Publikationen

2026

• Stuck on Suggestions: Automation Bias, the Anchoring Effect, and the Factors That Shape Them in Computational Pathology

  Rosbach, E., Ammeling, J., Ganz, J., Bertram, C. A., Conrad, T., Riener, A., & Aubreville, M. (2026). Stuck on Suggestions: Automation Bias, the Anchoring Effect, and the Factors That Shape Them in Computational Pathology. Machine Learning for Biomedical Imaging, 3, 126–147. http://doi.org/https://doi.org/10.59275/j.melba.2026-87b1 (Original work published 2026)

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Aubreville, Marc

  • Abteilung/Einrichtung

    • CIVU – Center for Interaction, Visualization and Usability

  • Forschungsfeld

    • Data Science und Künstliche Intelligenz

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Emely Rosbach
    • Jonas Ammeling
    • Jonathan Ganz
    • Christof Bertram
    • Thomas Conrad
    • Andreas Riener
    • Marc Aubreville

• Performance evaluation of deep learning models for image analysis: Considerations for visual assessment and statistical metrics

  Bertram, C. A., Ammeling, J., Bartel, A., Beamer, G., & Aubreville, M. (2026). Performance evaluation of deep learning models for image analysis: Considerations for visual assessment and statistical metrics. Veterinary Pathology, 03009858261461760. http://doi.org/10.1177/03009858261461760 (Original work published 2026)

  Abstract

  anzeigen ▸

  Deep learning-based automated image analysis (DL-AIA) has been shown to outperform trained pathologists in tasks related to feature quantification. Related to these capabilities, the use of DL-AIA tools is currently extending from proof-of-principle studies to routine applications, including the evaluation of patient samples (diagnostic pathology), regulatory safety assessment (toxicologic pathology), and recurrent research tasks. To ensure that DL-AIA applications are safe and reliable, it is critical to conduct a thorough and objective generalization performance assessment to evaluate an algorithm’s ability to accurately predict patterns of interest and possibly evaluate model robustness (ie, the algorithm’s capacity to maintain predictive accuracy on images from different sources). In this article, we review the practices for performance assessment in veterinary pathology publications by which 2 main approaches were identified: (1) exclusive visual performance assessment (ie, eyeballing algorithmic predictions) plus validation of the model’s application utilizing secondary performance indices and (2) statistical performance assessment (alongside other methods), which requires creation of a test set with ground truth labels that is a hold-out from model development. This article compares the strengths and weaknesses of statistical and visual performance assessment methods. Furthermore, we discuss relevant considerations for rigorous statistical performance evaluation including metric selection, test data set image composition, ground truth label quality, resampling methods such as bootstrapping, statistical comparison of multiple models, and evaluation of model stability. It is our conclusion that visual and statistical evaluation have complementary strengths and a combined approach provides the greatest insight into the DL model’s performance and sources of error.

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Aubreville, Marc

  • Forschungsprojekt

    • (mu)ROMI: Robust and Accurate Multi-Tumor, Multi-Species, Multi-Laboratory and Multi-Scanner Mitosis Detection with Large-Scale Datasets and Artificial Intelligence

  • Abteilung/Einrichtung

    • FLAIR - Flensburg Artificial Intelligence Research

  • Profilbereich

    • Informatik und Gestaltung

  • Forschungsfeld

    • Data Science und Künstliche Intelligenz

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Christof Bertram
    • Jonas Ammeling
    • Alexander Bartel
    • Gillian Beamer
    • Marc Aubreville

• Inherent uninterruptible power supply via directly grid-connected machines and variable inertia flywheels

  Rettig, A., Jahromi, S. N., Jauch, C., & Reese, L. (2026). Inherent uninterruptible power supply via directly grid-connected machines and variable inertia flywheels. Electric Power Systems Research, 256, 11. http://doi.org/https://doi.org/10.1016/j.epsr.2026.112880 (Original work published 2026)

  Abstract

  anzeigen ▸

  This paper analyses the potential of a novel hydraulic variable inertia flywheel to improve standby power systems when attached to directly grid-connected electrical machines. Conventional standby power systems often require several seconds to reach full operation after a grid outage. During this time, critical loads may be left without power, unless supported by an additional uninterruptible power supply system. By flange-mounting a hydraulic variable inertia flywheel to an existing grid-connected electrical machine, emergency power can be provided immediately without the need for power electronics. The ability of the hydraulic variable inertia f lywheel to discharge at a quasi-constant speed offers improved frequency and voltage stability. The performance of the hydraulic variable inertia flywheel is compared with traditional flywheels of equivalent inertia and energy content across the two most common machines. The simulation results identify the most appropriate machine- f lywheel configuration to complement standby power systems. Combining a hydraulic variable inertia f lywheel with an electrically excited synchronous machine provides the most stable voltage and frequency support. The application of the widely used induction machine is found to be not preferable. Beyond acting as supplementary technology for standby power systems, combining directly grid-connected electrical machines with flywheels also offers secondary and tertiary benefits.

