The stable status evaluation for female breast implant surgery by calculating related physics parameters

Cosmetic doctor utilizes the position, size and shapes of female's breast to judge whether the breast is under steady-state condition after breast implant plastic surgery. Since, doctor evaluates the breast condition with the subjective discrimination (such as vision, sense of touch) without using the objective physical parameters auxiliary. This study uses the 3D optics scanner editing 3D image to obtain full-scale 3D female breasts image. The CAD system converts the breast position, size and shapes, as the length of the curve between UBL (upper breast line) and NBL (nipple base line), the length of the curve between NBL and LBL (lower breast line), breast volume and breasts congruence rate. The stability after the breast implant plastic surgery is one of the important successful indexes of plastic surgery, so with the continuity analysis the breast curve length, volume and congruence rate can let the doctor really grasp the stability of the breast after plastic surgery.     

An improved GVF snake based breast region extrapolation scheme for digital mammograms

To accurately extrapolate the breast region from a mammogram is a crucial stage of breast mass analysis. It significantly influences the overall analysis accuracy and processing speed of the whole breast mass analysis. In this paper, a novel edge map adjusting gradient vector flow snake (EMA GVF snake) algorithm for extrapolation of breast region from mammograms is proposed. In the proposed algorithm, the median filter is used to filter out the noise in a mammogram, the scale down stage is used to resize down the mammogram size (hence speeding up the extrapolation). The binarization processing stage and the morphological erosion processing stage are used to find a rough breast border. Then a novel gradient adjusting stage is applied to get a modified edge map and the gradient vector flow snake (GVF snake) is used to get the accurate breast border from the rough breast border. The proposed algorithm is tested on 322 digital mammograms from the Mammogram Image Analysis Society database. The mean error function, misclassification error function and the relative foreground area error function are conducted to evaluate the results of the detected breast border and the extracted breast region. Experimental results show that the breast border extrapolated by the proposed algorithm approximately follows the breast border extrapolated by an expert radiologist. Experimental results also show that the proposed algorithm is more robust and precise than the traditional GVF snake scheme for the breast extrapolation on mammograms.     

A study of breast motion using non-linear dynamic FE analysis

This paper presents a new method to simulate non-linear breast motion by using a three-dimensional (3D) dynamic finite element model (FEM). The model consists of a thorax with two breasts and three skin layers with specific mechanical properties. Using free breast vibration, the viscous damping ratios were ascertained to be 0.215 for an 80B size breast. The shear modulus for the breast was derived as the value that gave the minimum difference between the FEM-predicted results and the experimental data. A hyper-elastic neo-Hookean material model simulated the large deformation of breast tissue. The mode shapes of breast motions at different natural frequencies were established. The highest breast displacement amplitude ratio relative to the thorax was at 4 Hz. The study showed that FEM can predict breast displacement with sufficient accuracy and thereby provide the basis by which bras may be engineered more ergonomically in the future.

Practitioner Summary: To facilitate a theoretical analysis of breast motion to enable the design of more supportive bras, a dynamic FEM based on reliable non-linear properties of breast tissues has been developed. The methods and findings have potential widespread benefit for developing new products to promote women's health and comfort.     

Potential of near-field microwave imaging in breast cancer detection utilizing tapered rectangular waveguide probes

Microwave imaging for medical applications has been of interest for many years. A novel near-field microwave non-invasive testing and evaluation (NIT&E) technique utilizing tapered rectangular waveguide probes is presented for breast cancer detection. Near-field microwave NIT&E techniques can be a successful candidate for the detection of breast cancer because of their potential in dealing with materials of low conductivity (i.e. lossy dielectrics like the breast tissue). The physical basis for breast tumor detection with microwave imaging is the contrast in dielectric properties of normal and malignant breast tissues. A method adopting Fourier transform matching (FTM) technique and utilizing the reflection coefficient at the aperture of a tapered rectangular waveguide sensor radiating into a breast is described resulting in microwave images that indicate the presence of a tumor. These images demonstrate the feasibility of detecting breast tumors using this approach.     

