乳腺肿瘤超声图像的特征分析

基于乳腺肿瘤良恶性在超声图像的不同特征,利用计算机自动识别,作为医生的辅助诊断.方法的步骤为本文先在常用超声仪上获得乳腺肿瘤超声图像,接着从图像中自动提取肿瘤边缘,然后自动提取不依赖于超声仪系统的特征参数,用特征选择器选择出最优特征矢量,最后经分类器判别乳腺肿瘤的良恶性.实验基于200例病例随机划分为训练集和测试集各半进行测试,获得结果Accuracy为0.960,Sensitivity为0.982,Specificity为0.935,PPV和NPV分别为0.946和0.977,结果表明本文方法泛化能力强,可以作为识别乳腺肿瘤良恶性的一种辅助手段.     

一种超声乳腺图像中提取肿瘤边缘的方法

乳腺肿瘤边缘的提取是基于肿瘤超声图像特征进行良恶性诊断的基础.为此提出一种基于小波变换和动态规划从超声乳腺图像中提取肿瘤边缘的方法.先采用小波变换法提取乳腺肿瘤超声图像的初始边缘,再结合图像梯度的动态规划法对初始边缘进行修正,提取更为准确的乳腺肿瘤边缘.通过对临床采集的30例超声乳腺图像肿瘤边缘的提取和分析研究,表明该方法可以有效地用于超声乳腺图像肿瘤边缘的提取.     

基于径向基神经网络的直线超声电机位置控制

针对直线超声电机的精密位置控制,提出了一种基于径向基神经网络的自适应控制机制。鉴于直线超声电机工作原理,其运行状态必然受到摩擦、强非线性和时变等不确定性因素的干扰,为了对这些不确定性因素进行有效的逼近,采用了径向基神经网络。为了提高控制机制的自适应能力,首先利用来自试验数据的训练样本按正交最小二乘算法确定径向基神经网络的隐层单元的个数和相关参数,再按递推最小二乘法在线调整隐层与输出层之间的权重。试验结果表明,基于径向基神经网络的自适应控制器的性能不仅优于传统的PID控制和误差反向传播神经网络控制,而且具有很好的抗干扰能力。     

基于扩展的径向基函数核支持向量机的产品销售预测模型

针对产品销售时序具有多维度、非线性的特征,通过设计一种扩展的径向基函数核函数,将其应用于支持向量机中,得到一种扩展的径向基函数核支持向量机;设计了一种改进的免疫优化算法对其参数进行寻优.将该方法应用于汽车销售预测实例中,并与反向传播神经网络、采用一般径向基函数核的支持向量机及多尺度支持向量机进行了比较.实验结果表明该方法可行有效,其预测精度优于其他三种方法.     

基于径向基函数神经网络的超分辨率图像重建

为了突破成像极限,经济可行地获取高质量的卫星图像,提出了一种基于径向基神经网络的超分辨率图像重建算法。以径向基神经网络为基础,依据卫星图像退化模型获取网络训练所需的学习样本图像,采用向量映射的方式加速网络收敛。其中,径向基函数的中心、宽度及网络的隐含层数、连接权值是决定径向基神经网络的关键参数,直接关系到网络的重建性能。采用最近邻聚类算法,动态地建立起基函数的中心及宽度,自适应地确定网络的隐含层数及连接权值。建立起的径向基函数神经网络显著地提高了图像重建性能和网络收敛速度(221s即可收敛)。仿真实验和泛化实验表明,训练好的径向基神经网络可以有效地进行卫星图像的超分辨率重建,效率高,误差小。     

基于萤火虫优化的核自动编码器在中介轴承故障诊断中的应用

随着科技进步与工业规模的快速壮大,现代工业监测领域步入大数据时代,如何自动地从大规模原始信号中提取故障特征并诊断是一个重要课题。为了提高深度自动编码网络处理非线性问题的能力,提出一种基于核函数与去噪自动编码器(Denosingauto-encoder,DAE)的深度神经网络方法。采用径向基核函数改进传统的去噪自动编码器,提出核去噪自动编码器(Kernel denosing auto-encoder, KDAE);构建包含一个KDAE层与多个AE层的深度神经网络对故障特征进行层层提取,并添加softmax分类层;采用误差反向传播算法对网络参数进行微调,并采用混沌萤火虫算法优化核参数与深度网络中的待定参数,得到故障诊断模型。针对传统自动编码器泛化性较差的问题,在目标函数中添加L2惩罚项。通过航空发动机中介轴承典型故障试验数据,验证了所提方法与传统去噪自动编码网络相比能够得到更高的准确率。     

