• 学术搜索
  • 科研智能体
    • Research Labs
    • AI 阅读
    • AI 文库
    • 深度研究
    • 学者亮点
  • 学术资源
    • AI2000
    • 期刊/会议
    • 学者库
    • 学术API
      MCP&Skill
    • 溯源树
    • 数据集
  • 知识沉淀
    • 学术空间
订阅小程序
旧版功能
aminer vip
开通会员低至0.73元/天
一次搞定AI科研
立即登录
  • English
  • 联系方式
    柏

    柏林工程应用科学大学

    HTW Berlin - University of Applied Sciences
    院校EST. 1994
    906论文总数
    7,446引用总数

    论文量&引用量时间轴

    机构学者

    论文(906)

    年份
    起
    –
    止
    排序
    1RamanBench: A Large-Scale Benchmark for Machine Learning on Raman Spectroscopy
    Mario Koddenbrock, Christoph Lange, Robin Legner,Martin Jaeger,Martin Kögler, Mariano N Bournazou,Peter Neubauer, Felix Bießmann,Erik Rodner

    Machine learning (ML) has transformed many scientific fields, yet key applications still lack standardized benchmarks. Raman spectroscopy, a widely used technique for non-invasive molecular analysis, is one such field where progress is limited by fragmented datasets, inconsistent evaluation, and models that fail to capture the structure of spectral data. We introduce RamanBench, the first large-scale, fully reproducible benchmark for ML on Raman spectroscopy, consisting of streamlined data access (https://pypi.org/project/raman-data/), evaluation protocols and code (https://pypi.org/project/raman-bench/), as well as a live leaderboard (https://huggingface.co/spaces/HTW-KI-Werkstatt/RamanBench). It unifies 77 datasets (including 17 first released with this benchmark) across four domains, comprising 325,862 spectra and spanning classification and regression tasks under diverse experimental conditions. We benchmark 28 models under a standardized protocol, including classical methods (e.g., PLS), Raman-specific (e.g., RamanNet), tabular foundation models (TFMs; e.g., TabPFN), and time-series approaches (e.g., ROCKET). TFMs consistently outperform domain-specific and gradient boosting baselines, while time-series models remain competitive. However, no single method dominates across these heterogeneous datasets, revealing that Raman spectroscopy remains an open challenge for machine learning. Therefore, we invite the community to contribute new approaches to our living benchmark, with the potential to accelerate advances in critical applications such as medical diagnostics, biological research, and materials science.

    2026NeurIPS 2026(2026)引用:4
    引用
    AI阅读
    加入学术空间
    2The Multimodal Anonymizer: a Fully Local Multi-Agent AI System for Medical Data Deidentification
    Anja Hirsch, Foo Wei Ten, Kyle Steven Krueger, Robin Geyer, Tobias Roeschl, Matthias Groeschel, Paul Rostin, Roland Eils, Martin Spott,Fabian Prasser, Alexander Meyer, Julian Madrid

    Background: Safe reuse of multimodal hospital data for AI development is limited by the absence of reliable, context-aware deidentification across multimodal data and longitudinal patient data. Existing approaches are largely modality-specific and can indiscriminately remove clinically important information. Methods: We developed the Multimodal Anonymizer, a modular, locally deployable multi-agent framework integrating multimodal large language models, task-specific neural networks and rule-based transformations. We evaluated 16 orchestrator model configurations on a benchmark built from publicly available data and hospital data from our institution. The benchmark dataset included data from different origins: 250 MIMIC-IV patients with synthetically injected personally identifiable information (PII) supplemented with head CT, face images, handwriting, audio, German clinical-text datasets and local data. Primary outcomes were deidentification sensitivity and preservation of clinically important content; secondary analyses examined model characteristics, reproducibility, and performance against leading market and open-source solutions. Results: The best local configuration (the orchestrator being Qwen3-VL-235B-A22B-Thinking) achieved near-complete deidentification across all datasets, with per-patient sensitivity of 98.80% (95%-CI 97.20; 100), and per-PII sensitivity of 99.82% (95%-CI 99.76; 99.88). Critical clinical preservation was 99.60% (95%-CI 98.80; 100) per-patient, and clinical preservation was 99.61% (95%-CI 99.51; 99.71) per-file. All modalities achieved at least 98.30% sensitivity (lower bound 95%-CI). On our local data, the system achieved a deidentification sensitivity of 100% per-patient and per-PII; and a critical clinical preservation of 100% per-patient as well as a clinical preservation of 99.97% (95%-CI 99.91; 100) per-file. When comparing orchestrators, the leading local models were similar to proprietary models (GPT-5.2) in deidentification sensitivity while showing higher deidentification specificity. The Multimodal Anonymizer outperformed previous tools on most modalities. Conclusion: Near-complete, utility-preserving deidentification of multimodal clinical data is achievable with a unified, locally deployable multi-agent system, enabling safer large-scale reuse of hospital data for research and AI development. ### Competing Interest Statement This study was supported by the Berlin Institute for the Foundations of Learning and Data (BIFOLD). AH, RG, AM, and JM received support from BIFOLD during the conduct of the study. TR and MG were supported by the BIH Digital Clinician Scientist Program of the Berlin Institute of Health at Charite. The funders had no role in the design or conduct of the study; the collection, management, analysis, or interpretation of the data; the preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: Publicly available datasets were used in accordance with their respective access and licensing terms. The Charite Ethics Committee approved the retrospective use and anonymization of the institution-specific partogram dataset from Charite Universitaetsmedizin Berlin (EA2/267/25) without requiring individual informed consent. Reproducibility materials are provided as permitted by data-use agreements, institutional policy, and patient privacy requirements. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data Availability Statement The data analyzed in this study were obtained from a combination of credentialed-access, publicly available, and institution-specific sources. MIMIC-IV and linked MIMIC-IV resources are available from their original providers subject to required training, data-use agreements, and credentialed access. Additional public datasets are available from their original sources under their respective licenses and access conditions. The institution-specific partogram dataset from Charite Universitaetsmedizin Berlin is not publicly available because it contains sensitive clinical information and is subject to legal and institutional restrictions. Derived benchmark annotations, synthetic-PII injection specifications and aggregate evaluation outputs are available from the corresponding author on reasonable request, where consistent with applicable data-use agreements and privacy regulations. Code Availability Statement The code used for benchmark construction, evaluation, and statistical analysis is available at https://github.com/charite-iaim/multimodal-anonymizer. A plug-and-play version for non-technical user is provided to reproduce the reported analyses, together with documentation, prompts, model settings, and package versions in the code repository. The framework is provided with an easy-to-use browser-based user interface in order to facilitate the wide adoption of the framework. BIFOLD BIH Digital Clinician Scientist Program of the Berlin Institute of Health at Charité

