A5065624271- Gene expression and cancer classification
- Statistical Mechanics and Entropy
- Advanced Statistical Methods and Models
- Bioinformatics and Genomic Networks
- Prostate Cancer Treatment and Research
- Machine Learning and Data Classification
- Bayesian Methods and Mixture Models
- Prostate Cancer Diagnosis and Treatment
- Statistical Methods and Inference
- Neural Networks and Applications
- Imbalanced Data Classification Techniques
- Genetic Mapping and Diversity in Plants and Animals
- Cerebrovascular and Carotid Artery Diseases
- Face and Expression Recognition
- Bayesian Modeling and Causal Inference
- Marine and fisheries research
- AI in cancer detection
- Optimal Experimental Design Methods
- Intracranial Aneurysms: Treatment and Complications
- Cancer, Lipids, and Metabolism
- Acute Ischemic Stroke Management
- Artificial Intelligence in Healthcare
- Complex Systems and Time Series Analysis
- Wildlife Ecology and Conservation
- Species Distribution and Climate Change
The Institute of Statistical Mathematics
2010-2023
Seikei University
2018-2023
University of Fukui
2014-2018
The receiver operating characteristic (ROC) curve is a fundamental tool to assess the discriminant performance for not only single marker but also score function combining multiple markers. area under ROC (AUC) measures intrinsic ability discriminate between controls and cases. Recently, partial AUC (pAUC) has been paid more attention than AUC, because suitable range of false positive rate can be focused according various clinical situations. However, existing pAUC-based methods handle few...
Summary Binary data are popular in ecological and environmental studies; however, due to various uncertainties complexities present sets, the standard generalized linear model with a binomial error distribution often demonstrates insufficient predictive performance when analysing binary proportional data. To address this difficulty, we propose an asymmetric logistic regression that uses new parameter account for complexity. We observe controls model's asymmetry is important adjusting weights...
We propose a new method for clustering based on local minimization of the gamma-divergence, which we call spontaneous clustering. The greatest advantage proposed is that it automatically detects number clusters adequately reflect data structure. In contrast, existing methods, such as K-means, fuzzy c-means, or model-based need to prescribe clusters. detect all minimum points by define cluster centers. A necessary and sufficient condition gamma-divergence have also derived in simple setting....
We discuss a one-parameter family of generalized cross entropy between two distributions with the power index, called projective entropy. The is essentially reduced to Tsallis if are taken be equal. Statistical and probabilistic properties associated extensively investigated including characterization problem which conditions uniquely determine up index. A close relation Lebesgue space Lp dual Lq explored, in escort distribution associates an interesting property. When we consider maximum...
Pancreatic ductal adenocarcinoma (PDAC) is the most life-threating disease among all digestive system malignancies. We developed a blood mRNA PDAC screening using real-time detection PCR to detect expression of 56 genes, discriminate from noncancer subjects. undertook clinical study assess performance system. collected whole RNA 53 patients, 102 subjects, 22 patients with chronic pancreatitis, and 23 intraductal papillary mucinous neoplasms in per protocol analysis. The sensitivity for...
We discuss a special class of generalized divergence measures by the use generator functions. Any measure in is separated into difference between cross and diagonal entropy. The entropy associates with model maximum distributions; leads to statistical estimation via minimization, for arbitrarily giving model. dualistic relationship minimum explored framework information geometry. distributions characterized be totally geodesic respect linear connection associated divergence. A natural...
While most proposed methods for solving classification problems focus on minimization of the error rate, we are interested in receiver operating characteristic (ROC) curve, which provides more information about performance than rate does. The area under ROC curve (AUC) is a natural measure overall assessment classifier based curve. We discuss class concave functions AUC maximization boosting-type algorithm including RankBoost considered, and Bayesian risk consistency lower bound optimum...
We aimed to analyse clinical and gene expression profiles predict pathologic complete response disease-free survival using two consecutive, prospective, preoperative chemotherapy trial cohorts. Clinicopathological data were evaluated in a cohort from consecutive phase II studies that included patients with stage IIA–IIIC breast cancer of all subtypes. Analysed specimens obtained before chemotherapy, cDNA microarray analyses performed the Affymetrix Gene Chip U133 plus 2.0. Between December...
