基于脑部磁共振图像三维局部模式变换特征提取进行阿尔茨海默病病程预测分类

本文提出一种三维局部模式变换提取进行纹理特征并与常规特征相融合的方法,基于脑部磁共振图像,对认知功能正常的健康人体（CN）、轻度认知障碍（MCI）患者和阿尔茨海默病（AD）患者进行预测分类.首先对46例CN对照组、61例MCI患者和25例AD患者的脑部磁共振图像提取感兴趣区域,然后提取双侧海马体组织、灰质和白质的三维局部模式变换纹理特征和常规特征,并将两类特征融合,使用支持向量机分类算法进行分类.结果显示利用本方法,基于双侧海马体组织对AD组和CN组进行分类的准确率为88.73%、敏感度为78.00%、特异度为95.7%、受试者工作特征（ROC）曲线下面积（AUC）为0.886 5;基于灰质的准确率为85.92%、敏感度为80.00%、特异度为86.6%、AUC为0.854 3.这证明基于海马体磁共振图像,利用本文提出的改进三维局部模式变换提取的纹理特征进行阿尔茨海默病病程分类效果较好,融合常规特征后更可提高分类预测的精度.     

结合小波融合和深度学习的脑胶质瘤自动分割

针对水肿区域边界模糊和瘤内结构复杂多变导致的脑胶质瘤分割不精确问题，本文提出了一种基于小波融合和3D-UNet网络的脑胶质瘤磁共振图像自动分割算法.首先，对脑胶质瘤磁共振图像的T1、T1ce、T2、Flair四种模态进行小波融合以及偏置场校正；然后，提取待分类的图像块；再利用提取的图像块训练3D-UNet网络以对图像块中的像素进行分类；最后加载损失率较小的网络模型进行分割，并采用基于连通区域的轮廓提取方法，以降低假阳性率.对57组Brats2018（Brain Tumor Segmentation 2018）磁共振图像测试集进行分割的结果显示，肿瘤的整体、核心和水肿部分的平均分割准确率（DSC）分别达到90.64%、80.74%和86.37%，这表明该算法分割脑胶质瘤准确率较高，与金标准相近.相比多模态图像融合前，该算法在减少输入网络数据量和图像冗余信息的同时，还一定程度上解决了胶质瘤边界模糊、分割不精确的问题，提高了分割的准确度和鲁棒性.     

基于红外和可见光图像逐级自适应融合的场景深度估计

从图像中恢复场景的深度是计算机视觉领域中的一个关键问题。考虑到单一类型图像在深度估计中受场景不同光照的限制，提出了基于红外和可见光图像逐级自适应融合的场景深度估计方法（PF-CNN）。该方法包括双流滤波器部分耦合网络、自适应多模态特征融合网络以及自适应逐级特征融合网络。在双流卷积中红外和可见光图像的滤波器部分耦合使两者特征得到增强；自适应多模态特征融合网络学习红外和可见光图像的残差特征并将两者自适应加权融合，充分利用两者的互补信息；逐级特征融合网络学习多层融合特征的结合，充分利用不同卷积层的不同特征。实验结果表明:PF-CNN在测试集上获得了较好的效果，将阈值指标提高了5%，明显优于其他方法。     

基于改进U-Net++的CT影像肺结节分割算法

卷积神经网络的语义分割模型未有效利用特征权重信息,导致在医学图像复杂场景中分割边界出现欠分割现象.针对该问题,基于融合自适应加权聚合策略提出一种改进的U-Net++网络,并将其应用于电子计算机断层扫描影像肺结节分割.该模型首先在卷积神经网络中提取出不同深度特征语义级别的信息,再结合权重聚合模块,自适应地学习各层特征的权...     

基于词向量一致性融合的遥感场景零样本分类方法

遥感场景类别的语义词向量与图像特征原型的距离结构不一致问题,严重影响遥感场景零样本分类效果。针对该问题,利用不同词向量间一致性,提出一种基于解析字典学习的语义词向量融合方法,以提升遥感场景零样本分类效果。首先,采用解析字典学习方法,提取场景类别的不同词向量的公共稀疏系数,并作为融合后的语义词向量;然后,同样采用解析字典学习方法,将场景类别的图像特征原型嵌入到融合后的词向量空间,与融合后的词向量进行结构对齐,降低距离结构的不一致性;最后,通过联合优化获得未知类的图像特征空间类别原型表示,并采用最近邻分类器完成未知类别遥感场景的分类。在3种遥感场景数据集和多种语义词向量上进行定量和定性实验。实验结果表明,通过词向量融合可以获得与图像特征原型结构更一致的语义词向量,从而显著提升遥感场景零样本分类的准确度。     

