检索结果-内蒙古大学图书馆

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Scheible, Raphael Frei, Johann Thomczyk, Fabian He, Henry Tippmann, Patric Knaus, Jochen Jaravine, Victor Kramer, Frank Boeker, Martin Institute for AI and Informatics in Medicine University Hospital rechts der Isar Technical University Munich Germany IT-Infrastructure for Translational Medical Research Faculty of Applied Computer Science University of Augsburg Germany Data Integration Center Faculty of Medicine University of Freiburg Germany School of Computation Information and Technology Technical University Munich Germany Institute of Medical Biometry and Statistics Medical Center Faculty of Medicine University of Freiburg Germany Freiburg Center for Data Analysis and Modeling University of Freiburg Germany Hengrui Europe Biosciences Zurich Switzerland

ISBN: (纸本)9798891761643

Pre-trained language models have significantly advanced natural language processing (NLP), especially with the introduction of BERT and its optimized version, RoBERTa. While initial research focused on English, single-language models can be advantageous compared to multilingual ones in terms of pre-training effort, overall resource efficiency or downstream task performance. Despite the growing popularity of prompt-based LLMs, more compute-efficient BERT-like models remain highly relevant. In this work, we present the first German single-language RoBERTa model, GottBERT, pre-trained exclusively on the German portion of the OSCAR dataset. Additionally, we investigated the impact of filtering the OSCAR corpus. GottBERT was pre-trained using fairseq and standard hyperparameters. We evaluated its performance on two Named Entity Recognition (NER) tasks (Conll 2003 and GermEval 2014) and three text classification tasks (GermEval 2018 fine and coarse, and 10kGNAD) against existing German BERT models and two multilingual models. Performance was measured using the F1 score and accuracy. The GottBERT base and large models showed competitive performance, with GottBERT leading among the base models in 4 of 6 tasks. Contrary to our expectation, the applied filtering did not significantly affect the results. To support the German NLP research community, we are releasing the GottBERT models under the MIT license. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

On structural and practical identifiability: Current status and update of results

引用

Current Opinion in Systems Biology 2025年 41卷

作者： Heinrich, Mio Rosenblatt, Marcus Wieland, Franz-Georg Stigter, Hans Timmer, Jens University Freiburg Institute of Physics Germany CIBSS Centre for Integrative Biological Signalling Studies Freiburg Germany Freiburg Center for Data Analysis and Modeling Freiburg Germany Wageningen University Wageningen Netherlands

Identifiability of parameters in dynamical systems is a fundamental concept of mathematical modelling in systems biology and systems medicine. Both the structurally inherent identifiability of parameters in models and the practical identifiability of parameters, which arises from insufficient available data, play crucial roles in the development of useful models. Here, we provide an overview of recent developments in the field of structural identifiability analysis of models based on ordinary differential equations, emphasising its importance for accurate parameter estimation. We extend an existing benchmark study by comparing the methods for structural identifiability analysis with the recently developed StrucID, showing it to be a fast, efficient and intuitive algorithm. Furthermore, this review highlights the challenges in practical identifiability analysis and the need for benchmarking with real-world models using experimental data. The potential benefits of standardising documentation for benchmarking models with experimental data and practical non-identifiabilities are stressed. © 2025 The Author(s)

关键词：

来源：评论

学校读者我要写书评

暂无评论

Investigating a domain adaptation approach for integrating different measurement instruments in a longitudinal clinical registry

arXiv

引用

arXiv 2023年

作者： Hackenberg, Maren Pfaffenlehner, Michelle Behrens, Max Pechmann, Astrid Kirschner, Janbernd Binder, Harald Institute of Medical Biometry and Statistics Faculty of Medicine and Medical Center University of Freiburg Freiburg Germany Freiburg Center for Data Analysis and Modeling University of Freiburg Freiburg Germany Department of Neuropediatrics and Muscle Disorders Faculty of Medicine and Medical Center University of Freiburg Freiburg Germany University of Freiburg Freiburg Germany

