Computer modeling of gait abnormalities in cerebral palsy: application to treatment planning

The treatment of gait abnormalities in persons with cerebral palsy is challenging. Theoretically, gait abnormalities can be diminished by decreasing the muscle forces that disrupt normal movement (e.g. via muscle–tendon lengthenings or tone-altering medications) and/or increasing the muscle and ground reaction forces that have the potential to improve movement (e.g. via strengthening exercises, orthoses, or derotational osteotomies). However, different patients exhibit varying degrees of neurologic impairment, spasticity, weakness, and bone deformity, suggesting that gait deviations arise from a variety of sources, each of which requires a different treatment. Treatment planning is further complicated because there is currently no scientific basis for determining how patients’ neuromusculoskeletal impairments contribute to abnormal movement. This paper describes how biomechanical models can be used, in combination with experimental data, to enhance our understanding of gait abnormalities and to provide a theoretical basis for planning treatments. Two examples are presented, and suggestions for future work are discussed.     

一种基于降维密度聚类的船舶异常轨迹识别方法

目的 有效分析和探索海洋船舶时空轨迹行为模式,提高船舶轨迹聚类的效率与质量,更好地检测真实船舶的异常行为。方法 针对当前船舶轨迹数据研究中存在的对多维特征信息利用不足、检测效率不高、检测精度较差等问题,提出一种精确度高、能自主识别分析多维特征的船舶异常轨迹识别方法。首先利用随机森林分类器评估多维特征重要性,构建轨迹特征的最优组合;然后提出一种降维密度聚类方法,将T–分布随机邻域嵌入(T–SNE)和自适应密度聚类(DBSCAN)模型结合,通过构建特征选择层和无监督聚类层实现对数据元素非线性关系的高效提取以及对聚类参数的智能选择;最后根据聚类结果构建类簇特征向量,计算距离阈值判别轨迹相似度,实现轨迹异常检测模型的构建。结果 以UCI数据集为例,降维密度聚类方法对4、13、30、64维特征数据集的F1分数能达到0.9 048、0.9 534、0.8 218、0.6 627,多个聚类指标均优于DBSCAN、K–Means等常见聚类算法的。结论 研究结果表明,降维密度聚类方法能有效提取数据多维特征结构,实现聚类参数自适应,弥补密度聚类中参数难以确定的问题,有效实现对多种类型船舶轨迹异常的识别。     

基于JACK虚拟仿真的学龄前脑瘫儿童助行器人机优化设计

目的 减少学龄前脑瘫儿童助行器的安全隐患,解决患儿在训练过程中的肢体二次损伤问题,提升助行器人机舒适性,对学龄前脑瘫儿童助行器进行优化设计。方法 通过JACK虚拟仿真软件的RULA工具,分析得到学龄前脑瘫儿童在使用助行器过程中存在的不合理姿势,并对其进行人机工学与受力分析,提出改进方法并进行设计优化,最后导入JACK虚拟仿真软件进行风险与舒适性验证。结果 优化后的学龄前脑瘫儿童助行器的使用受伤风险从4级降低到1级,患儿躯干弯曲力矩、L4/L5脊柱压缩力、剪切力的姿势载荷指数分别降低53%、47%、81%,上身各关节处于舒适角度。结论 经过JACK仿真分析及人机优化后的助行器,有效降低了学龄前脑瘫儿童在康复训练中的二次损伤风险,减轻了患儿步行训练中的身体负担,提升了助行器使用的舒适性。     

Dialysis disequilibrium syndrome presenting as a focal neurological deficit

We report a patient suffering from chronic kidney disease who presented to us with severe pulmonary edema. His clinical, laboratory, and sonological parameters were suggestive of end-stage renal disease. Hemodialysis was initiated, and after 48 hours (3 sessions of hemodialysis) he became drowsy and a neurological examination revealed left upper limb monoplegia with left facial palsy. Urgent computerized tomography scan of the brain revealed diffuse hypodensity in the cerebral white matter bilaterally, and brain magnetic resonance imaging showed diffuse hyperintensity in the cerebral white matter bilaterally, right internal capsule and external capsule on fluid attenuated inversion recovery and T2 sequences (hypointense on T1 sequence). He made a gradual but complete neurological recovery and was discharged 2 weeks later with normal neurological status. A repeat brain magnetic resonance imaging on follow-up 6 weeks later revealed complete resolution of the white matter abnormalities.     