一种以超分辨率理论为基础的磁共振眼球成像方法

磁共振成像（MRI）是一种无电离辐射的非介入性的眼内肿瘤检测方法,但分辨率和运动伪影是成像过程中不易克服的困难．以往的扫描方法或是不可避免的引入运动伪影,或是需要受试者做精确的配合,增加了成像的难度,给受试者带来不舒适的体验．本文提出了一种以超分辨率理论为基础的新的磁共振眼球成像方法,使用一种特制的眼球线圈,对眼部区域扫描一系列动态的图像,使得不同方向上的采集分辨率互补．最后经过预处理、配准、超分辨率重建等操作,得到高质量的磁共振眼球图像．实验结果表明,这种方法可以在不需要受试者做额外配合工作的情况下,得到更加清晰的磁共振眼球图像．     

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.     

基于编码-解码卷积神经网络的遥感图像语义分割

农村地区遥感图像语义分割是进行城乡规划、植被以及农用地检测的基础。农村地区高分辨率遥感图像含有较为复杂的地物信息,对其进行语义分割难度较大。基于此,提出一种改进的对称编码-解码网络结构SegProNet,利用池化索引与卷积融合语义信息及图像特征,通过1×1卷积构建Bottleneck层进一步提取细节、减少参数量,逐步加深过滤器深度以构建端到端的语义分割网络,改进激活函数进一步提升网络性能。实验结果表明,在CCF卫星数据集上,所提方法及经典语义分割网络U-Net、SegNet的准确率分别为98.4%,80.3%,98.1%,所提方法较其他方法更优。     

基于深度学习的光学元件表面疵病识别北大核心CSCD

精密光学元件表面疵病的人工检测分类方法效率低,且准确率易受疲劳等人工因素影响,而基于传统机器学习方法的分类准确率有待进一步提高。提出了一种基于深度学习卷积神经网络的光学大尺寸元件表面疵病识别方法。首先,通过现场实验采集并整理了大尺寸镜面疵病样本;接着,基于单通道灰度图像构建融合梯度的三通道图像,挖掘更深入的特征表达;最后,基于经典的LeNet网络,提出了面向激光惯性约束聚变(ICF)的光学元件表面疵病识别网络ICFNet,该网络不需要复杂的手工特征设计和提取,仅使用原始灰度图像就实现高效的疵病识别。实验结果表明:针对包含麻点、划痕和灰尘的三类疵病数据,ICFNet相较于使用多项特征和支持向量机的传统方法拥有较好的分类准确率。     

基于级联卷积神经网络的前列腺磁共振图像分类

针对深度学习训练成本高，以及基于磁共振图像的前列腺癌临床诊断需要大量医学常识且极为耗时的问题，本文提出了一种基于级联卷积神经网络（Convolutional Neural Network，CNN）和磁共振图像的前列腺癌（Prostate Cancer，PCa）自动分类诊断方法，该网络以Faster-RCNN作为前网络，对前列腺区域进行提取分割，用于排除前列腺附近组织器官的干扰；以基于ResNet改进的网络结构CNN40_bottleneck作为后网络，用于对前列腺区域病变进行分类.后网络由瓶颈结构串联组成，其中使用批量标准化（Batch Normalization，BN）、全局平均池化（Global Average Pooling，GAP）进行优化.实验结果证明，本文方法对前列腺癌诊断结果较好，而且缩减了训练时间和参数量，有效降低了训练成本.     

A 3D densely connected convolution neural network with connection-wise attention mechanism for Alzheimer's disease classification

PurposeAlzheimer's disease (AD) is a progressive and irreversible neurodegenerative disease. In recent years, machine learning methods have been widely used on analysis of neuroimage for quantitative evaluation and computer-aided diagnosis of AD or prediction on the conversion from mild cognitive impairment (MCI) to AD. In this study, we aimed to develop a new deep learning method to detect or predict AD in an efficient way.Materials and methodsWe proposed a densely connected convolution neural network with connection-wise attention mechanism to learn the multi-level features of brain MR images for AD classification. We used the densely connected neural network to extract multi-scale features from pre-processed images, and connection-wise attention mechanism was applied to combine connections among features from different layers to hierarchically transform the MR images into more compact high-level features. Furthermore, we extended the convolution operation to 3D to capture the spatial information of MRI. The features extracted from each 3D convolution layer were integrated with features from all preceding layers with different attention, and were finally used for classification. Our method was evaluated on the baseline MRI of 968 subjects from ADNI database to discriminate (1) AD versus healthy subjects, (2) MCI converters versus healthy subjects, and (3) MCI converters versus non-converters.ResultsThe proposed method achieved 97.35% accuracy for distinguishing AD patients from healthy control, 87.82% for MCI converters against healthy control, and 78.79% for MCI converters against non-converters. Compared with some neural networks and methods reported in recent studies, the classification performance of our proposed algorithm was among the top ranks and improved in discriminating MCI subjects who were in high risks of conversion to AD.ConclusionsDeep learning techniques provide a powerful tool to explore minute but intricate characteristics in MR images which may facilitate early diagnosis and prediction of AD.     