In a longitudinal clinical registry, different measurement instruments might have been used for assessing individuals at different time points. To combine them, we investigate deep learning techniques for obtaining a joint latent representation, to which the items of different measurement instruments are mapped. This corresponds to domain adaptation, an established concept in computer science for image data. Using the proposed approach as an example, we evaluate the potential of domain adaptation in a longitudinal cohort setting with a rather small number of time points, motivated by an application with different motor function measurement instruments in a registry of spinal muscular atrophy (SMA) patients. There, we model trajectories in the latent representation by ordinary differential equations (ODEs), where person-specific ODE parameters are inferred from baseline characteristics. The goodness of fit and complexity of the ODE solutions then allows to judge the measurement instrument mappings. We subsequently explore how alignment can be improved by incorporating corresponding penalty terms into model fitting. To systematically investigate the effect of differences between measurement instruments, we consider several scenarios based on modified SMA data, including scenarios where a mapping should be feasible in principle and scenarios where no perfect mapping is available. While misalignment increases in more complex scenarios, some structure is still recovered, even if the availability of measurement instruments depends on patient state. A reasonable mapping is feasible also in the more complex real SMA dataset. These results indicate that domain adaptation might be more generally useful in statistical modeling for longitudinal registry data. © 2023, CC BY.

关键词： Ordinary differential equations

来源：评论

学校读者我要写书评

暂无评论

COMBINING PROPENSITY SCORE METHODS WITH VARIATIONAL AUTOENCODERS FOR GENERATING SYNTHETIC data IN PRESENCE OF LATENT SUB-GROUPS

arXiv

引用

arXiv 2023年

作者： Farhadyar, Kiana Bonofiglio, Federico Hackenberg, Maren Zöller, Daniela Binder, Harald Institute of Medical Biometry and Statistics Faculty of Medicine and Medical Center Freiburg Center for Data Analysis and Modeling University of Freiburg Freiburg Germany National Research Council of Italy ISMAR La Spezia Italy Institute of Medical Biometry and Statistics Faculty of Medicine and Medical Center University of Freiburg Freiburg Germany

In settings requiring synthetic data generation based on a clinical cohort, e.g., due to data protection regulations, heterogeneity across individuals might be a nuisance that we need to control or faithfully preserve. The sources of such heterogeneity might be known, e.g., as indicated by sub-groups labels, or might be unknown and thus reflected only in properties of distributions, such as bimodality or skewness. We investigate how such heterogeneity can be preserved and controlled when obtaining synthetic data from variational autoencoders (VAEs), i.e., a generative deep learning technique that utilizes a low-dimensional latent representation. To faithfully reproduce unknown heterogeneity reflected in marginal distributions, we propose to combine VAEs with pre-transformations. For dealing with known heterogeneity due to sub-groups, we complement VAEs with models for group membership, specifically from propensity score regression. The evaluation is performed with a realistic simulation design that features sub-groups and challenging marginal distributions. The proposed approach faithfully recovers the latter, compared to synthetic data approaches that focus purely on marginal distributions. Propensity scores add complementary information, e.g., when visualized in the latent space, and enable sampling of synthetic data with or without sub-group specific characteristics. We also illustrate the proposed approach with real data from an international stroke trial that exhibits considerable distribution differences between study sites, in addition to bimodality. These results indicate that describing heterogeneity by statistical approaches, such as propensity score regression, might be more generally useful for complementing generative deep learning for obtaining synthetic data that faithfully reflects structure from clinical cohorts. © 2023, CC BY.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

In-hospital safety of cryoballoon and radiofrequency ablation in patients with atrial fibrillation—German nationwide analysis of more than 300,000 procedures

