Photogrammetry is considered to be an alternative to other methods and is believed to contribute to the follow-up of BA. This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Development of automated region of interest selection algorithms for surface-guided radiation therapy of breast cancer.

To investigate automation of the preparation of the region of interest (ROI) for surface-guided radiotherapy (SGRT) of the whole breast with two algorithms based on contour anatomies: using the body contour, and using the breast contour. The patient dataset used for modeling consisted of 39 breast cancer patients previously treated with SGRT. The patient's anatomical structures (body and ipsilateral breast) were retrieved from the planning system, and the clinical ROI (cROI) drawn by the planners was retrieved from the SGRT system for comparison. For the body-contour-based algorithm, a convolutional neural network (MobileNet-v2) was utilized to train a synthetic human model dataset to predict body joint locations. With the body joint location knowledge, an automated ROI (aROIbody ) can be created based on: (1) the superior-inferior (S-I) borders defined by the joint locations, (2) the left-right (L-R) borders defined with 3/4 of chest width, and (3) a curation of the ROI to avoid the ipsilateral armpit. For the breast-contour-based algorithm, an aROIbreast was created by first defining the ROI in the S-I direction with the ipsilateral breast boundaries. Other steps are the same as with the body-contour-based algorithm. Among the 39 patients, 24 patients were used to fine-tune the algorithm parameters, and the remaining 15 patients were used to evaluate the quality of the aROIs against the cROIs. A blinded evaluation was performed by three SGRT expert physicists to rate the acceptability and the quality (1-10 scale) of the aROIs and cROIs, and the dice similarity coefficient (DSC) was also calculated to compare the similarity between the aROIs and cROIs. The results showed that the average acceptability was 14/15 (range: 13/15-15/15) for cROIs, 13.3/15 (range: 13/15-14/15) for aROIbody , and 14.6/15 (range: 14/15-15/15) for aROIbreast . The average quality was 7.4 ± 0.8 for cROIs, 8.1 ± 1.2 for aROIbody , and 8.2 ± 0.9 for aROIbreast . The DSC with cROIs was 0.81 ± 0.06 for aROIbody , and 0.83 ± 0.04 for aROIbreast . The ROI creation time was ∼120s for clinical, 1.3s for aROIbody , and 1.2s for aROIbreast . The proposed automated algorithms can improve the ROI compliance with the SGRT protocol, with a shortened preparation time. It is ready to be integrated into the clinical workflow for automated ROI preparation.

Impact of GAN artifacts for simulating mammograms on identifying mammographically occult cancer.

Generative adversarial networks (GANs) can synthesize various feasible-looking images. We showed that a GAN, specifically a conditional GAN (CGAN), can simulate breast mammograms with normal, healthy appearances and can help detect mammographically-occult (MO) cancer. However, similar to other GANs, CGANs can suffer from various artifacts, e.g., checkerboard artifacts, that may impact the quality of the final synthesized image, as well as the performance of detecting MO cancer. We explored the types of GAN artifacts that exist in mammogram simulations and their effect on MO cancer detection. We first trained a CGAN using digital mammograms (FFDMs) of 1366 women with normal/healthy breasts. Then, we tested the trained CGAN on an independent MO cancer dataset with 333 women with dense breasts (97 MO cancers). We trained a convolutional neural network (CNN) on the MO cancer dataset, in which real and simulated mammograms were fused, to identify women with MO cancer. Then, a radiologist who was independent of the development of the CGAN algorithms evaluated the entire MO cancer dataset to identify and annotate artifacts in the simulated mammograms. We found four artifact types, including checkerboard, breast boundary, nipple-areola complex, and black spots around calcification artifacts, with an overall incidence rate over 69% (the individual incident rate ranged from 9% to 53%) from both normal and MO cancer samples. We then evaluated their potential impact on MO cancer detection. Even though various artifacts existed in the simulated mammogram, we found that it still provided complementary information for MO cancer detection when it was combined with the real mammograms. We found that artifacts were pervasive in the CGAN-simulated mammograms. However, they did not negatively affect our MO cancer detection algorithm; the simulated mammograms still provided complementary information for MO cancer detection when combined with real mammograms.

