Article
Mathematics, Interdisciplinary Applications
Zhen Meng, Na Li, Ao Yuan
Summary: In this study, two test statistics are proposed for the high-dimensional nonparametric Behrens-Fisher problem: U-statistic Rank-based Test (URT) and Cauchy Combination Test (CCT). Simulation results show that URT is robust to outliers while closed form for calculating the statistical significance of CCT has been provided.
JOURNAL OF SYSTEMS SCIENCE & COMPLEXITY
(2022)
Article
Health Care Sciences & Services
Zhen Meng, Qinglong Yang, Qizhai Li, Baoxue Zhang
Summary: This study proposes a directional-sum test for the nonparametric Behrens-Fisher problem, utilizing a division-combination strategy and a one-layer wild bootstrap procedure. Simulation studies demonstrate that the proposed test effectively controls type I error rates and outperforms existing rank-sum and maximum-type tests in most scenarios. Applications to a dietary intervention trial further confirm the performance of the proposed test.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2021)
Article
Mathematics
Jianqi Yu
Summary: This article defines the hierarchical data missing pattern and explores how to extend ideas for dealing with monotone missing data to hierarchical missing patterns. The study introduces a pivotal quantity similar to Hotelling T2 and uses the moment matching method to derive its approximate distribution. The precision of the approximation is demonstrated through Monte Carlo data simulation, showing that the approximate method is satisfactory even for moderately small samples.
JOURNAL OF MATHEMATICS
(2021)
Article
Mathematics
Fco Javier Giron, Carmen del Castillo
Summary: The paper presents a simple solution to the Behrens-Fisher problem based on Bayes factors, exploring its relation with the Behrens-Fisher distribution. The Bayes factor is constructed using a simple hierarchical model and a closed form based on the densities of general Behrens-Fisher distributions. Additionally, simple asymptotic approximations of the Bayes factor, functions of the Kullback-Leibler divergence between normal distributions, are provided along with evidence of its consistency, accompanied by examples and comparisons.
REVISTA DE LA REAL ACADEMIA DE CIENCIAS EXACTAS FISICAS Y NATURALES SERIE A-MATEMATICAS
(2021)
Article
Mathematics
Serim Hong, Carlos A. Coelho, Junyong Park
Summary: This paper presents approaches based on exact and near-exact distributions to solve the Behrens-Fisher problem of testing the equality of means of two normal distributions. The proposed tests show better statistical power than Welch's t-test, particularly for small and unbalanced sample sizes.
Article
Biology
Rui Wang, Wangli Xu
Summary: This paper presents an approximate randomization test procedure for comparing the population means of two groups of independent observations. The asymptotic behavior of the test statistic and randomized statistic is studied, showing that the proposed test procedure has good power behavior and favorable performance in numerical experiments.
Article
Health Care Sciences & Services
Kerstin Rubarth, Markus Pauly, Frank Konietschke
Summary: This study developed purely nonparametric methods for analyzing repeated measures designs with missing values, formulating hypotheses in terms of purely nonparametric treatment effects. Solutions to the nonparametric Behrens-Fisher problem were presented, along with global testing, multiple contrast test procedures, and simultaneous confidence intervals for treatment effects. These methods demonstrated satisfactory control of the nominal type-I error rate even for small sample sizes and high amounts of missing data, and were successfully applied to a real data set.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2022)
Article
Statistics & Probability
Tommaso Lando, Idir Arab, Paulo Eduardo Oliveira
Summary: This study examines the comparison of ageing patterns of k-out-of-n systems with i.i.d. components, focusing on determining stochastic ordering relationships between different order statistics. It introduces a hierarchy of reference functions to express varying knowledge precision about the distribution of components lifetimes, deriving sufficient dominance conditions depending on the identified class. The method's applicability is discussed, with a test for relative convexity characterizing its central role in the proposed approach.
