检索结果-内蒙古大学图书馆

Random logistic machine (RLM): Transforming statistical models into machine learning approach

Communications in Statistics - Theory and Methods 2024年第21期53卷 7517-7525页

作者： Li, Yu-Shan Guo, Chao-Yu Division of Biostatistics and Data Science Institute of Public Health College of Medicine National Yang Ming Chiao Tung University Taipei Taiwan

data science is booming with big data, and machine learning provides better predictive analyses. However, conventional statistical models can effortlessly interpret the effect estimates, and the prediction models are generally in a closed form. Therefore, this research integrates the logistic regression model’s core with the Random Forest structure to create a blended novel machine learning method, the Random Logistic Machine (RLM). In this way, the new approach preserves the statistical and machine learning advantages. Computer simulation studies examined the predictive ability of RLM, random forest, and Logistic Regression under various scenarios. The results showed that the RLM delivers a comparable performance to Random Forests and Logistic Regression. An application to the Breast Cancer Wisconsin (Diagnostic) data Set also demonstrates the superior performance of the new approach. © 2023 The Author(s). Published with license by Taylor & Francis Group, LLC.

关键词： Logistic regression

来源：评论

学校读者我要写书评

暂无评论

Construction of Optimal Mixed-Level Uniform Designs

引用

Journal of Systems science & Complexity 2024年第2期37卷 841-862页

作者： CHATTERJEE Kashinath LIU Min-Qian QIN Hong YANG Liuqing Department of Population Health Sciences Division of Biostatistics and Data ScienceAugusta UniversityGA 30912USA NITFID LPMC&KLMDASRSchool of Statistics and Data ScienceNankai UniversityTianjin 300071China School of Statistics and Mathematics Zhongnan University of Economics and LawWuhan 430073China

The theory of uniform design has received increasing interest because of its wide application in the field of computer *** generalized discrete discrepancy is proposed to evaluate the uniformity of the mixed-level factorial *** this paper,the authors give a lower bound of the generalized discrete discrepancy and provide some construction methods of optimal mixed-level uniform designs which can achieve this lower *** methods are all deterministic construction methods which can avoid the complexity of stochastic *** saturated mixed-level uniform designs and supersaturated mixed-level uniform designs can be obtained with these ***,the resulting designs are also χ^(2)-optimal and minimum moment aberration designs.

关键词： Generalized discrete discrepancy Hadamard matrix mixed-level design orthogonal array supersaturated design

来源：评论

学校读者我要写书评

暂无评论

Generalized Estimating Equations Boosting (GEEB) machine for correlated data

引用

Journal of Big data 2024年第1期11卷 20页

作者： Wang, Yuan-Wey Yang, Hsin-Chou Chen, Yi-Hau Guo, Chao-Yu Division of Biostatistics and Data Science Institute of Public Health College of Medicine National Yang Ming Chiao Tung University Taipei Taiwan Institute of Statistical Science Academia Sinica Taipei Taiwan

Rapid development in data science enables machine learning and artificial intelligence to be the most popular research tools across various disciplines. While numerous articles have shown decent predictive ability, little research has examined the impact of complex correlated data. We aim to develop a more accurate model under repeated measures or hierarchical data structures. Therefore, this study proposes a novel algorithm, the Generalized Estimating Equations Boosting (GEEB) machine, to integrate the gradient boosting technique into the benchmark statistical approach that deals with the correlated data, the generalized Estimating Equations (GEE). Unlike the previous gradient boosting utilizing all input features, we randomly select some input features when building the model to reduce predictive errors. The simulation study evaluates the predictive performance of the GEEB, GEE, eXtreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM) across several hierarchical structures with different sample sizes. Results suggest that the new strategy GEEB outperforms the GEE and demonstrates superior predictive accuracy than the SVM and XGBoost in most situations. An application to a real-world dataset, the Forest Fire data, also revealed that the GEEB reduced mean squared errors by 4.5% to 25% compared to GEE, XGBoost, and SVM. This research also provides a freely available R function that could implement the GEEB machine effortlessly for longitudinal or hierarchical data. © 2024, The Author(s).