Automated Breast Boundary Segmentation to Improve the Accuracy of Identifying Abnormalities in Breast Thermograms

Breast thermography is a non-invasive, painless, affordable, and safer method of detecting breast abnormalities. Thermal imaging techniques used for detecting breast anomalies rely on precise breast boundary segmentation. However, the segmentation accuracy of the breast boundary region is affected by unclear boundaries, a low signal-to-noise ratio, and poor contrast of the thermal images. To mitigate this, this paper proposes a novel Distance-based Metrics and High-Temperature Region-based Adaptive Thresholding (DM-HTRAT) method for breast boundary region segmentation and dissecting left and right breasts precisely. In the first section, the Upper Region Boundary (URB) and the intersection point of the breast are segmented using the Distance-based Metrics (DM) method. In the second section, a High-Temperature Region-based Adaptive Thresholding (HTRAT) method is used to segment the Lower Region Boundary (LRB). The result of the segmentation is analysed both quantitatively and qualitatively, also it is compared with the state-of-the-art methods. The proposed method has a significantly better accuracy of 96.5% than the other methods. This segmented image is then fed into the abnormality detection system developed via the asymmetrical analysis of segmented breast boundary regions. The proposed DM-HTRAT-based breast boundary segmentation is robust in segmenting all types of breast thermal images and aids in the more reliable and effective detection of breast abnormalities.

Abstract PD14-04: PD14-04 Whole mammogram image-based Cox regression improves 5-year breast cancer prediction

Abstract To efficiently capture data from mammographic breast images and classify long-term risk of breast cancer, we developed FLIP, a novel Cox regression-based framework that fully utilizes data in the mammograms beyond current density measures. FLIP use the extensive existing data that are currently ignored in the context of breast cancer risk stratification. More than 20 studies support texture features add value to risk prediction beyond breast density. However, the entire mammogram imaging data has a high dimension of pixels (~13 million per image), greatly exceeding the number of women in a cohort. FLIP was fitted and cross-validated within the Joanne Knight Breast Health Cohort excluding cases diagnosed in the first 6 months of entry. The Joanne Knight Breast Health Cohort is comprised of over 10,000 women undergoing repeated mammography screening at Siteman Cancer Center and followed since 2010. All women had baseline mammogram at entry, provided a blood sample and completed a risk factor questionnaire. Mammograms are all using the same technology (Hologic). During follow-up through October 2020, we identified 246 incident breast cancer cases (pathology confirmed) and matched them to controls from the perspective cohort based on month of mammogram and age at entry. We obtained an AUC of 0.68 (SE 0.03) including the whole mammogram image, age and BI-RADS (4th edition) density category; and AUC of 0.72 (SE 0.04) by adding in BMI and menopausal status to this model. These 5-year prediction performances exceed that of well-developed models based on epidemiologic risk factors (P &amp;lt; 0.001). FLIP offers standard statistical solutions and removes barriers to wider clinical use without prohibitive training data and extensive computational requirements, providing a transparent workflow ensuring high reproducibility. It should be accessible anywhere mammograms are used. We conclude that using full mammogram images for breast cancer risk prediction captures additional information on breast tissue characteristics that relate to cancer risk, and improves prediction classification. This prediction algorithm can run efficiently in real time (in seconds) with processing of digital mammograms. Thus, this model can be easily implemented in mammography screening services and other clinical settings to guide real-time risk stratification to improve precision prevention of the leading cancer in women world-wide. Further analysis will quantify the value of adding other breast cancer risk factors, including polygenic risk scores. Addition of repeated mammogram images over time should further increase classification performance. This approach has the potential to improve risk classification by using data already available for the vast majority of women already having repeated screening mammograms. Schema overview of FLIP The raw images are in the form of .dcm files before entering into FLIP. After automated processing and image alignment, the two CC-views (left and right) are average between the two breasts for characterization. The inputted 2D mammograms are first characterized with bivariate splines over triangulation to preserve spatial distribution of pixels and accommodate the irregular semi-circular breast boundary. The characterization is further optimized (see Supplemental Material) which provides a unique and closed-form solution. b. A simple Cox proportional hazards model is adopted using well-established risk factors (RF), including age, breast density (BI-RADS), BMI, menopausal status, parity, family history, and history of benign breast disease. The mammogram image acts as an additional risk factor in the Cox regression accompanied with a 2D coefficient surface. All inferential procedures with Cox regression are applicable to FLIP which provides a transparent workflow ensuring high reproducibility. h_i (t) denotes the hazard function at time t for individual i, and h_0 (t) denotes the nonparametric baseline hazard function. c. Women who are diagnosed with breast cancer within the first 6 month of their mammogram date have been removed from this analysis and we focus on the 5-year risk. Discriminatory performance is assessed with AUC and validated via a 10-fold cross-validation. Citation Format: Shu Jiang, Graham A. Colditz. PD14-04 Whole mammogram image-based Cox regression improves 5-year breast cancer prediction [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD14-04.