Independent component analysis (ICA) has been shown to be useful in many applications. However, most ICA methods are sensitive data contamination. In this article we introduce a general minimum U-divergence framework for ICA, which covers some standard as special cases. Within the U-family further focus on γ-divergence due its desirable property of super robustness outliers, gives proposed method γ-ICA. Statistical properties and technical conditions recovery consistency γ-ICA studied....
Reproductive output is one of the central attributes life history, and knowledge age-specific reproduction can enhance understanding population performance dynamics. Tardigrades are microscopic invertebrates that live in marine, freshwater terrestrial ecosystems. While changes fertility relation to age known occur other invertebrate groups, subject has not been specifically addressed tardigrades. The current study demonstrates for first time effect lifespan on reproductive characteristics...
Linear scores are widely used to predict dichotomous outcomes in biomedical studies because of their learnability and understandability. Such approaches, however, cannot be elucidate biodiversity when there is heterogeneous structure target population. Our study was focused on describing intrinsic heterogeneity predictions. Because can captured by a clustering method, integrating different information from clusters should yield better Accordingly, we developed quasi-linear score, which...
This paper discusses recent developments for pattern recognition focusing on boosting approach in machine learning. The statistical properties such as Bayes risk consistency several loss functions are discussed a probabilistic framework. There number of proposed different purposes and targets. A unified derivation is given by generator function U which naturally defines entropy, divergence function. class U-loss associates with the learning algorithms minimization, includes AdaBoost...
Summary In the classic discriminant model of two multivariate normal distributions with equal variance matrices, linear function is optimal both in terms log likelihood ratio and maximizing standardized difference (the t-statistic) between means distributions. a typical case–control study, normality may be sensible for control sample but heterogeneity uncertainty diagnosis suggest that more flexible needed cases. We generalize t-statistic approach by finding which maximizes data from one...
In genome-wide interaction studies, to detect gene-gene interactions, most methods are divided into two folds: single nucleotide polymorphisms (SNP) based and gene-based methods. Basically, the on gene more effective than a SNP. Recent years, kernel canonical correlation analysis (Classical CCA) U statistic (KCCU) has been proposed nonlinear relationship between genes. To estimate variance in KCCU, they have used resampling which highly computationally intensive. addition, classical CCA is...
The role of visual evoked potential (VEP) in direct clipping the paraclinoid internal carotid artery (ICA) aneurysm remains uncertain.To examine whether intraoperative neuromonitoring with VEP can predict deterioration function after ICA anterior clinoidectomy.Among consecutive 274 patients unruptured cerebral aneurysm, we enrolled 25 treated by clinoidectomy this study. We evaluated acuity loss (VAL) and field (VFL) before surgery, 1 month at final follow-up.The VAL surgery (VAL1M)...
Abstract In modeling biological and ecological processes from data, it is essential to deal with data selection bias properly in order obtain reliable reasonable predictions. To incorporate the mechanism of into a statistical analysis, propensity score (PS) widely employed as an inverse probability weight consistent estimation binary response variable interest. However, performance often becomes unstable due mis‐estimation PS. well stabilize performance, we propose new regression model that...
Detection of disease-associated markers plays a crucial role in gene screening for biological studies. Two-sample test statistics, such as the t-statistic, are widely used to rank genes based on expression data. However, resultant ranking is often not reproducible among different data sets. Such irreproducibility may be caused by disease heterogeneity. When we divided into two subsets, found that signs t-statistics were reversed. Focusing instability, proposed sign-sum statistic counts all...
Species distribution modeling plays an important role in estimating the habitat suitability of species using environmental variables. For this purpose, Maxent and Poisson point process are popular powerful methods extensively employed across various ecological biological sciences. However, computational speed becomes prohibitively slow when huge background datasets, which is often case with fine-resolution data or global-scale estimations. To address problem, we propose a computationally...