A progressive processing method for breast cancer detection via UWB based on an MRI-derived model

Ultra-wideband （UWB） microwave imaging is a promising method for breast cancer detection based on the large contrast of electric parameters between the malignant tumor and its surrounded normal breast organisms. In the case of multiple tumors being present, the conventional imaging approaches may be ineffective to detect all the tumors clearly. In this paper, a progressive processing method is proposed for detecting more than one tumor. The method is divided into three stages： primary detection, refocusing and image optimization. To test the feasibility of the approach, a numerical breast model is developed based on the realistic magnetic resonance image （MRI）. Two tumors are assumed embedded in different positions. Successful detection of a 3.6 mm-diameter tumor at a depth of 42 mm is achieved. The correct information of both tumors is shown in the reconstructed image, suggesting that the progressive processing method is promising for multi-tumor detection.     

Detection of objects underneath the water-sediment interface based on time reversal processing

The hypothesis test of detecting presence and absence of a target in a highly reverberant environment using techniques of time-reversal focusing with interaction of transmission and reception was studied. Time-reversal （TR） focusing with interaction of transmission and reception is as follows： a PS （Probe Source） is located near the object to generate signal, the SRA （Source Receiver Array） receives the signal transmitted by the PS and focuses the signal at the location of the object by physical TR focusing （i.e., inverse beamforming with multi-path compensated）. TR transmission array and reception array interactively focus the echo scattering by the object with algorithmic TR focusing （i.e., beamforming with multi-path compensated）. Meanwhile the conventional BS （broadside） transmission detection and TR detection are compared. The experiment was conducted in the Laboratory Waveguide to prove the theoretical analyses and results. It follows that the TR detection get more gain on echo-to-reverberation ratio than the conventional BS transmission detection.     

Infrared small target detection based on modified local entropy and EMD

Image entropy and empirical mode decomposition （EMD） are effective methods for target detection. EMD algorithm is a powerful tool for adaptive multiscale analysis of nonstationary signals. A new technique based on EMD and modified local entropy is proposed in small target detection under sea-sky background. With the EMD algorithm, it is valid to estimate the background and get the target image by removing the background from the original image and segmenting the target based on the modified local entropy method. The data analysis and experiments show the validity of the proposed algorithm.     

Novel algorithm for simultaneously detecting multiple vapor materials with multiple-wavelength differential absorption lidar

Differential absorption lidar (DIAL) has been successfully used to detect vapor material, however limited to detect single vapor using two closely spaced wavelengths. The progress in multiple-wavelength lasers motivates the need for detection and estimation algorithms that have the capability for simultaneous detection of multiple materials. In this paper, a simple and accurate algorithm is presented for simultaneously detecting and estimating multiple vapor materials with multiple-wavelength DIAL, which based on the maximum likelihood estimation (MLE) methodology. The performance of the algorithm is evaluated by simulation experiments, the results show that this algorithm can separately identify and quantify vapor material in mixtures and perform quite well.     

Thermal apoptosis analysis considering injection behavior optimization and mass diffusion during magnetic hyperthermia

Thermally induced apoptosis for tumors depends mainly on the intrinsic characteristics of biological tissues as well as treatment temperature profile during magnetic hyperthermia.Further,treatment temperature distribution inside tumor depends on the injection behavior of irregular tumors,such as the injection dose and the injection location of nanofluids.In order to improve the treatment effect,the simulated annealing algorithm is adopted in this work to optimize the nanofluid injection behavior,and the improved Arrhenius model is used to evaluate the malignant ablations for three typical malignant tumor cell models.In addition,both the injection behavior optimization and the mass diffusion of nanofluid are both taken into consideration in order to improve the treatment effect.The simulation results demonstrate that the injection behavior can be optimized effectively by the proposed optimization method before therapy,the result of which can also conduce to improving the thermal apoptosis possibility for proposed typical malignant cells.Furthermore,an effective approach is also employed by considering longer diffusion duration and correct power dissipation at the same time.The results show that a better result can then be obtained than those in other cases when the power dissipation of MNPs is set to be QMNP=5.4×10⁷W·m³ and the diffusion time is 16 h.     

Track before detect for point targets with particle filter in infrared image sequences

The problem of detecting and tracking point targets in a sequence of infrared images with very low signalto-noise ratio (SNR) is investigated in this paper. A track before detect algorithm for infrared (IR) point target is developed based on particle filter. The particle filter is used to estimate the state of the target in track stage. The unnormalized weights of the output of the filter are used to approximately construct the likelihood ratio for hypothesis test in detection stage. Experiment results with the real image sequences that SNR is about 2.0 show that the proposed algorithm can successfully detect and track point target.     