基于深度学习的乳腺X射线影像分类方法研究

X射线乳腺影像与自然图像相比,色彩较为单调,且乳腺肿块边缘模糊,良性肿块与恶性肿块纹理相似,区分度较小。基于卷积深度学习网络提出一种适用于X射线乳腺肿块影像分类的方法,主要贡献如下：（1）提出一种提取乳腺影像多个卷积粒度的特征图的方案,分别使用不同尺寸的卷积核来提取不同粒度的卷积特征图,获得更为丰富的乳腺影像特征;（2）将判别方法嵌入到优化模型中,即设计新的目标函数,对分类误差进行差异化放大,从而加大分类错误的惩罚力度,指导模型向着分类错误最小的方向演进。在公开的乳腺X射线影像数据集上进行训练,通过交叉验证,AUC达到0.712?9,优于最好的乳腺影像分类方法,具有较强的鲁棒性。     

Effect of sports bra type and gait speed on breast discomfort,bra discomfort and perceived breast movement in Chinese women

This study investigated the effect of sports bra type (encapsulation versus compression) and gait speed on perceptions of breast discomfort, bra discomfort and breast movement reported by Chinese women. Visual analogue scales were used to evaluate breast discomfort, bra component discomfort and perceived breast movement of 21 Chinese participants when they wore an encapsulation or a compression sports bra, while static and while exercising at three different gait speeds. Participants perceived less breast discomfort and breast movement when wearing a compression bra compared to an encapsulation bra at a high gait speed, suggesting that compression bras are likely to provide the most effective support for Chinese women. However, significantly higher bra discomfort was perceived in the compression bra compared to the encapsulation bra when static and at the lower gait speed, implying that ways to modify the design of sports bras, particularly the straps, should be investigated to provide adequate and comfortable breast support.     

A breast modeler based on analysis of breast scans

The advent of 3D scanning technology has allowed effective measurement and analysis of breast size and shape, attracting interests by plastic surgeons, brassier designers, etc. Much work remains, however, before 3D scanning systems can be successfully used in automated analysis and synthesis of the breast—filtering noise, filling holes, and, in case a statistical analysis is desired, finding correspondence among each scan data. Moreover, analysis of a sagged breast is difficult to obtain, due to occlusions. In this paper, we address the problems and specific issues of using 3D scan data for the analysis and synthesis of breast models. The goal of our work is to build a breast modeler which can help both surgeons and garment designers in analyzing breast volume and surface measurements. Given enough samples of scanned breasts, our modeler can generate highly realistic breast shape, with some expected and consistent variability. Copyright © 2007 John Wiley & Sons, Ltd.     

应用单调下降曲线模型自动提取乳房轮廓线

乳房外轮廓包含非常重要的信息,在图像定位、对比分析中起着非常重要的作用,在一定程度上影响着乳腺图像对比分析技术的性能。通过分析乳腺X线图像的特点,判别图像为左乳图像还是右乳图像,若是右乳图像则做水平镜像处理;应用单调下降曲线模型提取乳房轮廓点;选择最长下降曲线段上的轮廓点为基准点,分别向上和向下逐行修正其他扫描线上的轮廓点;判别伪轮廓曲线段,采用圆弧模型拟合正确的轮廓线,从而提取出完整的乳房轮廓线。     

Breast volume is affected by body mass index but not age

Abstract

Background: This study aimed to establish normative breast volume data for women of varying ages, body masses and breast sizes, and to determine the effect of age and body mass index (BMI) on breast volume.

Methods: The breast volume of 356 women (age range: 18.1–83.7 years; BMI range: 18.4–54.5 kg/m²) was measured using three-dimensional scanning in a prone position.

Results: Breast volumes ranged from 48 to 3100 mL. Although breast volume was not significantly affected by age, it was significantly affected by BMI, with the breast volume of overweight and obese women being two-to-three times greater than women with normal BMI’s.

Conclusion: It is recommended that bra cups must be designed to support the wide range and increasing magnitude of breast volumes exhibited by women.

Practitioner summary: This original research provides evidence for bra designers and manufacturers on the range of breast volumes of women and the significant effect of BMI on breast volume. Bra cups need to be designed to support the wide range and increasing magnitude of breast volumes exhibited by women.     

Measurement protocol of women's nude breasts using a 3D scanning technique

It has been difficult to obtain accurate surface data for women's breasts, since the borderline of the breast is ambiguous at the skin surface. In this paper, we scanned 37 women's nude breasts using 3D phase shifting moiré and explored a measurement protocol to get a reliable boundary of the breast and thereby provided new shape parameters for breasts. The folding line method was useful for finding a continuous and natural boundary for the breasts, so that the breasts base and volume could be measured more accurately. The global average radius of the curvature of the bottom breast line was suggested as a useful shape parameter for the design of a comfortable and form-fitting brassiere, especially for those who have a skewed bottom breast curve. It is also expected that data on breast volume based on natural anatomy would be a useful reference for surgical operation on the breast.     