    2026引用:1
    引用
    AI阅读
    加入学术空间
    3Synthetic Data Enables Human-Grade Microtubule Analysis with Foundation Models for Segmentation
    Mario Koddenbrock, Justus Westerhoff, Dominik Fachet,Simone Reber, Felix A Gers, Erik Rodner

    Studying microtubules (MTs) and their mechanical properties is central to understanding intracellular transport, cell division, and drug action. While important, experts still need to spend many hours manually segmenting these filamentous structures. The suitability of state-of-the-art methods for this task cannot be systematically assessed, as large-scale labeled datasets are missing. We address this gap by introducing the synthetic dataset SynthMT, produced by tuning a novel image generation pipeline on real-world interference reflection microscopy (IRM) frames of in vitro reconstituted MTs without requiring human annotations. In our benchmark, we evaluate nine fully automated methods for MT analysis in both zero- and Hyperparameter Optimization (HPO)-based few-shot settings. Across both settings, classical algorithms and current foundation models still struggle to achieve the accuracy required for biological downstream analysis on in vitro MT IRM images that humans perceive as visually simple. However, a notable exception is the recently introduced SAM3 model. After HPO on only ten random SynthMT images, its text-prompted version SAM3Text achieves near-perfect and in some cases super-human performance on unseen, real data. This indicates that fully automated MT segmentation has become feasible when method configuration is effectively guided by synthetic data. To enable progress, we publicly release the dataset, the generation pipeline, and the evaluation framework.

    2026PLoS computational biology(2026)引用:1
    引用
    AI阅读
    加入学术空间
    4Adaptive Segmented Regression for the Modeling of Temperature-Dependent Material Properties
    Nataliya Togobytska, Dmytro Stepanenko, Christoph Möhs

    This paper presents a mathematical approach for the modeling of the physical properties of steel slag, an industrial by-product stemming from the steelmaking process. Due to its substantial volume, steel slag poses a considerable risk as landfill waste and a potential source of pollution. Advancements in technology have led to the recycling and reuse of reprocessed, granulated slag as an eco-friendly construction material. Studying the physical properties of slag melts and understanding the factors that influence these properties hold significant importance. The modeling of temperature-dependent electrical conductivity and viscosity of slag melts is based on a piecewise linear regression approximation within temperature ranges defined by expert knowledge. To enhance the reliability of modeling outcomes, we propose the application of a genetic algorithm for derivative-free optimization of breakpoints in segmented regression. The practical implications of these results are significant in the selection of appropriate slags for various metallurgical processes.

    2026Progress in Industrial Mathematics at ECMI 2023(2026)
    引用
    AI阅读
    加入学术空间
    5Bidirectional 3-Km FSO Transmission Using Multi-Aperture Space Diversity and Digital Subcarrier Combining
    Aymeric Arnould,Abraham Johst,Robert Emmerich,Marcel Rothe,Peter Hanne,Colja Schubert,Nicolas Perlot, Markus Nölle,Georg Rademacher,Ronald Freund

    Digital subcarrier combining is used for bidirectional FSO transmission between single-and multi-aperture terminals, with only one transmitter and receiver per direction. Two digital combining schemes are evaluated on a 3-km link over 24 hours.

    20262026 Optical Fiber Communications Conference and Exhibition (OFC)(2026)
    引用
    AI阅读
    加入学术空间
    立即登录,查看全部 906 篇论文

    合作机构(100)

    柏林工业大学合作论文 41
    柏林洪堡大学合作论文 14
    Federal Institute For Materials Research and Testing合作论文 12
    德国亥姆霍兹研究中心协会合作论文 12
    柏林夏里特大学医学院合作论文 11
    德累斯顿工业大学合作论文 11
    弗劳恩霍夫协会合作论文 7
    安美特合作论文 7
    汉堡大学合作论文 6
    莱比锡大学合作论文 6

    机构统计