模糊神经网络在颅脑磁共振图像分割中的应用研究

由于颅脑磁共振图像分割的效果受制于解剖结构和成像过程引起的边缘模糊和噪声，本研究结合神经网络和模糊逻辑技术，提出一种基于模糊神经网络的颅脑磁共振图像的半自动分割算法。分割实验的结果表明，在同等条件下，模糊神经网络分割算法的收敛速度比BP神经网络分割算法快三倍以上；与最大似然法、模糊c均值聚类和BP神经网络分割算法相比，FNN分割算法的抗噪声和抗模糊能力更强。     

原发性肺癌孤立性结节的自动提取

研究自动分割和提取原发性肺癌肺部孤立性结节(SPN)特征的方法。对CT图像进行预处理后，首先分割出肺实质，然后用模糊C均值聚类方法对肺实质图像作进一步地细分割，提取感兴趣区域(ROI)，最后根据分形理论计算出分形维数结合灰度方差供分类判决。结果表明此方法能够有效地自动识别SPN。     

哲学视野下的医疗市场细分

满足患者需要是现代医院营销的核心。从哲学视角看,患者需求异质性的客观存在是进行医疗市场细分的唯物论基础,而患者需求普遍性和特殊性的辩证统一则是其辩证法基础。     

L —a new software system for the evaluation of functional magnetic resonance images of the human brain

This paper describes the non-commercial software system L that was developed for the processing of functional magnetic resonance images (fMRI) of the human brain. The analysis of fMRI data comprises various aspects including filtering, spatial transformation, statistical evaluation as well as segmentation and visualization. In L , particular emphasis was placed on the development of new visualization and segmentation techniques that support visualizations of individual brain anatomy so that experts can assess the exact location of activation patterns in individual brains. As the amount of data that must be handled is enormous, another important aspect in the development L was the efficiency of the software implementation. Well established statistical techniques were used whenever possible.     

新型农村合作医疗应用分段定额付费方式的实证研究

目的 控制医疗费用过快增长,探索建立适合新型农村合作医疗运行体制的按病种付费方式.方法 以安徽省新型农村合作医疗试点县为现场.选取5病种3年全部资料进行实证研究,并对5病种就诊总费用中进入新型农村合作医疗支付范围的费用研究进行校正和定量分析.结果按病种"分段定额"付费的制度体系包括病种的选择、付费标准的测算、结算模式、补偿方式、监督控制措施五大方面.结论 按病种"分段定额"付费是进行新型农村合作医疗医疗费用控制的有效模式,但其作用的发挥需要单病种临床路径的制定等五项配套措施的完善.     

Vertebral body segmentation with prior shape constraints for accurate BMD measurements

We propose a novel vertebral body segmentation approach, which is based on the graph cuts technique with shape constraints. The proposed approach depends on both image appearance and shape information. Shape information is gathered from a set of training shapes. Then we estimate the shape variations using a new distance probabilistic model which approximates the marginal densities of the vertebral body and its background in the variability region using a Poisson distribution refined by positive and negative Gaussian components. To segment a vertebral body, we align its 3D shape with the training 3D shape so we can use the distance probabilistic model. Then its gray level is approximated with a Linear Combination of Gaussians (LCG) with sign-alternate components. The spatial interaction between the neighboring voxels is identified using a new analytical approach. Finally, we formulate an energy function using both appearance models and shape constraints. This function is globally minimized using s/t graph cuts to get the optimal segmentation. Experimental results show that the proposed technique gives promising results compared to other alternatives. Applications on Bone Mineral Density (BMD) measurements of vertebral body are given to illustrate the accuracy of the proposed segmentation approach.     

Interactive-cut: Real-time feedback segmentation for translational research

In this contribution, a scale-invariant image segmentation algorithm is introduced that “wraps” the algorithm's parameters for the user by its interactive behavior, avoiding the definition of “arbitrary” numbers that the user cannot really understand. Therefore, we designed a specific graph-based segmentation method that only requires a single seed-point inside the target-structure from the user and is thus particularly suitable for immediate processing and interactive, real-time adjustments by the user. In addition, color or gray value information that is needed for the approach can be automatically extracted around the user-defined seed point. Furthermore, the graph is constructed in such a way, so that a polynomial-time mincut computation can provide the segmentation result within a second on an up-to-date computer. The algorithm presented here has been evaluated with fixed seed points on 2D and 3D medical image data, such as brain tumors, cerebral aneurysms and vertebral bodies. Direct comparison of the obtained automatic segmentation results with costlier, manual slice-by-slice segmentations performed by trained physicians, suggest a strong medical relevance of this interactive approach.     

Statistical segmentation of surgical instruments in 3-D ultrasound images

The recent development of real-time 3-D ultrasound (US) enables intracardiac beating-heart procedures, but the distorted appearance of surgical instruments is a major challenge to surgeons. In addition, tissue and instruments have similar gray levels in US images and the interface between instruments and tissue is poorly defined. We present an algorithm that automatically estimates instrument location in intracardiac procedures. Expert-segmented images are used to initialize the statistical distributions of blood, tissue and instruments. Voxels are labeled through an iterative expectation-maximization algorithm using information from the neighboring voxels through a smoothing kernel. Once the three classes of voxels are separated, additional neighboring information is combined with the known shape characteristics of instruments to correct for misclassifications. We analyze the major axis of segmented data through their principal components and refine the results by a watershed transform, which corrects the results at the contact between instrument and tissue. We present results on 3-D in-vitro data from a tank trial and 3-D in-vivo data from cardiac interventions on porcine beating hearts, using instruments of four types of materials. The comparison of algorithm results to expert-annotated images shows the correct segmentation and position of the instrument shaft.