Machine learning models for Parkinson’s disease detection and stage classification based on spatial-temporal gait parameters

BackgroundParkinson’s disease (PD) is a chronic and progressive neurodegenerative disease with no cure, presenting a challenging diagnosis and management. However, despite a significant number of criteria and guidelines have been proposed to improve the diagnosis of PD and to determine the PD stage, the gold standard for diagnosis and symptoms monitoring of PD is still mainly based on clinical evaluation, which includes several subjective factors. The use of machine learning (ML) algorithms in spatial-temporal gait parameters is an interesting advance with easy interpretation and objective factors that may assist in PD diagnostic and follow up.Research questionThis article studies ML algorithms for: i) distinguish people with PD vs. matched-healthy individuals; and ii) to discriminate PD stages, based on selected spatial-temporal parameters, including variability and asymmetry.MethodsGait data acquired from 63 people with PD with different levels of PD motor symptoms severity, and 63 matched-control group individuals, during self-selected walking speed, was study in the experiments.ResultsIn the PD diagnosis, a classification accuracy of 84.6 %, with a precision of 0.923 and a recall of 0.800, was achieved by the Naïve Bayes algorithm. We found four significant gait features in PD diagnosis: step length, velocity and width, and step width variability. As to the PD stage identification, the Random Forest outperformed the other studied ML algorithms, by reaching an Area Under the ROC curve of 0.786. We found two relevant gait features in identifying the PD stage: stride width variability and step double support time variability.SignificanceThe results showed that the studied ML algorithms have potential both to PD diagnosis and stage identification by analysing gait parameters.     

医学图像光线投影体绘制中一种改进的快速算法

光线投影算法是体绘制算法中图像效果比较好的方法，但存在运算量大，绘制速度慢的问题。为此本文提出了一种新的光线投影体绘制的加速算法，利用重采样点在两坐标系中的矩阵变换特性。减少矩阵运算量，加快重采样计算过程，并且通过将Bresenham算法扩展至三维，利用包围盒技术避免对空体元的采样，从而加速了光线投影的效率。实验结果表明改进后的加速算法，既能保证绘制质量，又能显著减少计算量，提高体绘制的速度。     

基于经验模式分解的心电特征提取算法

本研究应用基于经验模式分解的心电特征提取方法,利用第一本征模函数(intrinsic mode function,IMF)分量对QRS波进行定位,并通过减少分解层数、筛选次数、处理区域等策略实现了快速算法。利用MIT-BIT心律失常数据库的数据进行算法测试,取得较高的检测率,检测速度也有明显提高。实验结果表明,经验模式分解算法在QRS波定位中具有相当的优越性,临床应用中取得了良好的检测效果。     

Recurrent wavefront morphologies: A method for quantifying the complexity of epicardial activation patterns

We have developed a method for quantifying the complexity of activation patterns observed during ventricular fibrillation (VF) that is based on our previously reported methodology for decomposing epicardial mapping data into a set of isolated wavefronts. One-half second datasets are acquired from a 21×24 array of unipolar electrodes (1 mm spacing), and the wavefronts are isolated. A correlation technique is used to compute the similarity between all possible pairs of the isolated wavefronts. From these data, the wavefronts are sorted into clusters, each of which represents a recurring wavefront morphology. We define multiplicity (M) as the number of clusters needed to account for 90% of the total activations in the VF episode.M measures the complexity of the rhythm. In repetitive patterns (e.g., sinus rhythm),M=1, indicating that the same morphology repeatedly activates the mapped region. Typically, in VF,M>1, with larger numbers representing more complex, disorganized patterns. As an example, we computedM at 5, 10, 15, and 20 sec after electrical induction of VF in six pigs.M decreased significantly (p<0.001), suggesting increasing organization during this period.     

基于 Shapiro 模型推理算法的智能知识获取在中医知识库中的实现

本文应用现代计算机技术,将Shapiro模型推理算法运用于中医知识库中,使该知识库系统具有智能知识获取的能力.     

Pediatric Femoral Shaft Fracture: An Age-Based Treatment Algorithm

PurposeFractures of the femoral shaft in children are common. The rates of bone growth and remodeling in children vary according to their ages, which affect their respective management. MethodsThis paper evaluates the incidence and patterns of pediatric femoral shaft fracture and the current concepts of treatments available.ResultsThe type of fracture—closed or open; stable or unstable—needs to be taken into account. Child abuse should be suspected in fractures sustained by infants. For younger children, non-surgical management is preferred, which include Pavlik harness (< 6 months old) and early spica casting (6 months to 6 years old). Older children (> 6 years old) usually benefit from surgical treatments as outcomes of non-surgical alternatives are worse and are associated with prolonged recovery times. These operative measures for older children that are 6–12 years old include elastic stable intramedullary nailing and submuscular plating. Factors to be considered when devising an appropriate intervention include body mass, location of injury, and nature of fracture. For adolescent and skeletally mature teenagers (> 12 years old), rigid antegrade entry intramedullary fixation is indicated. In the event of open fractures or polytrauma, external fixation should be considered as a temporary treatment method for initial fracture stabilization.ConclusionAn age-based and evidence-based algorithm has been proposed to guide surgeons in the process of evaluating an appropriate treatment.     

利用独立准直器开展调强放疗算法研究

研究利用独立准直器调整射野强度分布。设计了两种递推的办法来计算射野处序列,采用模拟退火算法优化射野片的照射顺序。并以在Ｖａｒｉａｎ ６００Ｃ轲速器上实施三个临床调强射野为例,估计ＩＣ调强放疗的照射时间。照射时间一般在５至１０ｍｉｎ。它近似与机器剂量率成反比,并受调强算法,矩阵元素大小以及强度分级数的影响。     

基于聚类分析的径向基神经网络用于证候诊断的研究

目的:优化中医证候诊断模型,为中医证候诊断标准的研究提供可行性方法.方法:提出用于中医证候诊断的径向基(Radial Basis Function,RBF)神经网络,利用聚类分析确定RBF神经网络隐层的参数,运用最小二乘确定RBF神经网络输出层的参数.结果:通过模型检验,证候诊断模型判准率比BP网络模型判准率高;证候诊断模型训练速度比BP网络模型快.结论:基于聚类分析的RBF神经网络用于中医证候诊断的研究是可行的和有效的.     

Frontobasilar fractures: Proposal for image reviewing algorithm

ObjectiveThe aim of this study was to develop and test the utility of a novel systematic protocol to analyze CT images of patients with trauma in the anterior cranial base and upper midface.Material and MethodsThe radiological data and primary reports of 27 consecutive patients with a frontal skull base fracture treated in two tertiary care hospitals from 2007 to 2011 were scrutinized. A novel algorithm for systematic image reviewing was used to assess the CT images and the findings were compared with the primary radiological reports.ResultsThe systematic review detected a substantial number of fractures and defects in anatomical structures that had not been systematically reported in the primary, on-call reports. Anterior skull base fracture was not initially reported in 32% of the patients; however, the algorithm detected this in 93% of them. The corresponding rates for fracture through cribriform plate were 28% and 72% and for fracture through the sella or hypophyseal area 22% and 78%. There were two fractures of the clivus and these were initially missed.ConclusionsDespite the failure to identify these fractures radiologically in the primary setting, all patients were still considered to have received appropriate treatment, but, the use of an image-reviewing algorithm will enhance the specificity of CT in the diagnosis of frontobasilar fractures.