基于MATLAB GUI的零件图像识别

零件图像识别有多种方法,其关键是零件图像的特征提取,为此提出了基于图像边缘检测提取零件图像特征和用径向基神经网络实现识别的方法。首先对零件图像进行边缘检测．提取零件图像的边缘轮廓;然后将被检测的边缘轮廓图像分成若干个子区域并分别统计各子区域的边缘像素量,各子区域中的相对边缘像素系数作为零件的特征,将这些特征作为神经网络的输入样本,由径向基神经网络实现识别;最后由GUI完成零件图像的识别,实验结果证明是有效的。     

基于神经网络信息融合的发动机磨损磨粒识别

磨粒识别和分类是铁谱分析技术在发动机故障诊断和状态监测的关键环节.针对单一神经网络模型磨粒识别的局限性,提出了一种基于不同类型神经网络信息融合的磨粒智能识别方法.首先利用径向基函数(RBF)神经网络和反向传播(BP)神经网络对磨粒进行识别,得到2组初始识别结果, 归一化后作为2组基本概率分配函数,然后利用D-S证据理论对其融合得到最终识别结果.实例计算表明,与单一神经网络模型相比,提出的信息融合方法提高了磨粒识别的区分度和准确率,并具有良好的通用性和容错性.     

高阶统计量与RBF网络结合用于齿轮故障分类

提出一种基于高阶统计量特征提取的径向基函数网络齿轮故障分类方法,以齿轮箱振动信号的高阶统计量估计值作为齿轮故障特征,以径向基函数神经网络作为分类器,成功地对齿轮故障进行了分类。研究表明,高阶统计量和径向基函数神经网络相结合的齿轮故障分类方法是有效的。     

基于电极位移信号特征分析的电阻点焊质量监测

提出一种基于点焊过程信息采集和处理的焊点质量在线监测方法。通过对电极位移、动态电阻信号的实时采集和分析,利用电阻信号动态特征刻画熔核形成的不同阶段,从同步位移信号中提取12个与焊点质量相关的动态特征参量。通过对提取的特征参量与作为焊点质量评价指标的抗剪强度之间的相关性分析,选取相关性显著的特征参量作为输入和抗剪强度为输出,建立线性回归、非线性回归及径向基函数神经网络焊点质量监测模型。监测模型的有效性检验结果表明,建立的三种监测模型都可实现对焊点质量的在线监测。径向基函数神经网络模型的监测准确率高于其他两种模型,其平均验证误差为2.28%,最大验证误差低于10%。     

Optimizing the transfer-learning with pretrained deep convolutional neural networks for first stage breast tumor diagnosis using breast ultrasound visual images

Female accounts for approximately 50% of the total population worldwide and many of them had breast cancer. Computer-aided diagnosis frameworks could reduce the number of needless biopsies and the workload of radiologists. This research aims to detect benign and malignant tumors automatically using breast ultrasound (BUS) images. Accordingly, two pretrained deep convolutional neural network (CNN) models were employed for transfer learning using BUS images like AlexNet and DenseNet201. A total of 697 BUS images containing benign and malignant tumors are preprocessed and performed classification tasks using the transfer learning-based CNN models. The classification accuracy of the benign and malignant tasks is completed and achieved 92.8% accuracy using the DensNet201 model. The results thus achieved compared in state of the art using benchmark data set and concluded proposed model outperforms in accuracy from first stage breast tumor diagnosis. Finally, the proposed model could help radiologists diagnose benign and malignant tumors swiftly by screening suspected patients.     

A novel breast tumor classification algorithm using neutrosophic score features

A lot of studies confirmed the seriousness of breast cancer as the most tumors lethal to women worldwide. Early detection and diagnosis of breast cancer are of great importance to increase treatment options and patients’ survival rate. Ultrasound is one of the most frequently used methods to detect and diagnosis breast tumor due to its harmlessness and inexpensiveness. However, problems were found in the tumor diagnosis and classification as benign and malign on ultrasound image for its vagueness, such as speckle noise and low contrast. In this paper, we propose a novel breast tumor classification algorithm that combines texture and morphologic features based on neutrosophic similarity score. Then, a supervised feature selection technique is employed to reduce feature space. Finally, a support vector machine (SVM) classifier is employed to prove the discrimination power of the proposed features set. The proposed system is validated by 112 cases (58 malign, 54 benign). The experimental results show that such features set is promising and 99.1% classification accuracy is achieved.     