A Novel Model for Breast Cancer Detection and Classification

Breast cancer is a dreadful disease that affects women globally. The occurrences of masses in the breast region are the main cause of breast cancer development. It is important to detect breast cancer as early as possible as this might increase the survival rate. The existing research methodologies have the problems of increased computation complexity and low detection accuracy. To overcome such problems, this paper proposes an efficient breast cancer detection and classification system based on mammogram images. Initially, the mammogram images are preprocessed so unwanted regions and noise are removed and the contrast of the images is enhanced using Homo Morphic Adaptive Histogram Equalization (HMAHE). Then, the breast boundaries are identified with the use of the canny edge detector. After that, the pectoral muscles present in the images are detected and removed using the Global Pixel Intensity-based Thresholding (GPIT) method. Then, the tumors are identified and segmented by the Centroid-based Region Growing Segmentation (CRGS) algorithm. Next, the tumors are segmented and clustered and feature extraction is carried out from the clustered tumors. After that, the necessary features are selected by using the Chaotic Function-based Black Widow Optimization Algorithm (CBWOA). The selected features are utilized by the Convolutional Squared Deviation Neural Network Classifier (CSDNN) which classifies the tumors into six different categories. The proposed model effectively detects and classifies breast tumors and its efficiency is experimentally proved by comparison with the existing techniques.

Breast shape classification and discrimination driven by local features-focusing on Chinese women in their 20s

In order to analyze the breast shapes and provide basis for the development of bra sizing system and garment pattern-making, this study proposed a method to classify breast shapes from breast local features and overall shape based on the detailed breast morphological parameters. A definition for the breast boundary was determined to establish a measuring standard for the feature points, and then 18 related parameters were obtained based on the 3D point-cloud data of 158 unmarried young female college students aged between 18 and 25. Six shape parameters were selected for cluster analysis. In addition, the gathering ratio was induced to classify the breast shapes of subjects into nine categories that are FC-Middle, FC-Gathering, FC-Separating, UO-Middle, UO-Gathering, UO-Separating, PO-Middle, PO-Gathering and PO-Separating. Through verification, it shows that the error range of 95.71% UBP and 91.43% OBP are all less than 1.5 cm, while the error range of 81.43% UBP and 74.29% OBP are less than 1 cm. The accuracy of the classified results via the proposed method is 97.33%. The classification of breast shapes of this study can provide a reference for young women to choose the correct size of bra.

Breast tumor segmentation on medical images using combination of fuzzy clustering and threshold

Breast tumor segmentation and boundary detection are crucial stages in breast cancer therapy and follow-up. Radiologists can minimize the high workload associated with screening for breast cancer by automating this complex process. This article proposes a system for segmenting breast and unaffected areas (breast) tumors on medical images using a combination of fuzzy clustering and thresholding (FCMT) tools. This computer diagnostic method works with each section of the mammary gland without learning segmentation and definition of boundaries. As part of the approbation, two databases of breast mammogram images were used. To increase the image quality, we used pre-processing techniques such as contrast augmentation before applying the FCMT for segmentation. To assess the effectiveness of the devised approach, the Mean Square Error, dice coefficient, Structured Similarity Index, Peak Signal-to-Noise Ratio, accuracy, and sensitivity were computed. Using the proposed FCMT segmentation technique, a mean intersection over union (IoU) of 93.85 was attained. The presented approach is more resilient and accurate in segmenting tumor progression on medical images, according to the findings of the experiments.