基于自联想神经网络的良恶性肿瘤分类技术

尽管乳腺癌的诊断和处理技术在不断进步,但乳腺病灶的早期检测仍然是阻止癌症的主要方法。乳腺组织中肿块的存在是乳腺癌的重要特征。通过使用自联想神经网络和多层感知器技术研究了良恶性肿瘤的分类方法。该研究的实验结果显示,在DDSM数据库上进行训练和测试,得到了较高的CAD系统的灵敏度（TP）和较低的假阳性率（FP）;在100%的训练分类率上获得了91%的测试分类率;ROC曲线下方面积最大可达约0.948。     

基于EST和SVM的乳腺癌识别新方法

在分析了目前肿瘤分类检测所采用的方法基础上,提出了一种基于特征空间分离变换（Eigenspace Separation Transform）结合支持向量机的乳癌识别新方法。用UCI数据库提供的569例乳腺肿瘤患者,乳腺肿块细胞核显微图像的30个量化特征样本集,进行了分类识别实验,结果表明采用新方法检测乳癌正确识别率达98.3%,优于传统的其他分类识别方法。     

Computer-aided detection and classification of microcalcifications in mammograms: a survey

Breast cancer continues to be a significant public health problem in the world. Approximately, 182,000 new cases of breast cancer are diagnosed and 46,000 women die of breast cancer each year in the United States. Even more disturbing is the fact that one out of eight women in US will develop breast cancer at some point during her lifetime. Primary prevention seems impossible since the causes of this disease still remain unknown. Early detection is the key to improving breast cancer prognosis. Mammography is one of the reliable methods for early detection of breast carcinomas. There are some limitations of human observers, and it is difficult for radiologists to provide both accurate and uniform evaluation for the enormous number of mammograms generated in widespread screening. The presence of microcalcification clusters (MCCs) is an important sign for the detection of early breast carcinoma. An early sign of 30–50% of breast cancer detected mammographically is the appearance of clusters of fine, granular microcalcification, and 60–80% of breast carcinomas reveal MCCs upon histological examinations. The high correlation between the appearance of the microcalcification clusters and the diseases show that the CAD (computer aided diagnosis) systems for automated detection/classification of MCCs will be very useful and helpful for breast cancer control. In this survey paper, we summarize and compare the methods used in various stages of the computer-aided detection systems (CAD). In particular, the enhancement and segmentation algorithms, mammographic features, classifiers and their performances are studied and compared. Remaining challenges and future research directions are also discussed.     

Bra-breast forces generated in women with large breasts while standing and during treadmill running: Implications for sports bra design

This study aimed to determine the bra-breast forces generated in women with large breasts while these women wore different levels of breast support during both upright standing and treadmill running. The mean bilateral vertical component of the bra-breast force in standing was 11.7?±?4.6?N, whereas during treadmill running the mean unilateral bra-breast force was 8.7?±?6.4?N and 14.7?±?10.3?N in the high and low support conditions, respectively. Furthermore, breast mass was significantly correlated with vertical breast displacement (R(2)?=?0.62) in the high support condition. The wide range of breast masses of women with large breasts is an important consideration for designers of sports bras to ensure these bras can reduce force generation and breast discomfort by providing a high level of breast support while these women participate in physical activity.     

基于深度学习的乳腺癌病理图像分类研究综述

准确、高效的乳腺癌病理图像分类是计算机辅助诊断的重要研究内容之一。随着机器学习技术的发展，深度学习日渐成为一种有效的乳腺癌病理图像分类处理方法。分析了乳腺癌病理图像分类方法及目前存在的问题；介绍了四种相关的深度学习模型，对基于深度学习的乳腺癌病理图像分类方法进行梳理，并通过实验对比分析现有模型的性能；最后对乳腺癌病理图像分类的关键问题进行了总结，并讨论了未来研究的发展趋势。     

Optimal brassiere wire based on the 3D anthropometric measurements of under breast curve