Article
Statistics & Probability
Daojiang He, Huijun Shi, Kai Xu, Mingxiang Cao
Summary: This paper investigates the problem of testing the equality of the mean vectors of k populations with possibly unknown and unequal covariance matrices in high-dimensional settings. The proposed test offers much flexibility in practical applications by assuming only mild additional conditions and using the Welch-Satterthwaite chi 2-approximation to mimic the shape of the null distribution of the test statistic. Extensive simulation studies show that the proposed test outperforms existing methods in terms of both size and power.
JOURNAL OF NONPARAMETRIC STATISTICS
(2023)
Article
Automation & Control Systems
Fei Lu, Mauro Maggioni, Sui Tang
Summary: This paper discusses the inverse problem of learning interaction laws from observation data, and proposes an estimator based on minimizing a regularized least squares functional. Numerical simulations show that the learnability condition is satisfied in practical models, and the estimators are robust to noise in the observations. The estimators are also capable of producing accurate predictions of trajectories in large time intervals.
JOURNAL OF MACHINE LEARNING RESEARCH
(2021)
Article
Statistics & Probability
Rendao Ye, Bingni Fang, Zhongchi Wang, Kun Luo, Wenting Ge
Summary: This article focuses on the inference on location parameter for the skew-normal population under dependent samples. Bootstrap test statistics and Bootstrap confidence intervals are constructed for the Behrens-Fisher-type problem, when the scale parameter or skewness parameter is known. The Monte-Carlo simulation results indicate that the Bootstrap approach is generally better than the approximate approach. Real data examples of gross domestic product and stock closing price are used to illustrate the above approaches.
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
(2023)
Article
Engineering, Multidisciplinary
Anan Tang, Yuanman Ma, Xuelong Hu, Weidong Huang
Summary: This paper presents a discrete state nonparametric AEWMA chart based on the Wilcoxon signed-rank statistics. The chart takes into account the discrete nature of the statistics and the design of the chart structure. The statistical properties of the chart are accurately evaluated using a Markov chain approach. Simulation results show that the AEWMA signed-rank scheme performs well across various distributions and is effective in detecting both small and large shifts. A real data example is also provided to demonstrate the practical implications of the monitoring scheme.
QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL
(2022)
Article
Statistics & Probability
Fazil Aliev, Levent Ozbek, Mehmet Fedai Kaya, Coskun Kus, Hon Keung Tony Ng, Haikady N. Nagaraja
Summary: In this study, a nonparametric test procedure based on order statistics is proposed for testing the equality of two continuous distributions. The exact null distribution is obtained using enumeration method and combinatorial argument. The study provides methods to compute the mean, variance, and critical values of the distribution. The power performance of the new test is examined for different alternatives and compared to commonly used nonparametric tests. The study also discusses a generalization of the proposed procedure for unequal sample sizes and includes an illustrative example.
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
(2022)
Article
Computer Science, Artificial Intelligence
Berkan Kadioglu, Peng Tian, Jennifer Dy, Deniz Erdogmus, Stratis Ioannidis
Summary: This paper focuses on the rank regression setting, where a dataset of N samples with features is ranked through pairwise comparisons by a noisy oracle. By observing the comparison dataset, the learner aims to regress sample ranks and learn the model parameters with a certain level of accuracy.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Berkan Kadioglu, Peng Tian, Jennifer Dy, Deniz Erdogmus, Stratis Ioannidis
Summary: In this research, the ranking regression problem is discussed, and it is shown that learning through paired comparisons can regress the rank of samples.
PATTERN RECOGNITION
(2022)
Article
Multidisciplinary Sciences
Kerstin Rubarth, Paavo Sattler, Hanna Gwendolyn Zimmermann, Frank Konietschke
Summary: Clustered data often occur in practical applications, such as animal experiments. Standard methods like Linear Mixed Models and Generalized Estimating Equations are difficult to check for assumptions in scenarios with small sample sizes, highly skewed, count, and ordinal or binary data. Wilcoxon-Mann-Whitney type effects provide suitable alternatives that do not require specific data distribution and can handle missing data.