Track before detect for point targets with particle filter in infrared image sequences

The problem of detecting and tracking point targets in a sequence of infrared images with very low signalto-noise ratio (SNR) is investigated in this paper. A track before detect algorithm for infrared (IR) point target is developed based on particle filter. The particle filter is used to estimate the state of the target in track stage. The unnormalized weights of the output of the filter are used to approximately construct the likelihood ratio for hypothesis test in detection stage. Experiment results with the real image sequences that SNR is about 2.0 show that the proposed algorithm can successfully detect and track point target.     

Detecting overlapping communities in networks via dominant label propagation

Community detection is an important methodology for understanding the intrinsic structure and function of a realworld network.In this paper,we propose an effective and efficient algorithm,called Dominant Label Propagation Algorithm(Abbreviated as DLPA),to detect communities in complex networks.The algorithm simulates a special voting process to detect overlapping and non-overlapping community structure in complex networks simultaneously.Our algorithm is very efficient,since its computational complexity is almost linear to the number of edges in the network.Experimental results on both real-world and synthetic networks show that our algorithm also possesses high accuracies on detecting community structure in networks.     

Robust minimum-variance time-reversal beamforming for target detection and localization in uncertain environment

The target detection and localization in uncertain environment with robust time reversal （TR） technique was investigated. TR is a physical process as well as a method of signal processing. Therefore, it is natural to perform beamforming via TR, i.e., TR beamforming （TRBF）. To reduce the effects of environmental uncertainty on TR spatio-temporal focusing, transmitting TRBF with modeled instead of physical probe source was studied and robust minimum-variance receiving TRBF with diagonal loading was put forward. Both of them were applied to the detection and distance estimation of the target, which was dealt with in a waveguide experiment. The experimental results show the validity of the methods in uncertain environment.     

Detecting Overlapping Communities Based on Community Cores in Complex Networks

The identification of communities is significant for the understanding of network structures and functions. Since some nodes naturally belong to several communities, the study of overlapping community structures has attracted increasing attention recently, and many algorithms have been designed to detect overlapping communities. We propose a new algorithm. The main idea is first to find the core of a community by detecting maximal cliques and then merging some tight community cores to form the community. Experimental results on two real networks demonstrate that the present algorithm is more accurate for detecting overlapping community structures, compared with some well-known results and methods.     

早期乳腺肿瘤的超宽带微波稳健波束形成成像检测系统

提出了一种基于自主设计小型超宽带天线的微波稳健波束形成(RCB)成像肿瘤检测系统.仿真结果表明,该检测系统对肿瘤反射信号有很高的敏感度.在简单平面模型和核磁共振成像图(MRI)导出模型中进行仿真检测实验,并将天线阵列接收信号用RCB算法进行成像处理.从乳房重构图像中能够得到正确的肿瘤位置及大小信息,实现了平面模型中最小直径3 mm和MRI导出模型中最小直径4 mm的肿瘤检测,证实了该检测系统用于早期乳腺肿瘤检测的可行性.     

Contrast-enhanced microwave imaging of breast tumors: a computational study using 3-D realistic numerical phantoms

The detection of early-stage tumors in the breast by microwave imaging is challenged by both the moderate endogenous dielectric contrast between healthy and malignant glandular tissues and the spatial resolution available from illumination at microwave frequencies. The high endogenous dielectric contrast between adipose and fibroglandular tissue structures increases the difficulty of tumor detection due to the high dynamic range of the contrast function to be imaged and the low level of signal scattered from a tumor relative to the clutter scattered by normal tissue structures. Microwave inverse scattering techniques, used to estimate the complete spatial profile of the dielectric properties within the breast, have the potential to reconstruct both normal and cancerous tissue structures. However, the ill-posedness of the associated inverse problem often limits the frequency of microwave illumination to the UHF band within which early-stage cancers have sub-wavelength dimensions. In this computational study, we examine the reconstruction of small, compact tumors in three-dimensional numerical breast phantoms by a multiple-frequency inverse scattering solution. Computer models are also employed to investigate the use of exogenous contrast agents for enhancing tumor detection. Simulated array measurements are acquired before and after the introduction of the assumed contrast effects for two specific agents currently under consideration for breast imaging: microbubbles and carbon nanotubes. Differential images of the applied contrast demonstrate the potential of the approach for detecting the preferential uptake of contrast agents by malignant tissues.     