引用

Heart Rhythm 2025年

作者： Jäckel, Markus Kaier, Klaus Steinfurt, Johannes Gressler, Alexander Staudacher, Dawid Leander Oettinger, Vera Hilgendorf, Ingo Gjermeni, Diona Rilinger, Jonathan Eichenlaub, Martin Westermann, Dirk Arentz, Thomas von zur Mühlen, Constantin Maier, Alexander Department of Cardiology and Angiology University Heart Center Freiburg–Bad Krozingen Faculty of Medicine University of Freiburg Freiburg Germany Center for Big Data Analysis in Cardiology (CeBAC) Department of Cardiology and Angiology University Heart Center Freiburg–Bad Krozingen Faculty of Medicine University of Freiburg Freiburg Germany Institute of Medical Biometry and Statistics Medical Center Faculty of Medicine University of Freiburg Freiburg Germany Department of Medicine III (Interdisciplinary Medical Intensive Care) Medical Center Faculty of Medicine University of Freiburg Freiburg Germany

Background: Pulmonary vein isolation can be performed by radiofrequency (RF) or cryoballoon (CB) ablation. Guidelines do not favor one technique, and knowledge about complication rates is limited. Objective: We aimed to report the procedural safety of RF and CB ablation using data from a German nationwide real-world registry. Methods: By use of health records, all left atrial catheter ablation procedures with RF or CB ablation in Germany from 2013 to 2021 were analyzed. After adjustment for confounders, safety performance end points were compared. Results: From 2013 to 2021, RF ablation was performed in 184,616 patients and CB ablation in 118,984 patients with increasing trends in patient numbers and performing centers for both procedures. Patients with RF ablation had slightly more comorbidities. In-hospital mortality (RF, 0.08%;CB, 0.06%) and other investigated complications were rare. After adjustment for patient baseline characteristics, the risk of in-hospital mortality, serious bleeding, stroke, intracerebral bleeding, and acute kidney injury did not differ. The risk of pericardiocentesis (relative risk, 0.50;95% confidence interval, 0.46–0.55;P < .001), vascular complication (0.36;0.33–0.39;P < .001), and ventilation >48 hours (0.81;0.66–0.99;P = .042) was significantly lower for CB ablation. Pericardiocentesis risk negatively correlated with annual procedure numbers per center with a faster learning curve for CB ablation (both P < .01). Conclusion: RF and CB ablation had low overall procedural complication rates, with CB ablation showing a 50% reduced risk of pericardiocentesis. centers with higher volume provided a better in-hospital safety with a faster learning curve for CB ablation. © 2025 Heart Rhythm Society

关键词： Atrial fibrillation Cardiac tamponade Catheter ablation Complications Cryoballoon Radiofrequency Volume outcome

来源：评论

学校读者我要写书评

暂无评论

Improving prediction models by incorporating external data with weights based on similarity

arXiv

引用

arXiv 2024年

作者： Behrens, Max Farhadizadeh, Maryam Rohde, Angelika Rühle, Alexander Nicolay, Nils H. Binder, Harald Zöller, Daniela Institute of Medical Biometry and Statistics Faculty of Medicine and Medical Center University of Freiburg Stefan-Meier-Strase 26 Freiburg79104 Germany Freiburg Center for Data Analysis and Modeling University of Freiburg Ernst-Zermelo-Strase 1 Freiburg79104 Germany University of Freiburg Department of Mathematical Stochastics Ernst-Zermelo-Strase 1 Freiburg79104 Germany Department of Radiation Oncology University of Freiburg-Medical Center Robert-Koch-Strase 3 Freiburg79106 Germany Department of Radiation Oncology University of Leipzig Stephanstr. 9a Leipzig04103 Germany

In clinical settings, we often face the challenge of building prediction models based on small observational data sets. For example, such a data set might be from a medical center in a multicenter study. Differences between centers might be large, thus requiring specific models based on the data set from the target center. Still, we want to borrow information from the external centers, to deal with small sample sizes. There are approaches that either assign weights to each external data set or each external observation. To incorporate information on differences between data sets and observations, we propose an approach that combines both into weights that can be incorporated into a likelihood for fitting regression models. Specifically, we suggest weights at the data set level that incorporate information on how well the models that provide the observation weights distinguish between data sets. Technically, this takes the form of inverse probability weighting. We explore different scenarios where covariates and outcomes differ among data sets, informing our simulation design for method evaluation. The concept of effective sample size is used for understanding the effectiveness of our subgroup modeling approach. We demonstrate our approach through a clinical application, predicting applied radiotherapy doses for cancer patients. Generally, the proposed approach provides improved prediction performance when external data sets are similar. We thus provide a method for quantifying similarity of external data sets to the target data set and use this similarity to include external observations for improving performance in a target data set prediction modeling task with small data. © 2024, CC BY.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Detecting frequency modulation in stochastic time-series data