Article
Cardiac & Cardiovascular Systems
Leonard Bergau, Vanessa Sciacca, Vinitha Nesapiragasan, Kerstin Rubarth, Frank Konietschke, Thomas Fink, Mustapha El Hamriti, Guram Imnadze, Lilas Dagher, Martin Braun, Moneeb Khalaph, Denise Guckel, Johannes Heintze, Georg Noelker, Juergen Vogt, Philipp Sommer, Christian Sohns
Summary: This study analyzed data from consecutive patients treated with CB-PVI, showing that CB-PVI as a preferred treatment for AF achieved favorable long-term outcomes, and factors such as left atrium diameter may influence recurrence.
JOURNAL OF CARDIOVASCULAR ELECTROPHYSIOLOGY
(2022)
Article
Surgery
Safwan Omran, Verena Mueller, Larissa Schawe, Matthias Buerger, Sebastian Kapahnke, Leon Bruder, Haidar Haidar, Frank Konietschke, Andreas Greiner
Summary: This study evaluated the early and midterm outcomes of the Endurant II stent graft and compared the results based on the anatomic severity grade (ASG) scores. The study found that Endurant II was safe and effective in both low-score and high-score patients. However, patients in the high-score group had a higher risk of implant-related complications and midterm mortality.
JOURNAL OF ENDOVASCULAR THERAPY
(2023)
Article
Medicine, Research & Experimental
Milan Haffke, Helma Freitag, Gordon Rudolf, Martina Seifert, Wolfram Doehner, Nadja Scherbakov, Leif Hanitsch, Kirsten Wittke, Sandra Bauer, Frank Konietschke, Friedemann Paul, Judith Bellmann-Strobl, Claudia Kedor, Carmen Scheibenbogen, Franziska Sotzny
Summary: A subset of PCS patients exhibit evidence of endothelial dysfunction with altered endothelial biomarkers. Different associations of the RHI with clinical parameters and varying biomarker profiles may suggest distinct pathomechanisms among patient subgroups.
JOURNAL OF TRANSLATIONAL MEDICINE
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Nick Lasse Beetz, Matthias Haas, Alexander Baur, Frank Konietschke, Akash Roy, Charlie Alexander Hamm, Madhuri Monique Rudolph, Seyd Shnayien, Bernd Hamm, Hannes Cash, Patrick Asbach, Tobias Penzkofer
Summary: This study analyzed the inter-reader variability in classifying prostate lesions using PI-RADS versions 2 and 2.1. The results showed that version 2.1 had better inter-reader agreement in the peripheral zone but poorer agreement in the transition zone. Combining scores 1 and 2 improved inter-reader agreement in both zones, suggesting a need for refined definitions for these PI-RADS scores.
ROFO-FORTSCHRITTE AUF DEM GEBIET DER RONTGENSTRAHLEN UND DER BILDGEBENDEN VERFAHREN
(2022)
Article
Peripheral Vascular Disease
Anita Micheel, Frank Konietschke, Irene Hinterseher, Sebastian Kapahnke, Matthias Buerger, Ben Raude, Larissa Schawe, Safwan Omran, Andreas Greiner, Jan Paul Frese
Summary: Age and comorbid conditions are risk factors for adverse outcomes after CEA. The V-POSSUM model is better than POSSUM to predict postoperative death and stroke after CEA in patients with symptomatic carotid stenosis and a high preoperative physiological score.
VASA-EUROPEAN JOURNAL OF VASCULAR MEDICINE
(2022)
Article
Peripheral Vascular Disease
Safwan Omran, Steffen Groeger, Leon Bruder, Matthias Buerger, Sebastian Kapahnke, Haidar Haidar, Frank Konietschke, Andreas Greiner
Summary: Non-staged neoaortoiliac system reconstruction and cryopreserved human allografts show comparable short- and midterm results for treating aortic graft infections. However, both procedures remain challenging with high morbidity and mortality rates.