Multicriteria Evaluation of Deep Neural Networks for Semantic Segmentation of Mammographies

Breast segmentation plays a vital role in the automatic analysis of mammograms. Accurate segmentation of the breast region increments the probability of a correct diagnostic and minimizes computational cost. Traditionally, model-based approaches dominated the landscape for breast segmentation, but recent studies seem to benefit from using robust deep learning models for this task. In this work, we present an extensive evaluation of deep learning architectures for semantic segmentation of mammograms, including segmentation metrics, memory requirements, and average inference time. We used several combinations of two-stage segmentation architectures composed of a feature extraction net (VGG16 and ResNet50) and a segmentation net (FCN-8, U-Net, and PSPNet). The training examples were taken from the mini Mammographic Image Analysis Society (MIAS) database. Experimental results using the mini-MIAS database show that the best net scored a Dice similarity coefficient of 99.37% for breast boundary segmentation and 95.45% for pectoral muscle segmentation.

An automatic and efficient technique for tumor location identification and classification through breast MR images

AimThe basic objective of this paper is to develop a single structured algorithm to classify the breast tissues into normal or abnormal. Material & MethodologyFor this study, the breast MR images dataset of 448 images collected from Healthmap diagnostics centre, PGIMS, Rohtak, India. The proposed algorithm consists of several steps i.e. an integrated (Median Wiener & Median) filtering technique is used for de-noising; breast boundary region extraction via selection of nipple and mid- sternum points to make the image rotation invariant; determined the tumor region intensity by using morphological operations & hole filling; classify the normal and abnormal breast tissues by SVM using 14 texture features extracted through GLCM & 13 morphological or kinetic features; evaluated the exact location as well as area of abnormal tissues. ResultsThe proposed algorithm has been evaluated statistically as well as visually. The quality parameters achieved are accuracy, sensitivity and specificity with values 0.937, 0.956 and 0.872 respectively. The Jaccard Index coefficient achieved is 0.921, which indicates promising overlap between the predicted tumor and the manually done image by the radiologist so called ground truth image. ConclusionThis work may be taken as a second opinion by the radiologists. The evaluated results may give a basic foundation for optimization by selecting the features more precisely and also different evolutionary algorithms using multi-classifiers can be designed in future.

Electronic bandgap miniaturized UWB antenna for near-field microwave investigation of skin

Near-field microwave investigation and tomography has many practical applications, especially where the trend of fields and signals in different environments is vital. This article shows an elliptical patch ultra-wideband antenna fed by a transmission line for the near-field characterization of cancerous cells in the skin. The antenna comprises an elliptical patch, stub loading to shift the band to lower bands, and an electronic bandgap structure on the ground side. Even though the antenna has a low profile of 15 × 15 mm2, the proposed antenna has more promising results than recent studies. Furthermore, both simulated near-field and far-field results show a broad bandwidth of 3.9–30 GHz and a resonance at 2.4 GHz applicable for industrial, scientific, and medical band applications. The proposed antenna also illustrates a peak gain of 6.48 dBi and a peak directivity of 7.09 dBi. Free space and skin (on a layer of breast fat and a tumor with a diameter of 4 mm at the boundary of skin and breast) are used as test environments during the simulation and measurement of near-field and far-field investigations while considering a phantom breast shape. Both far-field and near-field microwave investigations are performed in Computer Simulation Technology studio, and results are then compared with the measured data. The simulated and measured results are in good agreement, and the focused energy around the tumor is completely reconstructed. Therefore, the proposed antenna can be an adequate candidate for the differentiation of breast skin and tumor to reconstruct the tumor’s image.

Abstract PS1-36: Outcomes and management of positive and close anterior margins following skin-sparing mastectomy

