基于空间近邻关系的非平衡数据重采样算法

为了提高非平衡数据集的分类精度,提出了一种基于样本空间近邻关系的重采样算法。该方法首先根据数据集中少数类样本的空间近邻关系进行安全级别评估,根据安全级别有指导的采用合成少数类过采样技术（Synthetic minority oversampling technique,SMOTE）进行升采样;然后对多数类样本依据其空间近邻关系计算局部密度,从而对多数类样本密集区域进行降采样处理。通过以上两种手段可以均衡测试数据集,并控制数据规模防止过拟合,实现对两类样本分类的均衡化。采用十折交叉验证的方式产生训练集和测试集,在对训练集重采样之后,以核超限学习机作为分类器进行训练,并在测试集上进行验证。在UCI非平衡数据集和电路故障诊断实测数据上的实验结果表明,所提方法在整体上优于其他重采样算法。      

ROC comparison of visualization of hand fractures using digital and conventional techniques

PURPOSE: To compare the accuracy of digital luminescence radiography (DLR) and conventional film-screen radiography (FSR) in diagnosing fractures. MATERIAL AND METHOD: Both conventional and digital radiographs were acquired from a consecutive series of 57 patients with suspected wrist or hand fractures. The digital images were obtained with a 30% dose reduction. A ROC-analysis (receiver-operating characteristics) was performed. RESULTS: The area under the curve was 0.89 for conventional FSR, 0.93 for DLR, "gray scale" and 0.94 for DLR, "edge enhanced". CONCLUSIONS: Although its spatial resolution is lower, DLR provided better results than conventional FSR, when contrast processing algorithms were optimised for the specific clinical question. The edge-enhanced version was superior to the non-edge enhanced version. The reason for this seems to be the higher contrast resolution of DLR compared to FSR.     

Diagnostic accuracy evaluation using ROC curve analysis

The diagnostic accuracy of a biochemical quantity is inversely related to the overlapping zone between the values of the population suffering from a disease and the population which does not. The ROC curves are an indirect measure of the overlapping zone between both populations. Specimens (plasma and urine) taken from 928 patients with symptoms of acute abdominal pain were used and the catalytic concentration of alpha-amylase, pancreatic alpha-amylase and triacylglycerol lipase (determined by two methods) were measured. Definitive diagnosis was obtained by following the directives of expert groups on the evaluation of diagnostic tests. Diagnostic accuracy was characterized by calculating the diagnostic sensitivity and specificity, by representing the ROC curves and by quantifying the areas under the ROC curves. The catalytic concentration of pancreatic alpha-amylase in plasma was the quantity with a greater area under the ROC curve (A = 0.9740) and then the one which had greatest diagnostic accuracy. If we considered the upper limit of the reference interval to be the cut-off value, the catalytic concentration of pancreatic alpha-amylase in plasma had a diagnostic sensitivity and specificity values of 0.96 and 0.88 respectively for the acute pancreatitis.     

Regularities of source recognition: ROC analysis.

Source memory has become the focus of a growing number of investigations in a variety of fields. An appropriate model for source memory is, therefore, of increasing importance. A simple 2-dimensional signal-detection model of source recognition is presented. The receiver operating characteristics (ROCs) obtained from 3 experiments are then used to test the model. The data demonstrate 3 regularities: convex ROCs, z-ROCs with linear slopes of 1.00, and slightly concave z-ROCs. Two of these regularities support the model. The 3rd requires a revision of the model. This revised model is fitted to the data. The implications of these regularities for other theories are also discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Resampling and cross-validation techniques: a tool to reduce bias caused by model building?

The process of model building involved in the analysis of many medical studies may lead to a considerable amount of over-optimism with respect to the predictive ability of the 'final' regression model. In this paper we illustrate this phenomenon in a simple cutpoint model and explore to what extent bias can be reduced by using cross-validation and bootstrap resampling. These computer intensive methods are compared to an ad hoc approach and to a heuristic method. Besides illustrating all proposals with the data from a breast cancer study we perform a simulation study in order to assess the quality of the methods.     

