超声射频信号的甲状腺结节智能诊断方法

由于人口老龄化的原因,甲状腺癌的发病率增长率在所有癌症中是最为显著的.因此,对存在癌变可能的甲状腺结节进行预检查显得尤为重要,而超声智能诊断系统在甲状腺结节早期筛查方面已展现出巨大的应用前景.该文的工作旨在提出一种基于超声原始射频信号的组织参数定征和人工神经网络相结合的甲状腺结节智能诊断方法,以提高临床超声诊察效率及准...     

非线性超声射频信号熵对乳腺结节良恶性的定征

本文提出了一种基于非线性超声射频(radio frequency, RF)信号熵对乳腺结节良恶性进行定征的方法.对306例乳腺结节样本(良性158例,恶性148例)提取了基于超声RF信号二次谐波的熵和加权熵,以及常规超声参数(图像灰度、纵横比、不规则度、乳腺结节大小、深度);采用t检验和线性分类器检测参数对乳腺结节良恶性的区分度;进一步将有效参数组合输入支持向量机对乳腺结节良恶性进行分类.结果表明:除图像灰度外,其余参数均在乳腺结节的良性与恶性间有显著差异.多参数结合输入支持向量机的良恶性分类的准确率、敏感性和特异性分别为81.4%, 78.4%和84.2%.本文工作表明非线性超声RF信号的熵可有效地定征乳腺结节的良恶性,有望成为乳腺结节良恶性定征新参量.     

微型快照式窄带多光谱成像宫颈癌筛查方法

白光阴道镜图像对比度较低,不利于医生鉴别不同病变程度组织,也不利于自动化宫颈癌筛查。利用癌变组织富含血红蛋白成分及血红蛋白具有特征波段这一特性,与传统高光谱空间扫描成像及分时获取不同波段多光谱成像方法相反,利用快照式多光谱窄带成像来加速光谱图像获取过程,提升不同病变程度组织之间灰度对比度同时,降低后续图像分析处理算法难度,实现对宫颈组织病变类型高帧率自动化分类。首先,使用微型快照式窄带多光谱摄像方法,在血红蛋白的两个强吸收峰(415±10)和(525±10) nm、一个反射带(620±10) nm和一个背景波段(450±10) nm共四个波段对宫颈组织进行快照式零时差获取四幅窄带光谱图像。而后,对所获取的光谱图像进行简单代数加减,以生成突显病变组织的融合图像,提高不同病变程度组织之间的对比度。最后,使用欧式距离分类算法,对光谱融合图像中不同病变级别进行分类,建立计算机辅助宫颈癌筛查方法。创新点在于实现了高帧率计算机辅助光学病理诊断方法。分别采用临床常规白光阴道镜及微型快照式窄带多光谱摄像对宫颈癌手术切下的新鲜组织进行彩色图像及光谱融合图像的高帧率采集,并使用同一个欧式距离分类算法对两种图像进行自动分类,分类结果都以组织病理诊断作为标准来计算正确率。通过对比两种分类结果正确率来检验光谱融合图像相对于彩色图像是否提升对比度,及其是否可以实现与组织病理诊断(金标准)结果一致的诊断。欧式距离分类算法对光谱融合图像分类准确率接近100%,远高于对白光阴道镜图像约50%的准确率。多位临床医生对基于微型快照式多光谱摄像头光谱融合图像的计算机自动分类结果表示接受。微型快照式窄带多光谱成像方法能有效提升光谱融合图像获取帧率及不同病变程度组织之间灰度对比度,能有效快速地将宫颈组织划分为与组织病理诊断结果一致的病变类型。由于诊断客观、无创伤、结果立等可得,该方法将有助于实现落后地区宫颈癌筛查的普及以及图像导航下的宫颈癌精准治疗手术。     

基于主成分分析和BP神经网络的柑橘黄龙病诊断技术

柑橘黄龙病的传统诊断方法主要依赖于人眼经验及生化技术,前者凭经验,诊断快,但准确性低;后者准确性高,但效率低和成本高.本文采用高光谱成像技术,获取5种症状柑橘叶片的高光谱图像,采用基于主成分分析和BP神经网络相结合的方法,对370~988nm波段范围内的柑橘叶片高光谱图像进行了病状的无损检测.研究结果表明,柑橘叶片的高光谱图像存在很大冗余,前四个主成分累积方差贡献率达到97.42%.数据建模分类得表明:BP神经网络的分类准确率达85%以上,经主成分后再利用BP神经网络的分类准确率绝大部分达到90%以上.因此,利用高光谱成像技术进行柑橘黄龙病的早期诊断具有较高的可行性.     