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Random Interaction Forest (RIF)-A Novel Machine Learning Strategy Accounting for Feature Interaction

引用

IEEE Access 2023年 11卷 1806-1813页

作者： Guo, Chao-Yu Lin, Yi-Jyun National Yang Ming Chiao Tung University Division of Biostatistics and Data Science Institute of Public Health College of Medicine Taipei112304 Taiwan

If an interaction exists in medical and health sciences, a proper statistical approach is required to avoid an erroneous conclusion. For example, different genders may introduce modified therapeutic effects of drugs, or an adverse interaction between two medicines changes the pharmacological activity, reduces the therapeutic effect, or induces toxicity. Therefore, if the analysis does not account for the impact of the interaction, it may introduce significant prediction errors or bias. Regression models deal with a two-way interaction by adding the product of the two interactive variables. Since machine learning models demonstrate a superior predictive ability to regression models, this study proposes a new method based on the random forest to account for interaction, called random interaction forest (RIF). This new strategy modifies the structure of the random forest, where the interaction features are forced to be in the first two nodes. Simulation studies examined the predictive ability of the linear regression model, logistic regression model, random forest, and the RIF under various scenarios. The results showed that the RIF consistently outperforms random forest and logistic regression when interactions are present. The RIF also performs better in many scenarios than the linear regression model. When the effect of interaction is more significant, the performance of RIF could be superior. © 2013 IEEE.

关键词： Linear regression

来源：评论

学校读者我要写书评

暂无评论

Characterization of a new fluorescence lifetime imaging ophthalmoscope

引用

ACTA OPHTHALMOLOGICA 2024年第S279期102卷

作者： Nycz, Julia Link, Dietmar Klemm, Matthias Klee, Sascha Haueisen, Jens Institute of Biomedical Engineering and Informatics Technische Universität Ilmenau Ilmenau Germany Division Biostatistics and Data Science Karl Landsteiner University of Health Sciences Krems Austria

Aims/Purpose: Fluorescence lifetime imaging ophthalmoscopy (FLIO) allows in vivo measurement of autofluorescence intensity decays of endogenous fluorophores in the ocular fundus. So far, only devices from Heidelberg Engineering based on the Spectralis system have been used in FLIO research. Here, we present and characterize a new FLIO device based on the RETImap system from Roland Consult. Methods: The device is based on a confocal scanning laser ophthalmoscope (35° field, 512 × 512 px). A ps diode laser (BDL-SMN 473 nm, Becker & Hickl GmbH, Berlin, Germany) excites autofluorescence. The fluorescence photons are split into a short (498–560 nm, SSC) and a long (560–720 nm, LSC) spectral channel (one HPM-100-40 detector [Becker & Hickl GmbH] each) and are detected by time-correlated single photon counting (SPC-160, Becker & Hickl GmbH). We determined the maximum laser power (ILT2400, International Light Technologies, Inc. Peabody, MA, USA) and analysed the instrument's behaviour at three different laser power levels (150 μW, 200 μW and max.) in terms of laser spectrum (CAS140CT, Instrument Systems GmbH, Munich, Germany) and instrument response function (IRF). The IRF was determined using a 25 μM Eosin Y solution, mixed with a 5 M solution of potassium iodide, placed in a flat cuvette (110-OS, Hellma GmbH & Co. KG, Müllheim, Germany) in front of the objective lens of the FLIO device. Fluorescence measurements of approximately 1-min duration were performed three times for all three laser powers. The IRF and the full width at half maximum (FWHM) were calculated using FLIMX software ( *** ). Results: The max. laser power was 280 μW. The peak wavelengths of the laser spectra were 467.6 (150 μW), 467.9 (200 μW) and 468.0 nm (280 μW). IRF FWHM in SSC were 298.6 ± 1.1 ps (150 μW), 341.0 ± 2.5 ps (200 μW) and 347.5 ± 6.0 ps (280 μW). In LSC, the IRF FWHM were 290.4 ± 3.8 ps (150 μW), 344.0 ± 3.4 ps (200 μW) and 358.8 ± 1.3 ps (280 μW). Results are mean ± standard devi

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic retinal vessel analysis with flicker light stimulation: Higher stimulation frequencies lead to higher vascular responses

引用

ACTA OPHTHALMOLOGICA 2025年第S284期103卷

作者： Link, Dietmar Freitag, Stefanie Klee, Sascha Department of Optoelectrophysiological Engineering Technische Universität Ilmenau Ilmenau Germany Department of General Health Studies Division Biostatistics and Data Science Karl Landsteiner University of Health Sciences Krems Austria