The use of the 'binormal' model for parametric ROC analysis of quantitative diagnostic tests

The binormal model is widely used for parametric receiver operating characteristic (ROC) analyses of data concerning the accuracy of medical diagnostic tests. Empirical evaluation of the performance of this model in the face of departures from binormality has been limited to interpretations of radiology-type examinations recorded on a rating scale. This paper extends the investigation to the performance of the model with biochemical and other tests recorded on an interval scale. In order to describe non-binormal pairs of distributions, a useful standardized graphical display is developed; this display also illustrates several features of ROC curves. We consider non-binormal pairs of distributions with or without a monotone likelihood ratio and show that by transformation of the underlying scale, one can make many such pairs resemble closely the binormal model. These findings justify Metz's use of the binormal model in the 'LABROC' software for ROC analyses of laboratory type data even when the raw data may 'look' decidedly non-Gaussian.     

Linguistic techniques and the analysis of emotionality in interviews.

In an attempt to analyze the degree of emotionality in the manner (as opposed to the content) of speech, excerpts from 6 interviews and a recording of a radio program were analyzed with regard to linguistic (junctures, stress, pitch) and paralinguistic phenomena (e.g., characteristic of vocalizations and voice quality). The therapeutic interviews were distinguished from the radio program conversation on the basis of linguistic phenomena only. However, linguistic phenomena were seen to be independent of emotional expressions contentwise. The results were interpreted in terms of the expression of emotion through content nonlanguage vocalizations. From Psyc Abstracts 36:04:4IF01D. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Evaluation of confounding effects in ROC studies

In many clinical studies, it is clear that external forces can affect the performance of diagnostic tests, as these factors influence the distributions of separator variables. A new estimator for the receiver operating characteristic (ROC) function is proposed; this estimator converges to the ROC function uniformly on the interval [0,1]. Using this new estimator, the author proposes to use Cox's proportional hazards regression model for the evaluation of confounding effects in ROC studies. The method can be used even when concomitant information is only available for the cases, for example, disease severity. A textbook example on prostate cancer is described for illustration.     

Application of probability techniques to the objective interpretation of veterinary clinical biochemistry data

Methods for the interpretation of veterinary clinical biochemistry have not developed as rapidly as biochemical technology. However, the results of clinical biochemistry tests are only of value when they are interpreted appropriately. A retrospective study was undertaken to investigate the equine biochemistry data which had been stored in a veterinary hospital database. By applying percentile analysis and Bayesian probability methods to the clinical biochemistry and corresponding diagnosis data, a novel method for the interpretation of clinical biochemistry data has been developed. The method allows clinicians to determine whether a biochemistry value is abnormal, its degree of abnormality, and the most likely associated diagnoses. The method could be used to investigate a practice-based population and may have significant implications for the interpretation of clinical biochemistry data in veterinary medicine in the future.     

Observer performance in detecting multiple radiographic signals. Prediction and analysis using a generalized ROC approach

This inquire demonstrated that transfusion malaria was not so rare as we thought. So 56 cases have been numbered between 1960 and 1974, and probably they do not reflect exactly the real situation. In our country, plasmodium falciparum, specifically responsible of the most serious accidents, is the form most frequently incriminated and its frequency increases among the years: 76,1% of the infects between 1970 and 1974. At last we have confirmed that a prevent of those accidents must consider as a subject any one having resided in an impaluded country, and that implies a constant knowledge of the epidemiologic map of this parisitism. This prevent will be increased so that the transfusion malaria will not get a illness disease in France.     

Integrative data analysis: The simultaneous analysis of multiple data sets.