Abstract Background: Skin-sparing mastectomy (SSM) has emerged as a safe oncologic technique for extirpation of breast tissue in the context of immediate breast reconstruction (IBR). Documented rates of local recurrence are comparable for SSM and conventional mastectomy and attention to the plane of dissection is essential with removal of skin overlying the tumor when clinically indicated. The incidence of positive or close margins is greatest for peripheral tumors at the breast boundaries and will influence rates of local recurrence. Management of positive or close margins is inconsistent and this audit aimed to determine the incidence of compromised margins and their impact on local recurrence and overall survival in SSM patients. Methods: A retrospective analysis was undertaken of breast patients with invasive or non-invasive breast cancer undergoing SSM and IBR at a single tertiary referral centre between January 2006 and December 2009. A total of 150 patients were included and all underwent resection of breast tissue with a peri-areolar incision. Clinical information was extracted from a prospectively maintained database. Data was collected on patient demographics, tumor characteristics, non-surgical treatment and outcome events (recurrence and death). The definition of a negative margin on histology during the study period was tumor ≥2mm from the edge of the specimen with close margins &amp;lt;2mm but no ink on tumor. Results: The mean age of patients was 51 years (range 24 - 75) and median duration of follow up 140 months (range 10 - 167). Amongst these 150 SSM patients, 25 (17%) had positive or close anterior margins (&amp;lt;2mm) with 125 patients having negative anterior margins (&amp;gt;2mm). None of these patients with close or positive margins underwent re-excision following initial SSM. Twenty-four patients (16%) developed either loco-regional (n=9) or distant recurrence (n=15) with 126 patients (84%) alive at 10 years. Although more patients with positive/close compared with negative margins had recurrence (20% versus 15%), this did not reach statistical significance (p=0.55) Similar proportions of patients in each margin category received post-mastectomy radiotherapy (p=0.66) and adjuvant/neo-adjuvant chemotherapy (p=0.66). There were no statistically significant differences in rates of local recurrence or survival between patients with positive/close and negative margins but the number of events is small and may represent a type II error. Conclusion: Twenty percent of patients (5/25) with positive or close margins after SSM develop local or distant recurrence with reduction of risk by PMRT of &amp;lt;50%. None of the patients with positive/close margins who developed recurrence survived for 10 years. Close/positive margins after SSM may portend a worse outcome from recurrent disease and further research is required to optimize management of this group of patients in an era of skin-preserving mastectomy. Citation Format: Primeera Wignarajah, Dorin Dumitru, John R Benson. Outcomes and management of positive and close anterior margins following skin-sparing mastectomy [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS1-36.

Breast Cancer Mass Detection in Mammograms Using Gray Difference Weight and MSER Detector

Breast cancer is a deadly and one of the most prevalent cancers in women across the globe. Mammography is widely used imaging modality for diagnosis and screening of breast cancer. Segmentation of breast region and mass detection are crucial steps in automatic breast cancer detection. Due to the non-uniform distribution of various tissues, it is a challenging task to analyze mammographic images with high accuracy. In this paper, background suppression and pectoral muscle removal are performed using gradient weight map followed by gray difference weight and fast marching method. Enhancement of breast region is performed using contrast limited adaptive histogram equalization (CLAHE) and de-correlation stretch. Detection of breast masses is accomplished by gray difference weight and maximally stable external regions (MSER) detector. Experimentation on Mammographic Image Analysis Society (MIAS) and curated breast imaging subset of digital database for screening mammography (CBIS-DDSM) show that the method proposed performs breast boundary segmentation and mass detection with best accuracies. Mass detection achieved high accuracies of about 97.64% and 94.66% for MIAS and CBIS-DDSM dataset, respectively. The method is simple, robust, less affected to noise, density, shape and size which could provide reasonable support for mammographic analysis.

Accurate construction of 3-D numerical breast models with anatomical information through MRI scans

The realistic numerical breast model is an important tool for verifying algorithms, improving design concepts, and exploring new technologies related to microwave breast cancer detection. Recently, several numerical breast models have been developed. However, these models do not include the real breast boundary and the chest region, which causes the electromagnetic wave propagation in the breast models to deviate from the actual situation. The proposed breast models in this paper overcome the significant deficiencies of the recent breast models in terms of structure completeness and authenticity. The model construction based on magnetic resonance imaging (MRI) scans is a multistep approach. Firstly, intensity inhomogeneity in MRI images is corrected by an improved nonparametric nonuniform intensity normalization (N4) algorithm. Then, a dual threshold and morphological transformation (DTMT) method is developed to segment the real breast region. Subsequently, a modified maximum inter-class variance (MICV) method is employed to automatically divide the breast region into the fat and the fibroglandular clusters with desired complexity. Finally, the dielectric properties of breast tissues are calculated by a piecewise weighted mapping method. The applicability and effectiveness of the proposed method are verified by constructing four breast models with varying tissue density, shape, and size. The constructed models in this study are significant to achieve reliable and precise effects for microwave breast cancer detection due to the accurate expression of the breast anatomical information and electromagnetic characteristics.