悬架筒式减振器异响鉴别系统设计

针对某车型悬架筒式减振器异响问题,分别进行整车路试和台架试验,在收集大量数据情况下,提出采用基于权重的聚类分析法鉴别减振器异响,并分析其中的原理。在此基础上,通过Lab VIEW虚拟仪器软件开发平台与MATLAB信号处理工具箱相结合,设计一套用于鉴别车辆悬架筒式减振器异响的测试系统,并对减振器进行异响鉴别。结果表明,该测试系统鉴别结果与减振器整车路试主观评价相关系数高于0.92,能够满足工厂企业产品质量控制方面检测要求,可为今后产品抽样检测、提高异响鉴别可靠度和改进生产过程提供参考。     

脑部螺旋CT血管造影的技术探讨和临床应用

５１例病人均采用ＰＩＣＫＥＲ６０００ＣＴ常规轴位及注射对比剂延时增强,并螺旋ＣＴＡ原始图像数据采集,病人分Ａ、Ｂ、Ｃ三组进行研究,表明脑部ＳＣＴＡ采用层厚３ｍｍ,层距３ｍｍ,加薄ＩＮＤＥＸ１．５ｍｍ,ＭＩＰ重建,图像质量均达到１级,大脑中动脉可显示至４级,认为脑部ＣＴＡ扫描参数及造影时间的选择很重要,它是一种无创伤性检查方法,值得推广应用。     

An efficient and automatic glioblastoma brain tumor detection using shift‐invariant shearlet transform and neural networks

The detection and segmentation of tumor region in brain image is a critical task due to the similarity between abnormal and normal region. In this article, a computer‐aided automatic detection and segmentation of brain tumor is proposed. The proposed system consists of enhancement, transformation, feature extraction, and classification. The shift‐invariant shearlet transform (SIST) is used to enhance the brain image. Further, nonsubsampled contourlet transform (NSCT) is used as multiresolution transform which transforms the spatial domain enhanced image into multiresolution image. The texture features from grey level co‐occurrence matrix (GLCM), Gabor, and discrete wavelet transform (DWT) are extracted with the approximate subband of the NSCT transformed image. These extracted features are trained and classified into either normal or glioblastoma brain image using feed forward back propagation neural networks. Further, K‐means clustering algorithm is used to segment the tumor region in classified glioblastoma brain image. The proposed method achieves 89.7% of sensitivity, 99.9% of specificity, and 99.8% of accuracy.     

Imaging manifestations of sepsis‐associated encephalopathy

The purpose of this study was to retrospectively analyze the clinical and imaging data of sepsis‐associated encephalopathy (SAE) following infantile diarrhea. Eight infants were diagnosed with SAE after diarrhea and assessed using computed tomography (CT) and magnetic resonance imaging (MRI). The main symptoms present in each of the eight patients were fever, diarrhea, seizures, and changes in consciousness. From cranial CT examination, five patients showed normal results, two displayed cerebral edema, and one displayed slightly lower density areas in the frontal and parietal lobes. There were four patients examined through MRI: one patient displayed slightly widened cerebral sulci; two displayed edema in the white cerebral matter and basal ganglia with gyriform enhancement after contrast; one displayed an abnormal bilateral signal in the occipital lobes and another in the frontoparietal lobe; and one displayed a diffuse abnormal signal in the white matter of both cerebral hemispheres and basal ganglia. Imaging manifestations of SAE included encephaledema or abnormal signals in the white matter and basal ganglia. CT examination can exclude other cerebral pathologies causing brain dysfunction in early stages. MRI examination can provide more information to aid in the early diagnosis of SAE.     

Glioma tumor detection in brain MRI image using ANFIS‐based normalized graph cut approach

Glioma is the severe type of brain tumor which leads to immediate death for the case high‐grade Glioma. The Glioma tumor patient in case of low grade can extend their life period if tumor is timely detected and providing proper surgery. In this article, a computer‐aided fully automated Glioma brain tumor detection and segmentation system is proposed using Adaptive Neuro Fuzzy Inference System (ANFIS) classifier based Graph cut approach. Initially, orientation analysis is performed on the brain image to obtain the edge enhanced abnormal regions in the brain. Then, features are extracted from the orientation enhanced image and these features are trained and classified using ANFIS classifier to classify the test brain image into either normal or abnormal. Normalized Graph cur segmentation methodology is applied on the classified abnormal brain image to segment the tumor region. The proposed Glioma tumor segmentation method is validated using the metric parameters as sensitivity, specificity, accuracy and dice similarity coefficient.     