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。     

Individual identification using multi-metric of DTI in Alzheimer's disease and mild cognitive impairment

Accurate identification of Alzheimer's disease(AD) and mild cognitive impairment(MCI) is crucial so as to improve diagnosis techniques and to better understand the neurodegenerative process. In this work, we aim to apply the machine learning method to individual identification and identify the discriminate features associated with AD and MCI. Diffusion tensor imaging scans of 48 patients with AD, 39 patients with late MCI, 75 patients with early MCI, and 51 age-matched healthy controls(HCs) are acquired from the Alzheimer's Disease Neuroimaging Initiative database. In addition to the common fractional anisotropy, mean diffusivity, axial diffusivity, and radial diffusivity metrics, there are two novel metrics,named local diffusion homogeneity that used Spearman's rank correlation coefficient and Kendall's coefficient concordance,which are taken as classification metrics. The recursive feature elimination method for support vector machine(SVM)and logistic regression(LR) combined with leave-one-out cross validation are applied to determine the optimal feature dimensions. Then the SVM and LR methods perform the classification process and compare the classification performance.The results show that not only can the multi-type combined metrics obtain higher accuracy than the single metric, but also the SVM classifier with multi-type combined metrics has better classification performance than the LR classifier.Statistically, the average accuracy of the combined metric is more than 92% for all between-group comparisons of SVM classifier. In addition to the high recognition rate, significant differences are found in the statistical analysis of cognitive scores between groups. We further execute the permutation test, receiver operating characteristic curves, and area under the curve to validate the robustness of the classifiers, and indicate that the SVM classifier is more stable and efficient than the LR classifier. Finally, the uncinated fasciculus, cingulum, corpus callosum, corona radiate, external capsule, and internal capsule have been regarded as the most important white matter tracts to identify AD, MCI, and HC. Our findings reveal a guidance role for machine-learning based image analysis on clinical diagnosis.     

One- and Two-Phase Software Requirement Classification Using Ensemble Deep Learning

Recently, deep learning (DL) has been utilized successfully in different fields, achieving remarkable results. Thus, there is a noticeable focus on DL approaches to automate software engineering (SE) tasks such as maintenance, requirement extraction, and classification. An advanced utilization of DL is the ensemble approach, which aims to reduce error rates and learning time and improve performance. In this research, three ensemble approaches were applied: accuracy as a weight ensemble, mean ensemble, and accuracy per class as a weight ensemble with a combination of four different DL models—long short-term memory (LSTM), bidirectional long short-term memory (BiLSTM), a gated recurrent unit (GRU), and a convolutional neural network (CNN)—in order to classify the software requirement (SR) specification, the binary classification of SRs into functional requirement (FRs) or non-functional requirements (NFRs), and the multi-label classification of both FRs and NFRs into further experimental classes. The models were trained and tested on the PROMISE dataset. A one-phase classification system was developed to classify SRs directly into one of the 17 multi-classes of FRs and NFRs. In addition, a two-phase classification system was developed to classify SRs first into FRs or NFRs and to pass the output to the second phase of multi-class classification to 17 classes. The experimental results demonstrated that the proposed classification systems can lead to a competitive classification performance compared to the state-of-the-art methods. The two-phase classification system proved its robustness against the one-phase classification system, as it obtained a 95.7% accuracy in the binary classification phase and a 93.4% accuracy in the second phase of NFR and FR multi-class classification.     

Deep learning based classification of focal liver lesions with contrast-enhanced ultrasound

Classification of liver masses is important to early diagnosis of patients. In this paper, a diagnostic system of liver disease classification based on contrast enhanced ultrasound (CEUS) imaging is proposed. In the proposed system, the dynamic CEUS videos of hepatic perfusion are firstly retrieved. Secondly, time intensity curves (TICs) are extracted from the dynamic CEUS videos using sparse non-negative matrix factorizations. Finally, deep learning is employed to classify benign and malignant focal liver lesions based on these TICs. Quantitative comparisons demonstrate that the proposed method outperforms the compared classification methods in accuracy, sensitivity and specificity.     