Evaluating attenuation correction strategies in a dedicated, single-gantry breast PET-tomosynthesis scanner

We are developing a dedicated, combined breast positron emission tomography (PET)-tomosynthesis scanner. Both the PET and digital breast tomosynthesis (DBT) scanners are integrated in a single gantry to provide spatially co-registered 3D PET-tomosynthesis images. The DBT image will be used to identify the breast boundary and breast density to improve the quantitative accuracy of the PET image. This paper explores PET attenuation correction (AC) strategies that can be performed with the combined breast PET-DBT scanner to obtain more accurate, quantitative high-resolution 3D PET images. The PET detector is comprised of a 32 × 32 array of 1.5 × 1.5 × 15 mm3 LYSO crystals. The PET scanner utilizes two detector heads separated by either 9 or 11 cm, with each detector head having a 4 × 2 arrangement of PET detectors. GEANT4 Application for Tomographic Emission simulations were performed using an anthropomorphic breast phantom with heterogeneous attenuation under clinical DBT-compression. FDG-avid lesions, each 5 mm in diameter with 8:1 uptake, were simulated at four locations within the breast. Simulations were performed with a scan time of 2 min. PET AC was performed using the actual breast simulation model as well as DBT reconstructed volumetric images to derive the breast outline. In addition to using the known breast density as defined by the breast model, we also modeled it as uniform patient-independent soft-tissue, and as a uniform patient-specific material derived from breast tissue composition. Measured absolute lesion uptake was used to evaluate the quantitative accuracy of performing AC using the various strategies. This study demonstrates that AC is necessary to obtain a closer estimate of the true lesion uptake and background activity in the breast. The DBT image dataset assists in measuring lesion uptake with low bias by facilitating accurate breast delineation as well as providing accurate information related to the breast tissue composition. While both the uniform soft-tissue and patient-specific material approaches provides a close estimate to the ground truth, <5% bias can be achieved by using a uniform patient-specific material to define the attenuation map.

Measuring Differential Volume Using the Subtraction Tool for Three-Dimensional Breast Volumetry: A Proof of Concept Study.

Background. Three-dimensional (3D) photography provides a promising means of breast volumetry. Sources of error using a single-captured surface to calculate breast volume include inaccurate designation of breast boundaries and prediction of the invisible chest wall generated by computer software. An alternative approach is to measure differential volume using subtraction of 2 captured surfaces. Objectives. To explore 3D breast volumetry using the subtraction of superimposed images to calculate differential volume. To assess optimal patient positioning for accurate volumetric assessment. Methods. Known volumes of breast enhancers simulated volumetric changes to the breast (n = 12). 3D photographs were taken (3dMDtorso) with the subject positioned upright at 90° and posteriorly inclined at 30°. Patient position, breathing, distance and camera calibration were standardised. Volumetric analysis was performed using 3dMDvultus software. Results. A statistically significant difference was found between actual volume and measured volumes with subjects positioned at 90° (P < .05). No statistical difference was found at 30° (P = .078), but subsequent Bland–Altman analysis showed evidence of proportional bias (P < .05). There was good correlation between measured and actual volumes in both positions (r = .77 and r = .85, respectively). Univariate analyses showed breast enhancer volumes of 195 mL and 295 mL to incur bias. The coefficient of variation was 5.76% for single observer analysis. Conclusion. Positioning the subject at a 30° posterior incline provides more accurate results from better exposure of the inferior breast. The subtraction tool is a novel method of measuring differential volume. Future studies should explore methodology for application into the clinical setting.

Improved Threshold Based and Trainable Fully Automated Segmentation for Breast Cancer Boundary and Pectoral Muscle in Mammogram Images