利用小波变换和神经网络对罕见病DMD的MRI进行分类识别

杜兴氏肌营养不良(DMD)是一种严重的儿童腿部神经肌肉罕见病。传统的诊断和检测方案一般为有创手段,会带给患儿极大的痛苦。基于受试者的磁共振图像(MRI),采用计算机辅助检测手段探索了有效的无创检测方法。实验分别选用sym4和db4两种小波基函数,对患儿组和健康对照组的MRI进行三种尺度的小波分解,从所得的分解图像中提取12个纹理特征参数,并利用人工神经网络(ANN)算法对图像参数进行分类识别。结果显示:在受试者的两类MRI加权图像(T1和T2)中,T1图像能更好地区分患儿与健康儿童;利用db4函数对图像进行小波分解,其效果略优于sym4函数,且在三种小波分解尺度中,以二层分解最优;利用ANN算法对图像进行分类识别,其灵敏度、特异度和准确率分别高达98.5%、97.3%和97.9%。该处理方法有望为临床提供客观有效的辅助诊断手段,可作为DMD疾病无创检测的尝试探索。     

基于改进迭代阈值的白细胞图像分割方法

人体外周血五类白细胞数量和所占比例反映了人体的健康状态,人工检查白细胞消耗医务人员的大量精力,如何用智能方法快速、准确进行白细胞智能分类是一个亟待解决的问题,其中白细胞分割的准确性是正确分类的关键。提出了一种改进的迭代阈值图像分割算法,对恢复有丝分裂连线的最小距离方法进行了基于数学和数模分析的预处理改良,提升了白细胞图像分割的精度和效率,解决了血小板粘连、白细胞边界不分明等问题。将白细胞从复杂的血液环境中分离出来,对有丝分裂细胞的多叶核进行最小距离判定并连线,然后定位到每个白细胞,制作数据集,最后用CNN分类。经测试,分割正确率达到了96%以上。实验结果验证了所提分割方法的准确性、高效性和实用性。     

基于超声射频信号的支持向量机双参量B线识别方法*

肺超声中的特殊征象B线对于临床诊断肺水肿等肺部疾病有重要意义,但诊断结果依赖于医生的主观判断,为了客观、自动地识别B线,提高诊断准确率,本文提出了一种基于超声回波射频信号的肺脏超声特殊征象B线识别方法。本文首先选取了射频信号的排列熵、信息熵、峰度、偏度、能量作为特征参数,利用独立样本t检验和单参数贝叶斯分类的方法检验超声射频数据中B线以及非B线所对应射频数据的各个参量的差异性以及各参数与B线识别的相关性。然后将不同的双参量组合输入非线性支持向量机（SVM）中进行分类,比较各个组合的分类效果。结果显示信息熵与排列熵参数组合基于射频信号的分类效果最好,分类灵敏度为90.521%,特异性为98.106%,准确率为96.328%,AUC等于0.95。在引入后处理算法后,B线识别效果有进一步提升,得到分类平均灵敏度为95.23%,平均特异性为97.22%,平均准确率为96.88%。研究结果表明基于射频数据的SVM双参量B线识别方法对辅助临床诊断具有重要价值,信息熵和排列熵的组合可以有效的对特殊征象B线进行高精度识别。     

用于宫颈癌筛查的光电联合检测系统

为了提高宫颈癌的早期筛查效率以合理分配医疗资源,研制了基于光电联合检测的宫颈癌早期筛查系统.系统顶端放置的光电极手持探头采用电脉冲、红光、绿光及近红外光交替分时刺激宫颈组织,通过光电激励和测量,计算得到组织的光电学特征参数——电压衰减常数和相对反射率,用组织分类算法实现了宫颈癌变的判别.将313例临床检验数据与组织病理检验金标准比对,建立了光电激励下组织响应临床数据库.组织分类结果表明,光电联合检测方法比单独光或电的检测在灵敏度、特异性和总符合率上有较大的提高,与组织病理检验结果的对比表明本文所研发系统总符合率达85.1%,对早期宫颈癌的筛查尤其是落后地区宫颈癌的筛查具有重要意义.     