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Jauch, Clemens

  • Forschungsprojekt

    • Inno!Nord-HYDRAD - Hydraulischer Schwungradspeicher

  • Abteilung/Einrichtung

    • Wind Energy Technology Institute

  • Profilbereich

    • Erneuerbare Energien und nachhaltige Entwicklung

  • Forschungsfeld

    • Netzintegration

  • Fachbereich

    • Fachbereich 2: Energy and Life Science

  Filter Tags

  • 2026
  • Journal Article
  • • Arne Rettig
    • Saeed Jahromi
    • Prof. Dr. Clemens Jauch
    • Lisanne Reese
  • • Electrical machine
    • Flywheel
    • Hydraulic variable inertia flywheel
    • Local grid
    • Renewable energy
    • Standby power system
    • Uninterruptible power supply

• Feasibility of confocal laser endomicroscopy as an optical biopsy method for SCC on the auricle : an exploratory study

  Müller-Diesing, F., Sievert, M., Panuganti, B., Aubreville, M., Porsche, N., Hackenberg, S., … Goncalves, M. (2026). Feasibility of confocal laser endomicroscopy as an optical biopsy method for SCC on the auricle : an exploratory study. Oral and Maxillofacial Surgery, 30(1), 21+. http://doi.org/10.1007/s10006-026-01506-y

  Abstract

  anzeigen ▸

  In the resection of cutaneous squamous cell carcinomas (cSCC) of the auricle, the challenge lies in ensuring a safe excision with adequate surgical margins in an aesthetically and functionally sensitive area, while keeping the resulting defect as small as possible. To contribute to achieving this goal, the present study is the first to investigate the potential of confocal laser endomicroscopy (CLE) as a method of optical biopsy for cSCC.

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Aubreville, Marc
    • Porsche, Nils

  • Forschungsprojekt

    • Optical Biopsy of Sinonasal Tumors using Confocal Laser Endomicroscopy: A Clinical and Deep Learning-based Assessment and Visualization

  • Abteilung/Einrichtung

    • FLAIR - Flensburg Artificial Intelligence Research

  • Profilbereich

    • Gesundheit und Life Sciences

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Flurin Müller-Diesing
    • Matti Sievert
    • Bharat Panuganti
    • Marc Aubreville
    • Nils Porsche
    • Stephan Hackenberg
    • Manuel Stöth
    • Agmal Scherzad
    • Jasper Meyer
    • Markus Wirth
    • Miguel Goncalves

• Intratumoral distribution of anisokaryosis in canine cutaneous mast cell tumors

  Bertram, C. A., Wilm, F., Parlak, E., Donovan, T. A., Janout, H., Bolfa, P., … Bartel, A. (2026). Intratumoral distribution of anisokaryosis in canine cutaneous mast cell tumors. Veterinary Pathology, 03009858261465447. http://doi.org/10.1177/03009858261465447 (Original work published Juli 2026)

  Abstract

  anzeigen ▸

  Automated measurements of anisokaryosis in canine cutaneous mast cell tumors (ccMCTs) have been shown to be predictive of survival, but questions remain regarding the intratumoral distribution of anisokaryosis. Whole-slide images of 96 ccMCTs were analyzed with a deep learning-based segmentation algorithm to quantify anisokaryosis using the standard deviation (SD) of the nuclear area. In 35/96 cases, \textgreater5\% of the non-overlapping 256 × 256 µm 2 regions were hotspots (SD ≥11.5 µm 2 ). Regions selected by 7 pathologists within these 35 cases matched hotspots in 32\% of the instances. Outcome analysis (tumor-related death) based on single tumor regions yielded an area under the curve (AUC) of 0.901 for pathologist-selected hotspots, falling between random region selection (AUC: 0.862) and 90th-percentile targeted selection (AUC: 0.956). Whole-slide analysis of the hotspot proportion predicted survival with an AUC of 0.956, with 20\% of hotspots as a prognostically meaningful threshold. Whereas pathologists-selected tumor regions are prognostically meaningful for nuclear morphometry, whole-slide analysis may provide additional prognostic information.