引用

Physical Review E 2022年第2期106卷 024204-024204页

作者： Adrian L. Hauber Christian Sigloch Jens Timmer Institute of Physics University of Freiburg 79104 Freiburg Germany Freiburg Center for Data Analysis and Modeling (FDM) University of Freiburg 79104 Freiburg Germany Developmental Biology Institute Biology I Faculty of Biology University of Freiburg 79104 Freiburg Germany Spemann Graduate School of Biology and Medicine (SGBM) University of Freiburg 79104 Freiburg Germany Centre for Integrative Biological Signalling Studies (CIBSS) University of Freiburg 79104 Freiburg Germany

We propose a statistical test to identify nonstationary frequency-modulated stochastic processes from time-series data. Our method uses the instantaneous phase as a discriminatory statistics with reliable critical values derived from surrogate data. We simulated an oscillatory second-order autoregressive process to evaluate the size and power of the test. We found that the test we propose is able to correctly identify more than 99% of nonstationary data when the frequency of the simulated data is doubled after the first half of the time series. Our method is easily interpretable, computationally cheap, and does not require choosing hyperparameters that are dependent on the data.

关键词： Statistical methods Stochastic processes & statistics

来源：评论

学校读者我要写书评

暂无评论

TotalSegmentator MRI: Robust Sequence-independent Segmentation of Multiple Anatomic Structures in MRI

arXiv

引用

arXiv 2024年

作者： D'Antonoli, Tugba Akinci Berger, Lucas K. Indrakanti, Ashraya K. Vishwanathan, Nathan Weiss, Jakob Jung, Matthias Berkarda, Zeynep Rau, Alexander Reisert, Marco Küstner, Thomas Walter, Alexandra Merkle, Elmar M. Boll, Daniel Breit, Hanns-Christian Nicoli, Andrew Phillip Segeroth, Martin Cyriac, Joshy Yang, Shan Wasserthal, Jakob Clinic of Radiology and Nuclear Medicine University Hospital Basel Petersgraben 4 BaselCH-4031 Switzerland Department of Diagnostic and Interventional Radiology Faculty of Medicine Medical Center University of Freiburg University of Freiburg Germany Department of Stereotactic and Functional Neurosurgery Faculty of Medicine Medical Center University of Freiburg University of Freiburg Freiburg Germany Medical Image and Data Analysis University Hospital of Tuebingen Tuebingen Germany Department of Biomedical Physics in Radiation Oncology German Cancer Research Center Heidelberg Germany Scientific Computing Center Karlsruhe Institute of Technology Karlsruhe Germany

Background Since the introduction of TotalSegmentator CT, there is demand for a similar robust automated MRI segmentation tool that can be applied across all MRI sequences and anatomic structures. Purpose To develop and evaluate an automated MRI segmentation model for robust segmentation of major anatomic structures independently of MRI sequence. Materials and Methods In this retrospective study, a nnU-Net model (TotalSegmentator) was trained on MRI and CT examinations to segment 80 anatomic structures relevant for use cases such as organ volumetry, disease characterization, surgical planning and opportunistic screening. Examinations were randomly sampled from routine clinical studies to represent real-world examples. Dice scores were calculated between the predicted segmentations and expert radiologist reference standard segmentations to evaluate model performance on an internal test set, two external test sets and against two publicly available models, and TotalSegmentator CT. The model was applied to an internal dataset containing abdominal MRIs to investigate age-dependent volume changes. Wilcoxon signed-rank test was used. Results A total of 1143 examinations (616 MRIs, 527 CTs) (median age 61 years, IQR 50-72) were split into training (n=1088, CT and MRI) and an internal test set (n=55;only MRI), two external test sets (AMOS, n=20;CHAOS, n=20;only MRI), and an internal aging-study dataset of 8672 abdominal MRIs (median age 59 years, IQR 45-70) were included. The model showed a Dice Score of 0.839 on the internal test set and outperformed two other models (Dice Score, 0.862 versus 0.759;and 0.838 versus 0.560;p Copyright © 2024, The Authors. All rights reserved.