基于卷积神经网络与显微高光谱的胃癌组织分类方法研究

为了探究高光谱技术在胃癌组织病理诊断中的应用,将高光谱成像与显微系统结合,采集胃部切片组织的高光谱图像。针对胃癌组织与胃部正常组织在410~910nm波段的光谱特性差异,提出了一种基于卷积神经网络模型的胃癌组织分类方法,对原始光谱进行S-G平滑和一阶导数等预处理,通过分析光谱数据的特点和模型的分类效率,确定了最佳的网络结构及参数。实验结果表明:该模型对胃部癌变和正常组织的分类准确率为96.53%,鉴别胃癌组织的灵敏度和特异性分别达到94.29%和97.14%;相比于浅层学习方法,卷积神经网络模型能够充分提取癌变组织的深层光谱特征,同时能有效避免过拟合现象。将深度学习理论与显微高光谱结合的方法为医学病理研究提供了新思路。     

焊缝超声相控阵检测数据深度学习降噪方法

超声相控阵检测技术在焊缝检测中具有广泛的应用.超声相控阵检测技术检测信号中常混入噪声导致检测成像时难以分辨真实的缺陷特征.这些噪声主要为无关的反射信号和局部相关的结构噪声,传统的超声图像降噪方法难以有效滤除这些噪声,且存在计算效率低、参数优化复杂等问题.该文提出了一种基于深度学习的焊缝超声相控阵检测技术检测S扫图像的降...     

Application of artificial neural networks to X-ray fluorescence spectrum analysis

X-ray fluorescence (XRF) is widely applied as a mature nondestructive testing method, and appropriate improvement of quantitative analysis methods can improve the accuracy of XRF. Artificial neural network is an intelligent information processing system, its developments and application in XRF are reviewed, and representative models (back propagation, radial basis function, genetic algorithm artificial neural network, and others) are discussed in more details in overfitting, generalization, and algorithm efficiency. Potential directions of developing artificial neural network applied in XRF are proposed in this review as a further study.     

两种前向神经网络在变压器故障诊断中的应用

针对依靠变压器油中溶解气体分析的传统故障诊断方法存在的不足以及未来智能诊断算法进一步发展的需要,在研究人工神经网络的基础上,介绍了两种前向神经网络在变压器故障诊断中的应用,并用大量的DGA样本数据做了仿真训练。首先讨论了几种常用变压器故障诊断方法的一些缺陷,通过分析现代智能诊断算法的局限性,得出改进人工神经网络的优势。然后结合两种前向神经网络的原理和结构,对变压器内部故障进行了分类和编码。分别设计了相应的故障诊断模型,在MATLAB中做了仿真测试,并给出了仿真程序。仿真结果表明,两种神经网络用于变压器故障识别较为理想,其中PNN网络在样本数量较多时效果更好,有效提高了变压器故障诊断的正确率。     

Thyroid nodule classification using ultrasound elastography via linear discriminant analysis