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Aubreville, Marc

  • Forschungsprojekt

    • (mu)ROMI: Robust and Accurate Multi-Tumor, Multi-Species, Multi-Laboratory and Multi-Scanner Mitosis Detection with Large-Scale Datasets and Artificial Intelligence

  • Abteilung/Einrichtung

    • FLAIR - Flensburg Artificial Intelligence Research

  • Profilbereich

    • Gesundheit und Life Sciences

  • Forschungsfeld

    • Data Science und Künstliche Intelligenz

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Christof Bertram
    • Frauke Wilm
    • Eda Parlak
    • Taryn Donovan
    • Hannah Janout
    • Pompei Bolfa
    • Michael Dark
    • Andrea Fuchs-Baumgartinger
    • Andrea Klang
    • Robert Klopfleisch
    • Barbara Richter
    • Marc Aubreville
    • Stephan Winkler
    • Matti Kiupel
    • Alexander Bartel

• Benchmarking Deep Learning and Vision Foundation Models for Atypical vs. Normal Mitosis Classification with Cross-Dataset Evaluation

  Banerjee, S., Weiss, V., Donovan, T. A., Fick, R., Conrad, T., Ammeling, J., … Bertram, C. A. (2026). Benchmarking Deep Learning and Vision Foundation Models for Atypical vs. Normal Mitosis Classification with Cross-Dataset Evaluation. Machine Learning for Biomedical Imaging, 3, 115–125. http://doi.org/https://doi.org/10.59275/j.melba.2026-6c1g

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Banerjee, Sweta
    • Porsche, Nils
    • Aubreville, Marc

  • Forschungsprojekt

    • (mu)ROMI: Robust and Accurate Multi-Tumor, Multi-Species, Multi-Laboratory and Multi-Scanner Mitosis Detection with Large-Scale Datasets and Artificial Intelligence

  • Abteilung/Einrichtung

    • FLAIR - Flensburg Artificial Intelligence Research

  • Profilbereich

    • Informatik und Gestaltung

  • Forschungsfeld

    • Data Science und Künstliche Intelligenz

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Sweta Banerjee
    • Viktoria Weiss
    • Taryn Donovan
    • Rutger Fick
    • Thomas Conrad
    • Jonas Ammeling
    • Nils Porsche
    • Robert Klopfleisch
    • Christopher Kaltenecker
    • Katharina Breininger
    • Marc Aubreville
    • Christof Bertram

• Data set creation for supervised deep learning–based analysis of microscopic images: Review of important considerations and recommendations

  Bertram, C. A., Weiss, V., Ammeling, J., Schabel, M., Donovan, T. A., Wilm, F., … Aubreville, M. (2026). Data set creation for supervised deep learning–based analysis of microscopic images: Review of important considerations and recommendations. Veterinary Pathology, 03009858261457959. http://doi.org/10.1177/03009858261457959

  Abstract

  anzeigen ▸

  Supervised deep learning (DL) receives great interest for automated analysis of microscopic images with an increasing body of literature supporting its potential. The development and testing of those DL models rely heavily on the availability of high-quality, large-scale data sets. However, creating such data sets is a complex and resource-intensive process, often hindered by challenges such as time constraints, domain variability, and risks of bias in image collection and label creation. This review provides a comprehensive guide to the critical steps in data set creation, including (1) image acquisition, (2) selection of annotation software, and (3) annotation creation. For image acquisition, besides ensuring a sufficiently large number, it is important to address sources of image variability (domain shifts), such as those related to slide preparation and digitization, that could lead to algorithmic errors if not adequately represented in the training data. For annotations, key quality criteria are the 3 “C”s: correctness, consistency, and completeness. For mitigation of annotation bias of a single annotator, this review explores advanced annotation methods (eg, computer-assisted annotations). To support data set creators, a standard operating procedure is provided as supplemental material, summarizing all important considerations for data set creation. Furthermore, this article underscores the importance of open data sets in driving innovation and enhancing reproducibility of DL research. By addressing the challenges and offering practical recommendations, this review aims to advance the creation and availability of high-quality, large-scale data sets, ultimately contributing to the development of generalizable and robust DL models for pathology applications.