一维空洞卷积神经网络的矿物光谱分类

矿物光谱综合反映了岩矿的物理化学特性、组分和内部结构特征,已被应用于岩矿识别研究.传统的矿物光谱分类方法需要先对矿物光谱进行预处理,再采用不同方法分析光谱特征,从而实现分类目的.但同时也会造成部分光谱信息丢失,导致最终分类精度不高且操作过程繁琐、效率低下,难以应对日益增长的大数据处理需求.因此,建立一个准确、高效的矿物...     

基于卷积神经网络的焊缝缺陷图像分类研究

为有效地对焊缝缺陷进行分类，从而判断焊接质量的等级，对传统卷积神经网络进行改进，提出一种多尺度压缩激励网络模型（SINet）。将4组两两串联的3×3卷积模块与Inception模块、压缩激励模块（SE block）相结合。通过多尺度压缩激励模块（SI module）将卷积层中的特征进行多尺度融合和特征重标定以提高分类准确率，并用全局平均池化层代替全连接层减少模型参数。此外考虑到焊接缺陷数量不平衡对准确率的影响，采用深度卷积对抗生成网络（DCGAN）进行数据集的平衡处理，并在该数据集上验证模型的有效性。与传统卷积神经网络相比，该模型具有良好的性能，在测试集上准确率达到96.77%，同时模型的参数个数也明显减少。结果表明该方法对焊缝缺陷图像能进行有效地分类。     

基于MRI和深度学习的桥小脑角区脑膜瘤与听神经瘤分类算法研究

桥小脑角区脑膜瘤与听神经瘤是两种常见的脑部肿瘤，它们的临床表现和影像学表现极为相似，在临床诊断时极易发生误诊．将影像数据与深度学习方法相结合，建立脑膜瘤与听神经瘤的判别模型，可以为两种脑肿瘤的及时准确诊断提供重要手段．本文采集了307名脑肿瘤患者的T₁W-SE序列图像，通过对原始图像进行限制对比度自适应直方图均衡化（Contrast Limited Adaptive Histogram Equalization，CLAHE）等预处理，提升数据集图像质量，再经过建立的三维卷积神经网络（3-Dimensional Convolutional Neural Network，3D CNN）深度学习框架中图像特征的学习，实现对脑膜瘤与听神经瘤的分类．图像增强参数与网络结构参数经过优化后，对脑膜瘤与听神经瘤分类的准确率达到0.918 0，曲线下面积（Area Under Curve，AUC）为0.913 4，实现了对桥小脑角区脑膜瘤与听神经瘤的有效判别．     

基于卷积神经网络与光谱特征的夏威夷果品质鉴定研究

夏威夷果含油量高,在开缝之后容易发生变质,现有关于夏威夷果品质鉴定的方法多为传统的破坏性检验,很难满足无损检测的需求。卷积神经网络(CNN)作为应用最广泛的深度学习网络模型之一,具有比浅层学习方法更强的特征提取与模型表达能力,在光谱数据方面的应用拥有很大潜力。基于夏威夷果在可见-近红外的光谱特征分析,研究用于提取夏威夷果光谱特征的卷积神经网络模型,并提出一种高效无损鉴定夏威夷果品质的方法。首先以三种不同品质的夏威夷果(好籽、哈籽及霉籽)为研究对象,分析样本在500~2 100 nm的光谱信息;在光谱数据预处理中引入白化处理方法,用以增强数据的相关性差异;然后在模型训练过程中,将样本随机分为训练集和预测集,探讨不同CNN结构、卷积层数、卷积核大小及个数、池化层类型、全连接层神经元个数以及激活函数对分类结果的影响,并采用激活函数ReLU和Dropout方法,预防样本数据过少引起的过拟合现象;最后通过分析模型分类准确率和计算效率,确定了一个6层结构的CNN模型： 输入层-卷积层-池化层-全连接层(200神经元)-全连接层(100神经元)-输出层。实验结果表明： 上述网络模型对校正集和预测集的分类准确率均达到100％。因此,改进后的卷积神经网络模型可充分学习夏威夷果的光谱特征并有效分类,将深度学习理论与光谱分析相结合的方法能够实现对夏威夷果品质的准确鉴定,同时为夏威夷果等坚果类食品的高效、无损、实时在线检测提供了新思路。