There are both quantitative and methodological techniques that foster the development and maintenance of a cumulative knowledge base within the psychological sciences. Most noteworthy of these techniques is meta-analysis, which allows for the synthesis of summary statistics drawn from multiple studies when the original data are not available. However, when the original data can be obtained from multiple studies, many advantages stem from the statistical analysis of the pooled data. The authors define integrative data analysis (IDA) as the analysis of multiple data sets that have been pooled into one. Although variants of IDA have been incorporated into other scientific disciplines, the use of these techniques is much less evident in psychology. In this article the authors present an overview of IDA as it may be applied within the psychological sciences, discuss the relative advantages and disadvantages of IDA, describe analytic strategies for analyzing pooled individual data, and offer recommendations for the use of IDA in practice. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of aging on urinary secretion of 3-hydroxyproline and its importance for cancer screening in the elderly with ROC analysis

OBJECTIVE: to observe the influence of aging on urinary secretion of 3-hydroxyproline (3-Hyp) in normal subjects and in cancer patients and study the effects of aging on the accuracy of urinary 3-Hyp as a general-purpose cancer screening test. DESIGN: We reanalyzed, from the aspect of gerontology, the amounts of 3-Hyp in the urine samples of the 211 healthy persons and 94 patients with cancer reported earlier. The two groups were divided in 3 age subgroups: adult, 30 to 44 years, middle-aged, 45 to 64 years, and aged, 65 years and older. The cancer stage was separated into 3 categories: stage 1, the early stage without any invasion, stage 2, locally invasive with no distant metastasis, and stage 3, advanced cancer. The correlation of age and urinary 3-Hyp was analyzed in normal subjects and cancer patients. Sensitivity and specificity were calculated for different cutoff points in the diverse groups to construct the receiver operating characteristic (ROC) curves. MAIN RESULTS: We observed that aging decreases the urinary secretion of 3-Hyp in normal subjects, with a correlation coefficient of -0.250 (p < 0.001). The mean levels of urinary 3-Hyp in patients with cancer were significantly higher than in the normal subjects, p = 0.005. The correlation coefficient of urinary 3-Hyp and age was not significant (0.100) in cancer patients. The cutoff points of 0.800 mg/g creatinine of urinary 3-Hyp for persons under 65 years of age and 0.600 mg/g creatinine for persons 65 years or older were the best thresholds for cancer screening. In the ROC analyses, we observed that 3-Hyp has higher accuracy for cancer screening in the aged group for all stages together and in the aged and middle-aged for early stages of cancer. CONCLUSIONS: The urinary 3-Hyp test was more effective for cancer screening in old persons (65 years or older), where cancer is more frequent and more difficult to discover. The authors stress the necessity and importance of verifying the effectiveness of urinary 3-Hyp for general cancer screening in a larger population and in a community set.     

The role of smoothing techniques in the interpretation of results from genomic scans using sib-pair data

BACKGROUND: Budesonide, a corticosteroid with high topical anti-inflammatory activity and low systemic activity, has been shown to prolong time to relapse in Crohn's disease. In the present study, the efficacy of budesonide in an oral pH-modified-release formulation was evaluated for maintenance treatment in patients with steroid-dependent ulcerative colitis. METHODS: Fourteen patients with steroid-dependent ulcerative colitis in the reduction phase of conventional glucocorticosteroids (c-GCS) following a severe attack, were treated with budesonide 3 mg t.d.s. for 6 months. The primary investigation parameters were changes in the clinical activity index (CAI) and in the daily dose of c-GCS. RESULTS: In 11 cases the CAI improved significantly and treatment with c-GCS could be terminated. Three patients experienced relapse and needed further c-GCS treatment. The average daily dose of c-GCS and the average value of the CAI before treatment with budesonide were significantly higher in the relapse group than in the remission group. CONCLUSIONS: In patients with c-GCS-dependent ulcerative colitis, a dose of 9 mg budesonide daily in an oral pH-modified-release formulation was well tolerated, significantly decreased the CAI, and rendered c-GCS unnecessary in the majority of cases.     

包钢营销分析数据仓库数据组织模式的研究

以包钢营销分析数据仓库的构建为背景,对其主题确定、数据结构、数据处理技术、元数据管理提出了解决方案.