超宽带微波检测早期乳腺肿瘤三维仿真

超宽带微波成像技术可作为有效的方法用于乳房早期乳腺肿瘤的检测. 该方法基于乳房组织和肿瘤之间较大的电学特性差异的特点进行成像, 能提供足够的分辨率以及足够的穿透深度. 本文采用时域有限差分方法建立超宽带微波信号在三维乳房组织中传播的模型, 并采用单极德拜模型完成了生物组织色散特性的模拟. 利用8发9收的天线阵列实现微波肿瘤探测, 利用共焦成像算法对乳房组织进行图像重构并进行肿瘤定位. 仿真结果显示共焦成像算法能够实现最小直径3 mm的肿瘤的检测, 同时证实了超宽带微波成像技术应用于早期乳腺肿瘤检测的有效性. 关键词： 微波成像 肿瘤检测 时域有限差分 共焦成像算法     

基于改进的Mask R-CNN的乳腺肿瘤目标检测研究

乳腺癌是全球女性死亡率最高的恶性肿瘤之一,早期发现有助于提升患者的存活率。本文利用深度学习中的目标检测网络对乳腺X线图像中的肿瘤病变区域进行定位和分类;然后选取Mask R-CNN网络作为目标检测模型,对Mask R-CNN的基准网络D-ShuffleNet进行改进,提出了一种新的网络——Mask R-CNN-II网络,并在Mask R-CNN-II网络中应用迁移学习算法。通过实验验证了Mask R-CNN-II网络比Mask R-CNN网络的检测精度更高,而且验证了所提基准网络、所使用的融合图像的思想以及迁移学习算法是有效的。Mask R-CNN-II有利于提高乳腺肿瘤的定位与分类,可为放射科医生提供辅助诊断意见,具有一定的临床应用价值。     

Computerized breast lesions detection using kinetic and morphologic analysis for dynamic contrast-enhanced MRI

To facilitate rapid and accurate assessment, this study proposed a novel fully automatic method to detect and identify focal tumor breast lesions using both kinetic and morphologic features from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). After motion registration of all phases of the DCE-MRI study, three automatically generated lines were used to segment the whole breast region of each slice. The kinetic features extracted from the pixel-based time-signal intensity curve (TIC) by a two-stage detection algorithm was first used, and then three-dimensional (3-D) morphologic characteristics of the detected regions were applied to differentiate between tumor and non-tumor regions. In this study, 95 biopsy-confirmed lesions (28 benign and 67 malignant lesions) in 54 women were used to evaluate the detection efficacy of the proposed system. The detection performance was analyzed using the free-response operating characteristics (FROC) curve and detection rate. The proposed computer-aided detection (CADe) system had a detection rate of 92.63% (88/95) of all tumor lesions, with 6.15 false positives per case. Based on the results, kinetic features extracted by TIC can be used to detect tumor lesions and 3-D morphology can effectively reduce the false positives.     

Comparison of optical detection system ,PVDF detection system, and PVDF needle hydrophone for optoacoustic tomography

Optoacoustic tomography (PAT) is a two-dimensional medical imaging method that has the advantage of optical contrast and resolution of ultrasonic waves. The detection systems with a high sensitivity can be used for detecting small tumors, located deeply in human tissues, such as the breast. In this study, the sensitivity of existing ultrasonic detection systems has been compared experimentally with that by using thermoelastic waves as a broadband ultrasonic source. For the comparison, an optical stress transducer(OST), a polyvinylidene difluoride (PVDF) sheet and a calibrated PVDF needle hydrophone were used.To ensure all of the detection systems interrogated by the same ultrasonic field, a small optical instrument that fixed the generating laser head was constructed. The sensitivity was evaluated by measuring signalto-noise ratios (SNRs) and noise equivalent pressures (NEPs). The PVDF system, with a 4-kPa NEP has a 22 dB better performance than the OST. The OST showed nearly the same sensitivity as the hydrophone for detecting ultrasound waves at a 1-cm distance in water. PVDF detection system provides a useful tool for imaging of soft tissues because of its high sensitivity and broad detection range.     