Segmentation of the breast region and pectoral muscle are fundamental subsequent steps in the process of Computer-Aided Diagnosis (CAD) systems. Segmenting the breast region and pectoral muscle are considered a difficult task, particularly in mammogram images because of artefacts, homogeneity among the region of the breast and pectoral muscle, and low contrast along the region of breast boundary, the similarity between the texture of the Region of Interest (ROI), and the unwanted region and irregular ROI. This study aims to propose an improved threshold-based and trainable segmentation model to derive ROI. A hybrid segmentation approach for the boundary of the breast region and pectoral muscle in mammogram images was established based on thresholding and Machine Learning (ML) techniques. For breast boundary estimation, the region of the breast was highlighted by eliminating bands of the wavelet transform. The initial breast boundary was determined through a new thresholding technique. Morphological operations and masking were employed to correct the overestimated boundary by deleting small objects. In the medical imaging field, significant progress to develop effective and accurate ML methods for the segmentation process. In the literature, the imperative role of ML methods in enabling effective and more accurate segmentation method has been highlighted. In this study, an ML technique was built based on the Histogram of Oriented Gradient (HOG) feature with neural network classifiers to determine the region of pectoral muscle and ROI. The proposed segmentation approach was tested by utilizing 322, 200, 100 mammogram images from mammographic image analysis society (mini-MIAS), INbreast, Breast Cancer Digital Repository (BCDR) databases, respectively. The experimental results were compared with manual segmentation based on different texture features. Moreover, evaluation and comparison for the boundary of the breast region and pectoral muscle segmentation have been done separately. The experimental results showed that the boundary of the breast region and the pectoral muscle segmentation approach obtained an accuracy of 98.13% and 98.41% (mini-MIAS), 100%, and 98.01% (INbreast), and 99.8% and 99.5% (BCDR), respectively. On average, the proposed study achieved 99.31% accuracy for the boundary of breast region segmentation and 98.64% accuracy for pectoral muscle segmentation. The overall ROI performance of the proposed method showed improving accuracy after improving the threshold technique for background segmentation and building an ML technique for pectoral muscle segmentation. More so, this article also included the ground-truth as an evaluation of comprehensive similarity. In the clinic, this analysis may be provided as a valuable support for breast cancer identification.

A Radiomic feature-based Nipple Detection Algorithm on Digital Mammography.

In the diagnosis and detection of breast lesions, the nipple is an important anatomical landmark which can be used for the registration on multiview mammograms. In this study, we propose a new detection algorithm for nipples on digital mammography (DM) by applying pixel classification based on geometric and radiomic features extracted from breast boundary regions. The imaging characteristics of nipples are closely related to the visibility on mammograms. To locate the nipple on mammogram, a searching area is first determined based on the breast boundary and chest wall orientation. Two different approaches are developed for obvious and subtle nipples, respectively. For obvious nipples, top hat transformation is employed to detect the nipple region, whose geometric center is regarded as the nipple position. For subtle nipples, the curved searching area near the breast boundary is mapped onto a Cartesian plane through a revised rubber band straightening transformation. On the straightened searching area, the geometric and radiomic features are calculated along the normal direction of the breast boundary, and a random forest classifier is trained for subtle nipple localization. Seven hundred and twenty-one DMs were collected for the evaluation of the proposed algorithm. The locations of nipples are manually identified by an experienced radiologist as the reference standard. The average Euclidean distance between the computed nipple position and the reference standard was 2.69mm (obvious) and 7.81mm (subtle), respectively. A total of 97.61% of the obvious nipples (613/628) and 88.17% of the subtle nipples (82/93) were detected within a 10-mm radius centered from the reference standard. The evaluation results show that the proposed method is effective for nipple detection on DM, especially for subtle nipple detection.

Three-dimensional Distorted Born Iterative Method Enhanced by Breast Boundary Extraction for Microwave Mammography.

Microwave mammography has several advantages e.g. portability, non-contact and cell-friendly measurement, and low cost for the equipment. Therefore, this technology has been regarded as a more frequent screening technique, than the X-ray based modality. This paper focuses on the dependence of the imaging accuracy, which is obtained using inverse scattering algorithms, on the estimation of the breast boundary. The Envelope method has been proposed as a method of using microwaves to estimate the location of the breast boundary. The accuracy of the boundary estimation via the Envelope method depends on that of the time-delay estimation, which is mostly processed by a filter based on waveform matching between the observed and reference signals. However, the coupling effect between the antenna and breast surface deforms the observed waveforms relative to the reference waveform. To mitigate this problem, this paper introduces the finite-difference time-domain (FDTD)-based waveform correction assuming the three-dimensional (3-D) distorted Born iterative method (DBIM) analysis. The 3-D numerical simulations based on realistic breast phantoms demonstrate that the proposed method enhances the accuracy for the 3-D reconstruction of the dielectric profile in highly heterogeneous cases.