A learning method for the class imbalance problem with medical data sets

In medical data sets, data are predominately composed of “normal” samples with only a small percentage of “abnormal” ones, leading to the so-called class imbalance problems. In class imbalance problems, inputting all the data into the classifier to build up the learning model will usually lead a learning bias to the majority class. To deal with this, this paper uses a strategy which over-samples the minority class and under-samples the majority one to balance the data sets. For the majority class, this paper builds up the Gaussian type fuzzy membership function and α-cut to reduce the data size; for the minority class, we use the mega-trend diffusion membership function to generate virtual samples for the class.Furthermore, after balancing the data size of classes, this paper extends the data attribute dimension into a higher dimension space using classification related information to enhance the classification accuracy. Two medical data sets, Pima Indians’ diabetes and the BUPA liver disorders, are employed to illustrate the approach presented in this paper. The results indicate that the proposed method has better classification performance than SVM, C4.5 decision tree and two other studies.     

一种基于粗集理论的遗传分类算法

本文提出了一种基于粗集理论的遗传分类算法,该算法可以无需任何辅助信息,只根据数据自身提供的信息对数据进行简化,提取有用的特征,并求得相应的规则.同时,还提出了一种基于属性重要度的分辨矩阵简化方法,该方法可提高对条件属性的约简效率.     

传统中国“家族医疗空间”的社会文化特质及其现代意义——中西社会结构比较的视阈

传统社会的“医疗空间”，不管是民间的“家族医疗空间”，还是宫廷的“皇家医疗空间”，均是与传统的社会结构相配合的，而且要严格遵照家族系统的伦理原则来建构。只要“高度特殊化”及高度“功能普化”的家族共同体系统无法突破。“高度普遍化”的社会公共空间就不可能生长发育，而孕育于其中的高度“功能专化”的现代医疗空间也就无法形成。从传统到现代的社会转型中，“家族医疗空间”历经了向“单位医疗空间”平稳过渡的转变，然而，以市场化为导向的医疗体制改革，产生了“制度变革”超越“结构转型”的后果。     

一个乡镇范围内卫生服务站点选址问题的数学模型及求解

目的为适应农村卫生体制改革，为农村乡镇卫生服务站点的选址提供依据。方法建立该问题的数学模型，并利用启发式算法求解。结果依据该模型得到：在有13个行政村的一个乡镇可设3个站点。结论既方便了群众就医，又节约了国家资源。     

注意缺陷多动障碍模型大鼠注意定势转移能力的行为学评价

目的了解注意缺陷多动障碍动物模型幼年自发性高血压大鼠(SHR)注意定势转移能力。方法雄性鼠龄21d的SHR(n=6),对照组WKY(n=6)。大鼠适应10d,完成在点食碗中寻找食物作为奖赏物的行为训练。观察指标:潜伏期;大鼠连续6次正确完成参数的循环数和总时间。结果SHR和WKY大鼠仅在Edshift阶段潜伏期分别为(3.5±2.35)s和(13±8.74)s,差异有显著性(P<0.05)。所有阶段SHR能够连续6次正确完成操作任务所需要的循环次数明显多于WKY大鼠,差异有显著性(P<0.05~0.01)。SHR完成SD与Edshift阶段所需要的循环数分别为(17.33±4.84)次和(12.67±3.50)次,两者存在高度相关(r=0.833,P=0.039);WKY完成SD与Edshift阶段所需要的循环数分别为(8.0±1.41)次和(6.67±1.21)次,两者无显著性相关(r=0.234,P=0.656)。SHR仅在CD和CDR阶段所需要的总时间长于WKY大鼠(P<0.05)。结论幼年大鼠可以完成注意定势转移能力的评定,幼年SHR大鼠存在注意定势转移能力缺陷。     