A deep learning model integrating convolution neural network and multiple kernel K means clustering for segmenting brain tumor in magnetic resonance images

In medical imaging, segmenting brain tumor becomes a vital task, and it provides a way for early diagnosis and treatment. Manual segmentation of brain tumor in magnetic resonance (MR) images is a time‐consuming and challenging task. Hence, there is a need for a computer‐aided brain tumor segmentation approach. Using deep learning algorithms, a robust brain tumor segmentation approach is implemented by integrating convolution neural network (CNN) and multiple kernel K means clustering (MKKMC). In this proposed CNN‐MKKMC approach, classification of MR images into normal and abnormal is performed by CNN algorithm. At next, MKKMC algorithm is employed to segment the brain tumor from the abnormal brain image. The proposed CNN‐MKKMC algorithm is evaluated both visually and objectively in terms of accuracy, sensitivity, and specificity with the existing segmentation methods. The experimental results demonstrate that the proposed CNN‐MKKMC approach yields better accuracy in segmenting brain tumor with less time cost.     

基于一类超球面支持向量机的机械故障诊断研究

针对机械故障诊断中故障类样本不易获取以及样本分布不均的问题,提出了基于一类超球面支持向量机(SVM)的故障诊断方法,该方法只需要对正常类样本进行训练.试验分析了异常类样本缺失对一类超球面支持向量机性能的影响,并提出模型参数优化选择方法,以提高分类模型的推广能力.分析了不同训练结果的分类能力,并对一类超球面支持向量机与一类超平面支持向量机的分类结果进行比较,验证了前者的正确性和有效性.     

机器学习分布式网络传输异常数据智能检测方法

对于链路状态数据库的网络传输异常数据检测存在检测数据不完整、较为敏感、检测效率差的问题,提出基于机器学习的分布式网络传输异常数据智能检测方法,通过K最近邻分簇算法对分布式网络节点实施分簇,利用贝叶斯分类算法检测簇头是否出现异常;确定异常簇后,选取小波阈值降噪方法对异常簇内数据进行降噪处理,在此基础上,采用遗传算法检测降噪处理后异常簇内的异常数据,通过群体内最佳个体与最差个体的适应度函数值的差值同既定阈值的比较结果得到最终异常数据。经实验证明,所提方法检测异常数据的平均时间为8.48 s,检测结果与实际结果相似性较高,且检测性能较为稳定,说明该方法具有较高的异常数据检测性能。     

Using Imbalanced Triangle Synthetic Data for Machine Learning Anomaly Detection

The extreme imbalanced data problem is the core issue in anomaly detection. The amount of abnormal data is so small that we cannot get adequate information to analyze it. The mainstream methods focus on taking fully advantages of the normal data, of which the discrimination method is that the data not belonging to normal data distribution is the anomaly. From the view of data science, we concentrate on the abnormal data and generate artificial abnormal samples by machine learning method. In this kind of technologies, Synthetic Minority Over-sampling Technique and its improved algorithms are representative milestones, which generate synthetic examples randomly in selected line segments. In our work, we break the limitation of line segment and propose an Imbalanced Triangle Synthetic Data method. In theory, our method covers a wider range. In experiment with real world data, our method performs better than the SMOTE and its meliorations.     

Hybrid GrabCut Hidden Markov Model for Segmentation

Diagnosing data or object detection in medical images is one of the important parts of image segmentation especially those data which is less effective to identify in MRI such as low-grade tumors or cerebral spinal fluid (CSF) leaks in the brain. The aim of the study is to address the problems associated with detecting the low-grade tumor and CSF in brain is difficult in magnetic resonance imaging (MRI) images and another problem also relates to efficiency and less execution time for segmentation of medical images. For tumor and CSF segmentation using trained light field database (LFD) datasets of MRI images. This research proposed the new framework of the hybrid k-Nearest Neighbors (k-NN) model that is a combination of hybridization of Graph Cut and Support Vector Machine (GCSVM) and Hidden Markov Model of k-Mean Clustering Algorithm (HMMkC). There are four different methods are used in this research namely (1) SVM, (2) GrabCut segmentation, (3) HMM, and (4) k-mean clustering algorithm. In this framework, on the one hand, phase one is to perform the classification of SVM and Graph Cut algorithm to create the maximum margin distance. This research use GrabCut segmentation method which is the application of the graph cut algorithm and extract the data with the help of scale-invariant features transform. On the other hand, in phase two, segment the low-grade tumors and CSF using a method adapted for HMkC and extract the information of tumor or CSF fluid by GCHMkC including iterative conditional maximizing mode (ICMM) with identifying the range of distant. Comparative evaluation is also performing by the comparison of existing techniques in this research. In conclusion, our proposed model gives better results than existing. This proposed model helps to common man and doctor that can identify their condition of brain easily. In future, this will model will use for other brain related diseases.     