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Aubreville, Marc

  • Abteilung/Einrichtung

    • FLAIR - Flensburg Artificial Intelligence Research

  • Profilbereich

    • Informatik und Gestaltung

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Christof Bertram
    • Viktoria Weiss
    • Jonas Ammeling
    • Maria Schabel
    • Taryn Donovan
    • Frauke Wilm
    • Christian Marzahl
    • Katharina Breininger
    • Marc Aubreville

• Reporting Transparency in Veterinary Pathology Deep Learning: A Systematic Review of Reproducibility-Critical Details

  Banerjee, S., Bertram, C. A., Weiss, V., Ammeling, J., Conrad, T., Porsche, N., … Aubreville, M. (2026). Reporting Transparency in Veterinary Pathology Deep Learning: A Systematic Review of Reproducibility-Critical Details. Veterinary Pathology. http://doi.org/10.1177/03009858261459452

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Banerjee, Sweta
    • Porsche, Nils
    • Aubreville, Marc

  • Forschungsprojekt

    • (mu)ROMI: Robust and Accurate Multi-Tumor, Multi-Species, Multi-Laboratory and Multi-Scanner Mitosis Detection with Large-Scale Datasets and Artificial Intelligence

  • Abteilung/Einrichtung

    • FLAIR - Flensburg Artificial Intelligence Research

  • Profilbereich

    • Informatik und Gestaltung

  • Forschungsfeld

    • Data Science und Künstliche Intelligenz

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Sweta Banerjee
    • Christof Bertram
    • Viktoria Weiss
    • Jonas Ammeling
    • Thomas Conrad
    • Nils Porsche
    • Robert Klopfleisch
    • Christoph Stoblberger
    • Christopher Kaltenecker
    • Katharina Breininger
    • Marc Aubreville

• Green hydrogen ambitions in Algeria and Morocco: contrasting conditions, relational paths

  Boukhatem, I., Blohm, M., & Furnaro, A. (2026). Green hydrogen ambitions in Algeria and Morocco: contrasting conditions, relational paths. Renewable and Sustainable Energy Transition, 9. http://doi.org/https://doi.org/10.1016/j.rset.2026.100150

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Abteilung/Einrichtung

    • Wind Energy Technology Institute

  • Profilbereich

    • Erneuerbare Energien und nachhaltige Entwicklung

  • Forschungsfeld

    • Wasserstoff

  • Fachbereich

    • Fachbereich 2: Energy and Life Science

  Filter Tags

  • 2026
  • Journal Article
  • • Imane Boukhatem
    • Marina Blohm
    • Andrea Furnaro

• A Subphase-Labeled Mitotic Dataset for AI-powered Cell Division Analysis

  Ivan, Z., Hirling, D., Grexa, I., Ammeling, J., Molnar, C., Micsik, T., … Horvath, P. (2026). A Subphase-Labeled Mitotic Dataset for AI-powered Cell Division Analysis. Scientific Data. http://doi.org/10.1038/s41597-026-07007-7

  Abstract

  anzeigen ▸

  Abstract Mitosis detection represents a critical task in digital pathology, as it plays an important role in the tumor grading and prognosis of patients. Manual determination is a labor-intensive task for practitioners with high interobserver variability, thus, automation is a priority. There has been substantial progress towards creating robust mitosis detection algorithms, primarily driven by the Mitosis Domain Generalization (MIDOG) challenges. Also, there has been growing interest in the molecular characterization of mitosis to achieve a more comprehensive understanding of its underlying mechanisms in a subphase-specific manner. We introduce a new mitotic figure dataset annotated with subphase information based on the MIDOG++ dataset as well as a previously unrepresented tumor domain to enhance the diversity and applicability. We envision a new perspective for domain generalization by improving model performance with subtyping mitosis, complemented with an atypical mitotic class. Our work has implications in two main areas: subtyping information can provide helpful information in mitosis detection, while also providing promising new directions in answering biological questions, such as molecular analysis of subphases.

  • Suche auf Google Scholar
  • DOI Verweis

  Exportformate

  • BibTeX
  • EndNote X3 XML
  • EndNote 7 XML
  • EndNote tagged
  • RIS

  Weiterführende Links

  anzeigen ▸

  • Personenseiten

    • Aubreville, Marc

  • Abteilung/Einrichtung

    • FLAIR - Flensburg Artificial Intelligence Research

  • Profilbereich

    • Gesundheit und Life Sciences

  • Forschungsfeld

    • Data Science und Künstliche Intelligenz

  • Fachbereich

    • Fachbereich 3: Information und Kommunikation

  Filter Tags

  • 2026
  • Journal Article
  • • Zsanett Ivan
    • Dominik Hirling
    • Istvan Grexa
    • Jonas Ammeling
    • Csaba Molnar
    • Tamas Micsik
    • Katalin Dobra
    • Levente Kuthi
    • Farkas Sukosd
    • Janos Fillinger
    • Judit Moldvay
    • Erika Toth
    • Marc Aubreville
    • Vivien Miczan
    • Peter Horvath