The non-surgical diagnosis of thyroid nodules is currently made via a fine needle aspiration (FNA) biopsy. It is estimated that somewhere between 250,000 and 300,000 thyroid FNA biopsies are performed in the United States annually. However, a large percentage (approximately 70%) of these biopsies turn out to be benign. Since the aggressive FNA management of thyroid nodules is costly, quantitative risk assessment and stratification of a nodule’s malignancy is of value in triage and more appropriate healthcare resources utilization. In this paper, we introduce a new method for classifying the thyroid nodules based on the ultrasound (US) elastography features. Unlike approaches to assess the stiffness of a thyroid nodule by visually inspecting the pseudo-color pattern in the strain image, we use a classification algorithm to stratify the nodule by using the power spectrum of strain rate waveform extracted from the US elastography image sequence. Pulsation from the carotid artery was used to compress the thyroid nodules. Ultrasound data previously acquired from 98 thyroid nodules were used in this retrospective study to evaluate our classification algorithm. A classifier was developed based on the linear discriminant analysis (LDA) and used to differentiate the thyroid nodules into two types: (I) no FNA (observation-only) and (II) FNA. Using our method, 62 nodules were classified as type I, all of which were benign, while 36 nodules were classified as Type-II, 16 malignant and 20 benign, resulting in a sensitivity of 100% and specificity of 75.6% in detecting malignant thyroid nodules. This indicates that our triage method based on US elastography has the potential to substantially reduce the number of FNA biopsies (63.3%) by detecting benign nodules and managing them via follow-up observations rather than an FNA biopsy.     

人工神经网络用于光度法同时测定三组分染料混合物

应用人工神经网络原理,以快速BP算法,对紫外可见吸收光谱严重重叠的三组分的染料溶液同时进行含量测定。在200～590nm的范围内,以7个特征波长处的吸收值作为网络特征参数,通过网络训练,复品红、结晶紫、藏红T的相对标准偏差分别为0.34％,0.67％,1.03％,三者的回收率在95.5％～104％之间。实验表明,该算法速度快,预测结果准确,可望用人工神经网络和光度法结合定量测定混合染料。     

Sensor and Actuator Fault Diagnosis for Robot Joint Based on Deep CNN

This paper proposes a data-driven method-based fault diagnosis method using the deep convolutional neural network (DCNN). The DCNN is used to deal with sensor and actuator faults of robot joints, such as gain error, offset error, and malfunction for both sensors and actuators, and different fault types are diagnosed using the trained neural network. In order to achieve the above goal, the fused data of sensors and actuators are used, where both types of fault are described in one formulation. Then, the deep convolutional neural network is applied to learn characteristic features from the merged data to try to find discriminative information for each kind of fault. After that, the fully connected layer does prediction work based on learned features. In order to verify the effectiveness of the proposed deep convolutional neural network model, different fault diagnosis methods including support vector machine (SVM), artificial neural network (ANN), conventional neural network (CNN) using the LeNet-5 method, and long-term memory network (LTMN) are investigated and compared with DCNN method. The results show that the DCNN fault diagnosis method can realize high fault recognition accuracy while needing less model training time.     

基于聚焦光声层析技术的甲状腺离体组织成像

对人体甲状腺内的病变组织进行定位和成像对于准确诊断和有效治疗甲状腺疾病是至关重要的. 本文评估了利用光声层析技术对离体甲状腺组织进行成像的可行性, 并利用基于30 MHz超声换能器的聚焦光声成像系统对甲状腺进行扫描成像. 实验中成像系统的横向分辨率和纵向分辨率分别达到了350 μupm和74 μupm. 分别对正常离体甲状腺组织和模拟病变甲状腺组织进行光声成像. 实验结果表明, 本成像系统能够有效区分和鉴别正常甲状腺组织和病变组织. 此项技术有望进一步提高甲状腺疾病诊断的准确率, 以便更为有效地指导疾病的治疗, 具有潜在的临床应用前景.     

人工神经网络结合多光谱遥感数据用于立地质量评价

多光谱遥感数据蕴含着大量的地表立地信息,而传统立地质量评价体系主要使用了人工地面调查数据。为了建立一套有效的立地质量评价体系,以内蒙古赤峰市旺业甸林场为研究对象,基于研究区域的多光谱遥感数据结合地面小班调查数据,采用一种改进的反向传播人工神经网络(back Propagation artificial neural network,BPANN)模型,以落叶松为例,建立了遥感光谱因子结合立地因子与地位指数关系的神经网络模型,对研究区域的小班进行立地质量评价研究。通过训练数据集的敏感度分析剔除弱相关或不相关的因子,简化了神经网络的规模,提高了网络的训练效率,得到了最优的地位指数预测模型,模型的预测精度达到95.36%,与使用传统小班调查数据建立的神经网络模型的预测结果进行了比较,精度提高了9.83%,说明使用多光谱遥感数据+小班调查数据确定的落叶松地位指数预测模型具有最高的预测精度。多光谱遥感数据十分适用于森林立地质量评价,改进BP神经网络具有理想的预测精度,充分证实了该方法的有效性和优越性。     