Cloud‐based decision support system for the detection and classification of malignant cells in breast cancer using breast cytology images

The advancement of computer‐ and internet‐based technologies has transformed the nature of services in healthcare by using mobile devices in conjunction with cloud computing. The classical phenomenon of patient–doctor diagnostics is extended to a more robust advanced concept of E‐health, where remote online/offline treatment and diagnostics can be performed. In this article, we propose a framework which incorporates a cloud‐based decision support system for the detection and classification of malignant cells in breast cancer, while using breast cytology images. In the proposed approach, shape‐based features are used for the detection of tumor cells. Furthermore, these features are used for the classification of cells into malignant and benign categories using Naive Bayesian and Artificial Neural Network. Moreover, an important phase addressed in the proposed framework is the grading of the affected cells, which could help in grade level necessary medical procedures for patients during the diagnostic process. For demonstrating the e effectiveness of the proposed approach, experiments are performed on real data sets comprising of patients data, which has been collected from the pathology department of Lady Reading Hospital of Pakistan. Moreover, a cross‐validation technique has been performed for the evaluation of the classification accuracy, which shows performance accuracy of 98% as compared to physical methods used by a pathologist for the detection and classification of the malignant cell. Experimental results show that the proposed approach has significantly improved the detection and classification of the malignant cells in breast cytology images.     

Microscopic brain tumor detection and classification using 3D CNN and feature selection architecture

Brain tumor is one of the most dreadful natures of cancer and caused a huge number of deaths among kids and adults from the past few years. According to WHO standard, the 700,000 humans are being with a brain tumor and around 86,000 are diagnosed since 2019. While the total number of deaths due to brain tumors is 16,830 since 2019 and the average survival rate is 35%. Therefore, automated techniques are needed to grade brain tumors precisely from MRI scans. In this work, a new deep learning‐based method is proposed for microscopic brain tumor detection and tumor type classification. A 3D convolutional neural network (CNN) architecture is designed at the first step to extract brain tumor and extracted tumors are passed to a pretrained CNN model for feature extraction. The extracted features are transferred to the correlation‐based selection method and as the output, the best features are selected. These selected features are validated through feed‐forward neural network for final classification. Three BraTS datasets 2015, 2017, and 2018 are utilized for experiments, validation, and accomplished an accuracy of 98.32, 96.97, and 92.67%, respectively. A comparison with existing techniques shows the proposed design yields comparable accuracy.     

基于双流卷积神经网络的肌电信号手势识别方法

面向高性能的肌电控制系统,提出一种基于双流卷积神经网络的肌电信号手势识别方法,其从原始表面肌电信号中提取离散小波变换系数,与原始表面肌电信号分别作为双流卷积神经网络两个分支的输入进行高层特征学习,最终通过一个高层特征融合模块对两个分支学习得到的高层特征进行融合.所提方法在3个包含50～52类手势动作表面肌电信号的大规模...     

基于深度学习的导波特征提取及其激光超声检测

针对激光超声检测中波场的三维数据处理计算量大且损伤特征提取难的问题,提出了一种基于深度学习模型的导波波场分析方法.首先,以VGG-Net网络为框架,建立了基于VGG11(A-LRN)的残差网络模型,用于挖掘时间-空间波场数据中的导波特征;其次,以局部波数特征为物理机理,采用导波传播的解析式生成训练样本,解决了深度学习大数据获取的问题,获得了波场特征提取的神经网络模型;最后,以激光超声系统在含损伤结构中的实验数据作为测试样本,验证了所提出的网络模型能够提取表征损伤的导波特征,实现了结构的损伤成像,其损伤成像精度均在67%以上,损伤形貌的可视化效果好。     

A new thermographic NDT for condition monitoring of electrical components using ANN with confidence level analysis