Aims/Purpose: Dynamic retinal vessel analysis (DVA) with flicker light is an established method of retinal vessel analysis. Stimulation is generally performed according to a standard protocol at 12.5 Hz. Technologically, a full-field shutter is used, which is located in the illumination path. In 2002, Polak et al. investigated the flicker-induced vascular response at different frequencies (2–64Hz). This showed that from frequencies > 4Hz, smaller responses tend to result. However, current developments in light stimulation are moving towards spatio-temporal stimulators. We therefore used a spatial light modulator (SLM) in a DVA setup. In this work, we investigated vascular responses at a lower and higher stimulation frequency compared to 12.5Hz. The dilation maxima were determined and compared. Methods: We studied 12 subjects (7w, 5m, 26.6 ± 3.7yr, 1 eye). We measured two primary vessels each: a superior temporal artery and a vein. All subjects were free of ocular and systemic diseases. A high-power LED (515nm) and a transmissive liquid crystal display (Sony LCX017AL) were connected to a mydriatic fundus camera. The SLM was controlled via standard graphics card and a customized software. The field of view was set to 30°. Each subject was measured two times. One measurement was carried out at 3.75Hz and one at 30Hz flicker frequency, randomized among the subjects. As in Polak et al. (2002), measurements were taken with 60s baseline, 60s flicker stimulation. Decay measurement was 120s. The pause time between two measurements was 15min. The mean value and standard deviation were calculated for each stimulation frequency and vessel type. The differences were tested for normal distribution and either the paired t-test or the Wilcoxon signed-rank test was performed. Results: All subjects were able to successfully complete the study ( n = 12). Arteries showed a mean maximum dilatation of 2.8 ± 1.2% at 3.75Hz and 4.0 ± 1.6% at 30Hz ( p = 0.023, Cohen's d = 0.765). Veins show

关键词：

来源：评论

学校读者我要写书评

暂无评论

The polarity effect of direct current stimulation on the vascular regulation of the inner retina

引用

ACTA OPHTHALMOLOGICA 2025年第S284期103卷

作者： Freitag, Stefanie Link, Dietmar Klee, Sascha Department of Optoelectrophysiological Engineering Technische Universität Ilmenau Ilmenau Germany Division Biostatistics and Data Science Department of General Health Studies Karl Landsteiner University of Health Sciences Krems Austria

Aims/Purpose: Electrical stimulation is not only a suitable method for researching the human nervous system but also for treating neuronal disorders. In ophthalmology, it is applied to patients suffering from various retinal diseases like retinitis pigmentosa. It is known that in the application of direct current stimulation (DCS), the polarity of the current determines the increase or decrease in neuronal activity and enables different therapeutic approaches. However, the polarisation-dependent effect of the current on the vascular processes in the retina has not yet been investigated. This study examined the effect of anodal and cathodal DCS on the temporal vessels of the inner retina. Methods: The study was conducted on 24 healthy subjects (14w, 10m, mean age 25.5 ± 3.2yr) who were randomly assigned to either an anodal or a cathodal DCS group (symmetrical design). To test the effect of the DCS (800μA; ring-shaped electrode placed surrounding the eye, square electrode placed at the occiput) on vascular regulation, the DCS was combined with visual flicker stimulation using the Dynamic Vessel Analyzer (DVA, Imedos Systems GmbH; standard protocol). Exclusion criteria were ocular and systemic diseases. We investigated two primary vessel segments (superior temporal artery (STa) and vein (STv)) located between 0.5 and 2.0 disc diameters from the optic disc. For each subject the DVA was performed two times (first baseline DVA1, followed by a 20-minute DCS (different polarity depending on the group), and directly afterwards a second DVA2). For statistical analysis, the t-test for paired samples was used and Bonferroni corrected. Results: The mean vascular responses (STa/STv) for the anodal DCS group were DVA1: 3.7 ± 1.7%/3.4 ± 1.6%; DVA2: 3.5 ± 1.3%/5.4 ± 3.2% and for the cathodal DCS group DVA1: 3.2 ± 2.0%/4.0 ± 2.6%; DVA2: 3.2 ± 1.4%/2.5 ± 1.3%. For the venous vascular response, we found a significant increase after anodal DCS ( p = 0.019) and a significant decrease aft