Article
Oncology
Serge Leyvraz, Frank Konietschke, Caroline Peuker, Moritz Schuette, Thomas Kessler, Sebastian Ochsenreither, Marc Ditzhaus, Erin D. Spruenken, Gina Doerpholz, Mario Lamping, Damian T. Rieke, Konrad Klinghammer, Susen Burock, Claas Ulrich, Gabriela Poch, Reinhold Schaefer, Frederick Klauschen, Antonia Joussen, Marie-Laure Yaspo, Ulrich Keilholz
Summary: The study evaluated the feasibility of a precision oncology strategy for treating metastatic uveal melanoma in routine clinical practice. The findings showed that selecting matched therapies based on molecular aberrations can provide clinical benefits for some patients.
EUROPEAN JOURNAL OF CANCER
(2022)
Article
Health Care Sciences & Services
Claus P. Nowak, Tobias Muetze, Frank Konietschke
Summary: This article investigates interim analysis in late phase clinical trials and proposes methods for repeated confidence intervals and simplified sample size determination. The study finds that the Wilcoxon-Mann-Whitney test, the Brunner-Munzel test, and a test procedure based on the log win odds show good applicability in the nonparametric comparison of two treatment arms. Simulations confirm the adequacy of the proposed methods, and the methodology is applied to a clinical trial in patients with relapse-remitting multiple sclerosis.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2022)
Article
Mathematical & Computational Biology
Sophie K. Piper, Dario Zocholl, Ulf Toelch, Robert Roehle, Andrea Stroux, Johanna Hoessler, Anne Zinke, Frank Konietschke
Summary: Statistical planning plays a crucial role in animal experiments, and unified biostatistical standards are essential for improving the quality of preclinical research projects.
BIOMETRICAL JOURNAL
(2023)
Article
Oncology
Nick Lasse Beetz, Franziska Drager, Charlie Alexander Hamm, Seyd Shnayien, Madhuri Monique Rudolph, Konrad Frobose, Sefer Elezkurtaj, Matthias Haas, Patrick Asbach, Bernd Hamm, Samy Mahjoub, Frank Konietschke, Maximilian Wechsung, Felix Balzer, Hannes Cash, Sebastian Hofbauer, Tobias Penzkofer
Summary: This study aimed to analyze how many MRI-targeted biopsy cores are needed to establish the most relevant histopathologic diagnosis of the index lesion and to build a prediction model. The results showed that the most relevant histopathologic diagnosis of the index lesion could be made on the basis of three MRI-targeted biopsy cores in 97% of patients.
PROSTATE CANCER AND PROSTATIC DISEASES
(2023)
Article
Multidisciplinary Sciences
Angelique Wolter, Christian H. Bucher, Sebastian Kurmies, Viktoria Schreiner, Frank Konietschke, Katharina Hohlbaum, Robert Klopfleisch, Max Loehning, Christa Thoene-Reineke, Frank Buttgereit, Joerg Huwyler, Paulin Jirkof, Anna E. Rapp, Annemarie Lang
Summary: In animal experiments, adequate pain management is necessary and should last for the expected pain period without repeated application. The current depot formulations of Buprenorphine have limited duration of action, but a new sustained-release formulation (BUP-Depot) has been developed as a potential alternative in Europe. It provides effective analgesia for 72 hours, similar to Tramadol in the drinking water.