关键词： Computerized tomography

来源：评论

学校读者我要写书评

暂无评论

RecycleNet: Latent Feature Recycling Leads to Iterative Decision Refinement

RecycleNet: Latent Feature Recycling Leads to Iterative Deci...

引用

IEEE Workshop on Applications of Computer Vision (WACV)

作者： Gregor Koehler Tassilo Wald Constantin Ulrich David Zimmerer Paul F. Jaeger Jörg K. H. Franke Simon Kohl Fabian Isensee Klaus H. Maier-Hein Division of Medical Image Computing German Cancer Research Center (DKFZ) Heidelberg Germany Helmholtz Information and Data Science School for Health Karlsruhe/Heidelberg Germany Helmholtz Imaging DKFZ National Center for Tumor Diseases (NCT) NCT Heidelberg a Partnership Between DKFZ University Medical Center Heidelberg Interactive Machine Learning Group DKFZ Machine Learning Lab University of Freiburg Freiburg Germany Latent Labs (***) London UK Applied Computer Vision Lab DKFZ Pattern Analysis and Learning Group Heidelberg University Hospital Heidelberg Germany

Despite the remarkable success of deep learning systems over the last decade, a key difference still remains between neural network and human decision-making: As humans, we can not only form a decision on the spot, but also ponder, revisiting an initial guess from different angles, distilling relevant information, arriving at a better decision. Here, we propose RecycleNet, a latent feature recycling method, instilling the pondering capability for neural networks to refine initial decisions over a number of recycling steps, where outputs are fed back into earlier network layers in an iterative fashion. This approach makes minimal assumptions about the neural network architecture and thus can be implemented in a wide variety of contexts. Using medical image segmentation as the evaluation environment, we show that latent feature recycling enables the network to iteratively refine initial predictions even beyond the iterations seen during training, converging towards an improved decision. We evaluate this across a variety of segmentation benchmarks and show consistent improvements even compared with top-performing segmentation methods. This allows trading increased computation time for improved performance, which can be beneficial, especially for safety-critical applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adapting deep generative approaches for getting synthetic data with realistic marginal distributions

arXiv

引用

arXiv 2021年

作者： Farhadyar, Kiana Bonofiglio, Federico Zöller, Daniela Binder, Harald Institute of Medical Biometry and Statistics Faculty of Medicine and Medical Center University of Freiburg Freiburg79104 Germany Freiburg Center for Data Analysis and Modeling University of Freiburg Freiburg79104 Germany National Research Council of Italy Ismar Forte Santa Teresa Lerici - La Spezia19032 Italy

Synthetic data generation is of great interest in diverse applications, such as for privacy protection. Deep generative models, such as variational autoencoders (VAEs), are a popular approach for creating such synthetic datasets from original data. Despite the success of VAEs, there are limitations when it comes to the bimodal and skewed marginal distributions. These deviate from the unimodal symmetric distributions that are encouraged by the normality assumption typically used for the latent representations in VAEs. While there are extensions that assume other distributions for the latent space, this does not generally increase flexibility for data with many different distributions. Therefore, we propose a novel method, pre-transformation variational autoencoders (PTVAEs), to specifically address bimodal and skewed data, by employing pre-transformations at the level of original variables. Two types of transformations are used to bring the data close to a normal distribution by a separate parameter optimization for each variable in a dataset. We compare the performance of our method with other state-of-the-art methods for synthetic data generation. In addition to the visual comparison, we use a utility measurement for a quantitative evaluation. The results show that the PTVAE approach can outperform others in both bimodal and skewed data generation. Furthermore, the simplicity of the approach makes it usable in combination with other extensions of VAE. Copyright © 2021, The Authors. All rights reserved.

关键词： Metadata

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：