Nonlinear analysis using Lyapunov exponents in breast thermograms to identify abnormal lesions

Breast diseases are one of the major issues in women’s health today. Early detection of breast cancer plays a significant role in reducing the mortality rate. Breast thermography is a potential early detection method which is non-invasive, non-radiating, passive, fast, painless, low cost, risk free with no contact with the body. By identifying and removing malignant tumors in early stages before they metastasize and spread to neighboring regions, cancer threats can be minimized. Cancer is often characterized as a chaotic, poorly regulated growth. Cancerous cells, tumors, and vasculature defy have irregular shapes which have potential to be described by a nonlinear dynamical system. Chaotic time series can provide the tools necessary to generate the procedures to evaluate the nonlinear system. Computing Lyapunov exponents is thus a powerful means of quantifying the degree of the chaos.In this paper, we present a novel approach using nonlinear chaotic dynamical system theory for estimating Lyapunov exponents in establishing possible difference between malignant and benign patterns. In order to develop the algorithm, the first hottest regions of breast thermal images are identified first, and then one dimensional scalar time series is obtained in terms of the distance between each subsequent boundary contour points and the center of the mass of the first hottest region. In the next step, the embedding dimension is estimated, and by time delay embedding method, the phase space is reconstructed. In the last step, the Lyapunov exponents are computed to analyze normality or abnormality of the lesions. Positive Lyapunov exponents indicates abnormality while negative Lyapunov exponents represent normality. The normalized errors show the algorithm is satisfactorily, and provide a measure of chaos. It is shown that nonlinear analysis of breast thermograms using Lyapunov exponents may potentially capable of improving reliability of thermography in breast tumor detection as well as the possibility of differentiating between different classes of breast lesions.     

Relationship between lesion size and signal enhancement on subtraction fat-suppressed MR imaging of the breast

OBJECTIVE: To investigate the relationship between size and whole lesion enhancement of breast neoplasms. MATERIALS AND METHODS: Fat-suppressed subtraction MRI was performed in 94 breast lesions (44 malignant, 50 benign) with pathologically confirmed diagnoses. Of these, all malignant lesions and 31 of the 50 benign lesions showed enhancement. The degree of enhancement was quantified by using an ROI tracing around the whole lesion and calculated as the percentage increase in signal intensity between the corresponding precontrast and postcontrast images. RESULTS: The 44 malignant lesions showed enhancement percentage of 38.3% to 186.4% (mean 109.9%), and the 31 benign lesions showed enhancement percentage of 12.8% to 180.2% (mean 79.5%). The difference is statistically significant (P = .002). In 54 small lesions (28 malignant, 26 benign) with enhancing pixel areas of <300 mm(2) corresponding to a diameter of approximately 19.5 mm, an enhancement exceeding 75% of baseline separated malignant lesions (mean enhancement 116.7%) from benign ones (mean enhancement 72.8%) (P = .0001). This gave a sensitivity of 100% and a specificity of 69%, a positive predictive value of 78%, negative predictive value of 100% and an accuracy of 85% in using >75% enhancement increase in detecting malignancy in small (<300 mm(2)) enhancing lesions. CONCLUSION: The high sensitivity in the detection of small malignant lesions suggests a potential for the method to be used in assessment of small enhancing breast lesions.     

基于多模态MRI与深度学习的乳腺病变良恶性鉴别

为提高基于动态增强磁共振成像（DCE-MRI）的计算机辅助（CAD）方法对乳腺病变良恶性鉴别的精度,本文基于多模态特征融合,提出一种联合非对称卷积和超轻子空间注意模块的卷积神经网络AC_Ulsam_CNN.首先,采用迁移学习方法预训练模型,筛选出对乳腺病变良恶性鉴别最为有效的DCE-MRI扫描时序.而后,基于最优扫描时序图像,搭建基于AC_Ulsam_CNN网络的模型,以增强分类模型的特征表达能力和鲁棒性.最后,将影像特征与乳腺影像数据报告和数据系统（BI-RADS）分级、表观扩散系数（ADC）和时间-信号强度曲线（TIC）类型等多模态信息进行特征融合,以进一步提高模型对病灶的预测性能.采用五折交叉验证方法进行模型验证,本文方法获得了0.826的准确率（ACC）和0.877的受试者工作曲线下面积（AUC）.这表明该算法在小样本量数据下可较好区分乳腺病变的良恶性,而基于多模态数据的融合模型也进一步丰富了特征信息,从而提高病灶的检出精度,为乳腺病灶良恶性的自动鉴别诊断提供了新方法.