SCIENTIFIC REPORTS
(2023)
Article
Clinical Neurology
Susanna Asseyer, Nasrin Asgari, Jeffrey Bennett, Omer Bialer, Yolanda Blanco, Francesca Bosello., Anna Camos-Carreras., Edgar Carnero Contentti., Sara Carta, John Chen, Claudia Chien, Mashina Chomba, Russell C. Dale, Josep Dalmau, Kristina Feldmann, Eoin P. Flanagan, Caroline Froment Tilikete, Carolina Garcia-Alfonso, Joachim Havla, Mark Hellmann, Ho Jin Kim, Philipp Klyscz, Frank Konietschke, Chiara La Morgia, Marco Lana-Peixoto, Maria Isabel Leite, Netta Levin, Michael Levy, Sara Llufriu, Pablo Lopez, Itay Lotan, Alessandra Lugaresi, Romain Marignier, Sara Mariotto, Susan P. Mollan, Cassandra Ocampo, Frederike Cosima Oertel, Maja Olszewska, Jacqueline Palace, Lekha Pandit, Jose Luis Peralta Uribe, Sean Pittock, Sudarshini Ramanathan, Natthapon Rattanathamsakul, Albert Saiz, Sara Samadzadeh, Bernardo Sanchez-Dalmau, Deanna Saylor, Michael Scheel, Tanja Schmitz-Huebsch, Jemal Shifa, Sasitorn Siritho, Pia S. Sperber, Prem S. Subramanian, Alon Tiosano, Adi Vaknin-Dembinsky, Alvaro Jose Mejia Vergara, Adi Wilf-Yarkoni, Luis Alfonso Zarco, Hanna G. Zimmermann, Friedemann Paul, Hadas Stiebel-Kalish
Summary: This is a global study aimed at investigating the effect of early high-dose corticosteroid treatment on optic neuritis. The study will collect various data to evaluate treatment strategies and the accuracy of diagnostic criteria.
FRONTIERS IN NEUROLOGY
(2023)
Article
Medicine, General & Internal
Safwan Omran, Larissa Schawe, Frank Konietschke, Stefan Angermair, Benjamin Weixler, Sascha Treskatsch, Andreas Greiner, Christian Berger
Summary: This retrospective study evaluated perioperative and ICU variables to predict colonic ischemia after infrarenal RAAA surgery. The study found that most patients developed CI within the first 3 postoperative days. Laboratory findings such as serum lactate, minimum pH, serum bicarbonate, and platelet count were associated with the development of CI.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Medicine, General & Internal
Carmen Scheibenbogen, Judith Theresia Bellmann-Strobl, Cornelia Heindrich, Kirsten Wittke, Elisa Stein, Christiana Franke, Harald Pruess, Hannah Pressler, Marie-Luise Machule, Heinrich Audebert, Carsten Finke, Hanna Gwendolyn Zimmermann, Birgit Sawitzki, Christian Meisel, Markus Toelle, Anne Krueger, Anna C. Aschenbrenner, Joachim L. Schultze, Marc D. Beyer, Markus Ralser, Michael Muelleder, Leif Erik Sander, Frank Konietschke, Friedemann Paul, Silvia Stojanov, Lisa Bruckert, Dennis M. Hedderich, Franziska Knolle, Gabriela Riemekasten, Maria J. G. T. Vehreschild, Oliver A. Cornely, Uta Behrends, Susen Burock
Summary: The sequela of COVID-19, known as post-COVID-19 condition or syndrome (PCS), encompasses a range of symptoms caused by immune dysregulation, autoimmunity, endothelial dysfunction, viral persistence, and viral reactivation. However, it is unclear whether different clinical subgroups of PCS can be distinguished based on the expression of biomarkers. There is an overlap between symptoms and mechanisms of PCS and postinfectious myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). Currently, there are no curative therapies available for ME/CFS or PCS, but the identified mechanisms provide targets for therapeutic interventions. To accelerate the development of therapies, it is proposed to evaluate drugs targeting different mechanisms in clinical trial networks using harmonized diagnostic and outcome criteria and subgroup patients based on comprehensive clinical profiling, including diagnostic and biomarker phenotyping.
FRONTIERS IN MEDICINE
(2023)