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic Subgroup Identification in Covariate-adjusted Response-adaptive Randomization Experiments 38

Dynamic Subgroup Identification in Covariate-adjusted Respon...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Li, Yanping Wang, Jingshen Wei, Waverly School of Statistics and Data Science Nankai University China Division of Biostatistics University of California Berkeley United States Department of Data Sciences and Operations University of Southern California United States

Identifying subgroups with differential responses to treatment is pivotal in randomized clinical trials, as tailoring treatments to specific subgroups can advance personalized medicine. Upon trial completion, identifying best-performing subgroups-those with the most beneficial treatment effects-is crucial for optimizing resource allocation or mitigating adverse treatment effects. However, traditional clinical trials are not customized for the goal of identifying best-performing subgroups because they typically pre-define subgroups at the beginning of the trial and adhere to a fixed subgroup treatment allocation rule, leading to inefficient use of experimental efforts. While some adaptive experimental strategies exist for the identification of the single best subgroup, they commonly do not enable the identification of the best set of subgroups. To address these challenges, we propose a dynamic subgroup identification covariate-adjusted response-adaptive randomization (CARA) design strategy with the following key features: (i) Our approach is an adaptive experimental strategy that allows the dynamic identification of the best subgroups and the revision of treatment allocation towards the goal of correctly identifying the best subgroups based on collected experimental data. (ii) Our design handles ties between subgroups effectively, merging those with similar treatment effects to maximize experimental efficiency. In the theoretical investigations, we demonstrate that our design has a higher probability of correctly identifying the best set of subgroups compared to conventional designs. Additionally, we prove the statistical validity of our estimator for the best subgroup treatment effect, demonstrating its asymptotic normality and semiparametric efficiency. Finally, we validate our design using synthetic data from a clinical trial on cirrhosis. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Prediction Inference Using Generalized Functional Mixed Effects Models

arXiv

引用

arXiv 2025年

作者： Zhou, Xinkai Cui, Erjia Sartini, Joseph Crainiceanu, Ciprian Department of Biostatistics Johns Hopkins University United States Division of Biostatistics and Health Data Science University of Minnesota United States

We introduce inferential methods for prediction based on functional random effects in generalized functional mixed effects models. This is similar to the inference for random effects in generalized linear mixed effects models (GLMMs), but for functional instead of scalar outcomes. The method combines: (1) local GLMMs to extract initial estimators of the functional random components on the linear predictor scale;(2) structural functional principal components analysis (SFPCA) for dimension reduction;and (3) global Bayesian multilevel model conditional on the eigenfunctions for inference on the functional random effects. Extensive simulations demonstrate excellent coverage properties of credible intervals for the functional random effects in a variety of scenarios and for different data sizes. To our knowledge, this is the first time such simulations are conducted and reported, likely because prediction inference was not viewed as a priority and existing methods are too slow to calculate coverage. Methods are implemented in a reproducible R package and demonstrated using the NHANES 2011-2014 accelerometry data. © 2025, CC BY-NC-SA.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

BAMITA: Bayesian Multiple Imputation for Tensor Arrays

arXiv

引用

arXiv 2024年

作者： Jiang, Ziren Li, Gen Lock, Eric F. Division of Biostatistics and Health Data Science University of Minnesota United States Department of Biostatistics University of Michigan United States

data increasingly take the form of a multi-way array, or tensor, in several biomedical domains. Such tensors are often incompletely observed. For example, we are motivated by longitudinal microbiome studies in which several timepoints are missing for several subjects. There is a growing literature on missing data imputation for tensors. However, existing methods give a point estimate for missing values without capturing uncertainty. We propose a multiple imputation approach for tensors in a flexible Bayesian framework, that yields realistic simulated values for missing entries and can propagate uncertainty through subsequent analyses. Our model uses efficient and widely applicable conjugate priors for a CANDECOMP/PARAFAC (CP) factorization, with a separable residual covariance structure. This approach is shown to perform well with respect to both imputation accuracy and uncertainty calibration, for scenarios in which either single entries or entire fibers of the tensor are missing. For two microbiome applications, it is shown to accurately capture uncertainty in the full microbiome profile at missing timepoints and used to infer trends in species diversity for the population. Documented R code to perform our multiple imputation approach is available at https://***/lockEF/MultiwayImputation. © 2024, CC BY.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：