ED-NM-MO三联法对丹参三七配比的多目标优化研究

[目的]采用ED-NM-MO三联法对经基线等比增减设计的丹参、三七不同配比的药效学数据进行非线性拟合和多目标优化。[方法]对经基线等比增减设计的丹参、三七不同配比的药效学数据,以心肌缺血程度Σ-ST、心肌缺血程度(缺血区左室)等7个分别反映心肌缺血、心脏状态和血流动力学的指标作为待优化的药效目标,进行非线性拟合和多目标优化。[结果]分别得到针对7个药效指标和6个药效指标(不包含血清中心肌钙蛋白)的Pareto最优配比。[结论]ED-NM-MO三联法是一种适合复方特点的优化方法,可以应用于由多饮片多组分多成分复方药物的剂量配比优化。     

Frosted cylindrical lens induced artefact on Humphrey automated perimetry

Background: The aim of this study was to assess the effect of a cylindrical corrective lens with frosted sides on Humphrey automated perimetry performed with a translucent spectacle occluder in the contralateral eye. Methods: Ten healthy volunteers with cylindrical refractive errors of more than 1.50 DC had Humphrey visual fields (30‐2) measured with the contralateral eye occluded with either a translucent spectacle occluder or an opaque patch. The order of testing with each occluder was randomised. Additionally, visual field examination with the opaque patch was performed on three volunteer subjects with deep‐set eyes to establish the clinical significance of the frosted cylindrical lens induced artefact (CLA). The back vertex distances for the correcting lenses were measured for all tests. Results: Eight out of 10 volunteers tested with the translucent occluder demonstrated CLA, which was absent when the opaque patch was used. The presence of an artefact corresponding to the axis of the cylinder was observed, when the vertex distance was more than 15 mm. Three subjects with deep‐set eyes presented with CLA, even with the use of an opaque patch. Discussion: Cylindrical lens artefacts can occur with increasing vertex distance when using a translucent spectacle occluder instead of an opaque patch. CLA is also seen in subjects with deep‐set eyes.     

RANDOM SYSTEMS OF HARD PARTICLES:MODELS AND STATISTICS

1　Ｉｎｔｒｏｄｕｃｔｉｏｎ　　Ｉｎｍａｎｙｓｉｔｕａｔｉｏｎｓ ,ｅｎｇｉｎｅｅｒｓａｎｄｓｃｉｅｎｔｉｓｔｓａｒｅｃｏｎ ｆｒｏｎｔｅｄｗｉｔｈｉｒｒｅｇｕｌａｒｓｐａｔｉａｌｓｔｒｕｃｔｕｒｅｓｗｈｉｃｈｃａｎｂｅｉｎ ｔｅｒｐｒｅｔｅｄａｎｄｍｏｄｅｌ     

Voluntary modification of automatic arm movements evoked by motion of a visual target

We have investigated whether the processes underlying the visually evoked, automatic adjustments to a reach are: (1) modifiable by the subject's intention, and (2) available to initiate movement of a stationary arm. Unpredictable movement of a target (80 m/s through 10 cm, left or right in a third of trials) either evoked a mid-flight adjustment of a reaching movement or else acted as a trigger to start an arm movement. Subjects were instructed to respond as rapidly as possible by moving their finger either in the same or in the opposite direction to the target. The target shift evoked an early (125-160 ms) and/or a later (> 160 ms) class of response in the reaching arm. The early response was highly automatic in that it could not be reversed (move opposite) by the subjects' intention. However, the subjects' intention did influence the frequency of occurrence and the size of this early response. The later response was totally modifiable in that it changed direction according to the subjects' intention. Similar classes of response were observed in stationary limbs, but the early, more automatic response was substantially weaker than that elicited during a reach. Two possible mechanisms are proposed to explain these results. The first is a dual-pathway model, which assumes that the two response classes are each generated by separate visuo-motor processes with different properties. The second model assumes both responses are generated by a single visuo-motor mechanism that is under the control of a higher, attentional process.     

医疗设备信息数据集成与应用设计

目的构建以医疗设备个案信息为内容的主题数据集中间库,实现医疗设备由经验型管理向基于网络数据信息的科学管理转变。方法分析医疗设备全寿命周期活动;梳理归纳基于HIS的医疗设备业务信息;抽象主题数据集子集及设置数据项;核查数据质量;构建中间库数据结构表等。结果构建了包括医疗设备基本信息等11个模块为内容的医疗设备主题数据集中间库。满足了各级医疗设备管理者日常管理和重大决策的需求。