Nanoparticle-based in vivo investigation on blood-brain barrier permeability following ischemia and reperfusion

The blood-brain barrier (BBB) represents a significant impediment to a large variety of central nervous system-active agents. In the current study, we applied fluorescent polystyrene nanospheres (20 nm) to study the BBB permeability following cerebral ischemia and reperfusion. A microdialysis probe was implanted in the cerebral cortex of an anesthetized rat injected with fluorescent polystyrene nanospheres. The circulating nanospheres extravasating to the brain extracellular fluids were collected by the probe. Fluorescence intensity in the microdialysates throughout the course of cerebral ischemia/reperfusion was measured. Cerebral ischemia and reperfusion induced transient accumulations of extracellular nanospheres in the brain. The accumulation of nanospheres may result from their extravasation from the blood vessels. The concurrent cerebral oxygen levels monitored using oxygen-dependent quenching of phosphorescence decreased following ischemia and returned to their original levels after reperfusion. In conclusion, we demonstrated that high temporal resolution measurements of BBB permeability in vivo can be obtained using fluorescence polystyrene nanospheres and that these data correlate with changes of cerebral oxygen concentration. This present investigation indicates that nanoparticles have potential clinical applications involving drug delivery and determination of therapeutic efficacy and on-site diagnosis.     

产品声音的交互式聚类设计研究

目的 声音是产品和用户之间的一种沟通媒介,为了增进设计师对产品声音的理解、合成与设计匹配,提出一种交互式可视化产品声音数据聚类分析框架。方法 首先通过神经网络将设计师感官描述式信息与声音的特征参数进行融合嵌套;其次基于高斯混合模型来描述非线性几何分布的产品声音数据;最后设计师输入个人先验知识经验参与交互聚类。结果 基于Python的Anaconda3包开发了产品声音交互式聚类的可视化分析实验工具,得到最优化产品声音聚类结果。结论 该产品声音交互聚类可视化分析工具融合了声音技术参数和人脑听觉反应机制,在聚类过程中允许用户参与交互并融入用户的先验知识,并行视图可以实时显示数据元素的流向和判别类别的稳定性。同时,可视化分析可以帮助用户横向比较各聚类结果的异同,样本的比例分布与合理性,以期寻求最优聚类结果。     

Nonparametric variational learning of multivariate beta mixture models in medical applications

Clustering as an essential technique has matured into a capable solution to address the gap between the growing availability of data and deriving the knowledge from them. In this paper, we propose a novel clustering method “variational learning of infinite multivariate Beta mixture models.” The motivation behind proposing this technique is the flexibility of mixture models to fit the data. This approach has the capability to infer the model complexity and estimate model parameters from the observed data automatically. Moreover, as a label‐free method, it could also address the problem of high costs of medical data labeling, which can be undertaken just by medical experts. The performance of the model is evaluated on real medical applications and compared with other similar alternatives. We demonstrate the ability of our proposed method to outperform widely used methods in the field as it has been shown in experimental results.     

Blood-flow models of the circle of Willis from magnetic resonance data

Detailed knowledge of the cerebral hemodynamics is important for a variety of clinical applications. Cerebral perfusion depends not only on the status of the diseased vessels but also on the patency of collateral pathways provided by the circle of Willis. Due to the large anatomical and physiologic variability among individuals, realistic patient-specific models can provide new insights into the cerebral hemodynamics. This paper presents an image-based methodology for constructing patient-specific models of the cerebral circulation. This methodology combines anatomical and physiologic imaging techniques with computer simulation technology. The methodology is illustrated with a finite element model constructed from magnetic resonance image data of a normal volunteer. Several of the remaining challenging problems are identified. This work represents a starting point in the development of realistic models that can be applied to the study of cerebrovascular diseases and their treatment.     

基于层次的K-means初始化算法

K-means算法是一种常用的聚类算法,但是聚类中心的初始化是其中的一个难点。笔者提出了一个基于层次思想的初始化方法。一般聚类问题均可看作加权聚类,通过层层抽样减少数据量,然后采用自顶向下的方式,从抽样结束层到原始数据层,每层都进行聚类,其中每层初始聚类中心均通过对上层聚类中心进行换算得到,重复该过程直到原始数据层,可得原始数据层的初始聚类中心。模拟数据和真实数据的实验结果均显示基于层次抽样初始化的K-means算法不仅收敛速度快、聚类质量高,而且对噪声不敏感,其性能明显优于现有的相关算法。     

基于语音交互的脑瘫康复训练系统的设计

研究了使用人机交互的方式对脑瘫患儿进行康复训练的技术,设计了基于语音交互的脑瘫康复训练系统。该系统采用51单片机驱动整个系统,芯片采用LD3320语音芯片,下位机程序采用C语言编程,上位机采用Lab VIEW编程,通过串口传送的方式,把下位机接收到的命令传送到上位机,患儿通过识别上位机界面的图片,对应说出所要识别的文字,语音芯片进行识别,评价系统自动评判出患儿说的正确与否,并记录患儿每次识别所用的时间。案例实验验证了本设计的有效性。