Infrared thermography technology is one of the most effective non-destructive testing techniques for predictive faults diagnosis of electrical components. Faults in electrical system show overheating of components which is a common indicator of poor connection, overloading, load imbalance or any defect. Thermographic inspection is employed for finding such heat related problems before eventual failure of the system. However, an automatic diagnostic system based on artificial neural network reduces operating time, human efforts and also increases the reliability of system. In the present study, statistical features and artificial neural network (ANN) with confidence level analysis are utilized for inspection of electrical components and their thermal conditions are classified into two classes namely normal and overheated. All the features extracted from images do not produce good performance. Features having low performance reduce the diagnostic performance. The study reveals the performance of each feature individually for selecting the suitable feature set. In order to find the individual feature performance, each feature of thermal image was used as input for neural network and the classification of condition types were used as output target. The multilayered perceptron network using Levenberg–Marquardt training algorithm was used as classifier. The performances were determined in terms of percentage of accuracy, specificity, sensitivity, false positive and false negative. After selecting the suitable features, the study introduces the intelligent diagnosis system using suitable features as inputs of neural network. Finally, confidence percentage and confidence level were used to find out the strength of the network outputs for condition monitoring. The experimental result shows that multilayered perceptron network produced 79.4% of testing accuracy with 43.60%, 12.60%, 21.40, 9.20% and 13.40% highest, high, moderate, low and lowest confidence level respectively.     

基于多分量卷积神经网络的多晶硅晶片颜色差异检测

具有复杂纹理的多晶硅晶片颜色差异检测是太阳能电池片制造过程中的一个挑战.针对传统的色差检测算法不适用于颜色差异类别变化大的场合,且分类结果不精确的问题,基于不同分量的颜色特征提出了一种多分量卷积神经网络的检测算法.通过分析多晶硅晶片图像在H SV颜色空间的特征分布,发现颜色特征在H、S和V分量中表现不同;基于全卷积神经...     

基于邻域属性重要度与主成分分析的齿轮箱故障特征约简

为有效降低齿轮箱故障特征的维数并提高诊断效率,提出了基于邻域属性重要度与主成分分析法相结合的齿轮箱故障特征约简方法,并利用支持向量机和BP神经网络对诊断的准确率进行对比分析。针对齿轮箱中具有不同程度裂纹的齿轮,选取其时域、频域和基于希尔伯特变换的36个特征;将邻域模型引入到特征属性的约简,构造前向贪心算法,以邻域属性重要度较大的9个特征作为特征集,提取累积贡献率达到95%以上的主成分,分别输入支持向量机和BP神经网络分类器中进行分类识别,并与不经过特征优选的主成分特征融合相对比。结果表明,采用基于邻域属性重要度与主成分分析法相结合的特征约简方法,既可以降低齿轮箱故障特征的维数,又不影响对其运行状态的表征,有助于识别不同裂纹水平的齿轮,与不经过特征优选直接进行融合的方法相比,所提出方法诊断准确率更高,训练时间更短。     

Imprint cytology‐based breast malignancy screening: an efficient nuclei segmentation technique

Imprint cytology (IC) refers to one of the most reliable, rapid and affordable techniques for breast malignancy screening; where shape variation of H&E stained nucleus is examined by the pathologists. This work aims at developing an automated and efficient segmentation algorithm by integrating Lagrange's interpolation and superpixels in order to delineate overlapped nuclei of breast cells (normal and malignant). Subsequently, a computer assisted IC tool has been designed for breast cancer (BC) screening. The proposed methodology consists of mainly three subsections: gamma correction for preprocessing, single nuclei segmentation and segmentation of overlapping nuclei. Single nuclei segmentation combines histogram‐based thresholding and morphological operations; where segmentation of overlapping nuclei includes concave point detection, Lagrange's interpolation for overlapping arc area detection and the fine segmentation of overlapped arc area by superpixels. Total 16 significant features (p < 0.05) quantifying shape and texture of nucleus were extracted, and random forest (RF) classifier was skilled for automated screening. The proposed methodology has been tested on 120 IC images (approximately 12 000 nuclei); where 98% segmentation accuracy and 99% classification accuracy were achieved. Besides, performance evaluation was studied by using Jaccard's index (= 94%), correlation coefficient (= 95%), Dice similarity coefficient (= 97%) and Hausdorff distance (= 43%). The proposed approach could offer benefit to the pathologists for confirmatory BC screening with improved accuracy and could potentially lead to a better shape understanding of malignant nuclei.