Acoustic Radiation Force Impulse-Imaging for the evaluation of the thyroid gland: a limited patient feasibility study

Purpose

Real-time tissue elastography, a qualitative elastography method, has shown promising results in the diagnostic work up of thyroid nodules. However, to our knowledge no study has evaluated a quantitative elastography method in the thyroid gland. The present study is a feasibility study evaluating Acoustic Radiation Force Impulse-Imaging, a novel quantitative elastography method in the thyroid gland.

Methods

ARFI-imaging involves the mechanical excitation of tissue using short-duration acoustic pulses to generate localized displacements in tissue. The displacements induce a lateral shear-wave propagation which is tracked using multiple laterally positioned ultrasound “tracking“ beams. Inclusion criteria were: thyroid nodules ?1 cm, non-functioning or hypo-functioning on radionuclide scanning, and cytological/histological assessment of thyroid nodule as reference method. All patients received conventional ultrasound, and examination of the thyroid gland including Power Doppler Ultrasound using a 9 MHz linear transducer, in addition real-time elastography (RTE) was performed at 9 MHz frequency and ARFI-imaging was performed at 4 MHz using Siemens (ACUSON S2000) B-mode-ARFI combination transducer.

Results

Sixty nodules in 55 patients were analyzed. Three nodules were papillary carcinoma. The stiffer the tissue the faster the shear wave propagates. The results obtained indicated that the shear wave velocity in thyroid lobes ranged between 0.5 and 4.9 m/s. The median velocity of ARFI-imaging in the healthy nodule-free thyroid gland, as well as in benign and malignant thyroid nodules was 1.98 m/s (range: 1.20-3.63 m/s), 2.02 m/s (range: 0.92-3.97 m/s), and 4.30 m/s (range: 2.40-4.50 m/s), respectively. While no significant difference in median velocity was found between healthy thyroid tissue and benign thyroid nodules, a significant difference was found between malignant thyroid nodules on the one hand and healthy thyroid tissue (p = 0.018) or benign thyroid nodules (p = 0.014) on the other hand. Specificity of ARFI-imaging for the differentiation of benign and malignant thyroid nodules was comparable with RTE (91-95%).

Conclusions

ARFI can be performed in the thyroid tissue with reliable results.     

基于邻域粗糙集和并行神经网络的故障诊断

针对目前使用神经网络诊断故障时出现的输入向量选择困难、网络结构复杂、对并发故障诊断效果不好等问题,提出了基于邻域粗糙集和并行神经网络的故障诊断方法。先利用邻域粗糙集对初始征兆进行约简,留下有价值的征兆作为神经网络的输入向量,然后针对每种故障类型设计一个神经网络。用多个训练好的神经网络来并行诊断故障,综合每个神经网络的结果给出最终的诊断结论。用转子实验台的实验数据对这种故障诊断方法进行验证,结果显示该方法能优化神经网络结构,且神经网络具有训练速度快、诊断正确率高的特点。     

Control and diagnostic of the complex impedance of selected perovskite compounds

In this work the diagnosis control of the complex impedance of selected perovskite compounds versus artificial neural network model optimized with the Levenberg–Marquardt algorithm is performed as detection of aging and degradation of materials usually requires destructive testing. The artificial neural network optimized by the Levenberg–Marquardt algorithm used in this work allows us to monitor the materials (LaNd) SrMnCrO₃ in a non-destructive manner. This intelligent control is done by calculating the complex impedance which reveals reliable information on the phenomenon of transport in materials. The method overcomes the problem of the lack of a mathematical expression between the input parameters (temperature, doping, and frequency) and the necessary parameters for computing the impedance (bulk resistance, grain boundary resistance, and the two parameters of the constant phase element impedance A₀ and P). The robustness and performance of the artificial neural network model was verified by introducing additional noise and by using the root mean square error and the R-square.