It has been difficult to quantify the fit of brassiere wire to the natural nude under breast line due to the difficulty of obtaining women's nude under breast line in 3D image. The objectives of this study are firstly to design the ergonomic brassiere wire that should fit to the curve of under breast and thorax and secondly to apply the 3D measurement of human body structure to the product design and evaluation using phase shifting moiré topography. Eight brassiere wires were evaluated and key factors of comfortable wires were selected depending on the types of women's under breast those are considered as backbones of the brassiere structure. Twenty-one middle-aged women participated in the subjective wear test, and 17 volunteers of them continued to participate in the 3D image recording. 3D morphological data of the breast before and after wearing experimental brassiere was examined by phase shifting moiré topography. As results, it was found that the fitting of the global average radius of curvature of wire to that of the body line of the under breast curve was an important variable for the design of wires. Conventional shape of brassiere wire was suitable to the majority of the subjects. However, for those who have skewed under breast lines, the wire of larger global average radius of curvature along the inner part of the under breast line was rated higher in terms of wear comfort and appearance. The recommended magnitude of the torsional rigidity of the brassiere wire was about 0.015 N/mm2mm. The observation on women's under breast curve and their comfortable breast wire as discussed in this paper would provide with comfortable and stable fit for women in special needs such as soldiers, athletes and actress.     

经典YOLO系列目标检测算法及其在乳腺癌检测中的应用

目前乳腺癌已取代肺癌成为年发病率最高的癌症, 基于深度学习的目标检测技术可对乳腺X线、乳腺超声和乳腺核磁共振等非侵入式成像进行自动病变检测, 已成为乳腺癌辅助诊断的首选途径. YOLO (you only look once)系列算法是基于深度学习的目标检测算法, 经典YOLO算法在速度和精准度具有优势, 被广泛应用于计算机视觉各领域, 最新YOLO算法是计算机视觉领域的SOTA (state of the art)模型, 如何利用YOLO系列算法提高乳腺癌检测速度和准确率, 已经成为研究者关注的焦点之一. 基于此, 本文介绍经典YOLO系列算法的原理, 梳理经典YOLO系列算法在乳腺癌图像检测中的应用现状, 并归纳总结现存问题, 同时对YOLO系列算法在乳腺癌检测的进一步应用进行展望.     

基于乳腺钼靶X线图像等值面特征的乳腺区域阈值分割算法

乳腺钼靶X线图像中乳腺区域的分割可以帮助对图像进行深入分析和处理,从而提高乳腺疾病的诊断准确率。提出一种能有效提取乳腺区域的算法。算法分析了乳腺钼靶X线图像等值面面积变化不连续的特征并将其用于分割阈值的精确计算。该算法使用基于扫描线的方法来获得含乳腺区域的连通区域,比种子填充法效率更高。为了获得更纯粹的乳腺区域,通过一些精细地处理对乳腺区域相连的未曝光图像边框作了剥离。实验结果表明算法在乳腺区域分割的精度和执行效率上都有更好的表现。     

基于纹理特征的超声图像乳腺肿块识别

针对乳腺超声图像,提出一种基于图像纹理特征提取的乳腺肿块识别方法,从而有助于使用计算机辅助鉴别的方法判断乳腺肿块是否发生癌变,辅助放射科医生对影像的性质作出预判.首先对乳腺超声图像进行最大响应滤波处理,在保证一定边缘组织结构完整的同时去除主要的噪声干扰.在此基础上,提取乳腺图像的一阶和二阶纹理特征,然后用人工神经网络对...     

融合空间和通道特征的高精度乳腺癌分类方法

组织病理学图像是鉴别乳腺癌的黄金标准,所以对乳腺癌组织病理学图像的自动、精确的分类具有重要的临床应用价值。为了提高乳腺组织病理图像的分类准确率,从而满足临床应用的需求,提出了一种融合空间和通道特征的高精度乳腺癌分类方法。该方法使用颜色归一化来处理病理图像并使用数据增强扩充数据集,基于卷积神经网络（CNN）模型DenseNet和压缩和激励网络（SENet）融合病理图像的空间特征信息和通道特征信息,并根据压缩-激励（SE）模块的插入位置和数量,设计了三种不同的BCSCNet模型,分别为BCSCNetⅠ、BCSCNetⅡ、BCSCNetⅢ。在乳腺癌癌组织病理图像数据集（BreaKHis）上展开实验。通过实验对比,先是验证了对图像进行颜色归一化和数据增强能提高乳腺的分类准确率,然后发现所设计的三种乳腺癌分类模型中精度最高为BCSCNetⅢ。实验结果表明,BCSCNetⅢ的二分类准确率在99.05%~99.89%,比乳腺癌组织病理学图像分类网络（BHCNet）提升了0.42个百分点;其多分类的准确率在93.06%~95.72%,比BHCNet提升了2.41个百分点。证明了BCSCNet能准确地对乳腺癌组织病理图像进行分类,同时也为计算机辅助乳腺癌诊断提供了可靠的理论支撑。