@unpublished{QiuAubrevilleWilmetal.2024, author = {Qiu, Jingna and Aubreville, Marc and Wilm, Frauke and {\"O}ttl, Mathias and Utz, Jonas and Schlereth, Maja and Breininger, Katharina}, title = {Leveraging Image Captions for Selective Whole Slide Image Annotation}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2407.06363}, year = {2024}, language = {en} } @article{GlahnHaghoferDonovanetal.2024, author = {Glahn, Imaine and Haghofer, Andreas and Donovan, Taryn and Degasperi, Brigitte and Bartel, Alexander and Kreilmeier-Berger, Theresa and Hyndman, Philip S. and Janout, Hannah and Assenmacher, Charles-Antoine and Bartenschlager, Florian and Bolfa, Pompei and Dark, Michael and Klang, Andrea and Klopfleisch, Robert and Merz, Sophie and Richter, Barbara and Schulman, F. Yvonne and Ganz, Jonathan and Scharinger, Josef and Aubreville, Marc and Winkler, Stephan M. and Bertram, Christof}, title = {Automated Nuclear Morphometry: A Deep Learning Approach for Prognostication in Canine Pulmonary Carcinoma to Enhance Reproducibility}, volume = {11}, pages = {278}, journal = {Veterinary Sciences}, number = {6}, publisher = {MDPI}, address = {Basel}, issn = {2306-7381}, doi = {https://doi.org/10.3390/vetsci11060278}, year = {2024}, abstract = {The integration of deep learning-based tools into diagnostic workflows is increasingly prevalent due to their efficiency and reproducibility in various settings. We investigated the utility of automated nuclear morphometry for assessing nuclear pleomorphism (NP), a criterion of malignancy in the current grading system in canine pulmonary carcinoma (cPC), and its prognostic implications. We developed a deep learning-based algorithm for evaluating NP (variation in size, i.e., anisokaryosis and/or shape) using a segmentation model. Its performance was evaluated on 46 cPC cases with comprehensive follow-up data regarding its accuracy in nuclear segmentation and its prognostic ability. Its assessment of NP was compared to manual morphometry and established prognostic tests (pathologists' NP estimates (n = 11), mitotic count, histological grading, and TNM-stage). The standard deviation (SD) of the nuclear area, indicative of anisokaryosis, exhibited good discriminatory ability for tumor-specific survival, with an area under the curve (AUC) of 0.80 and a hazard ratio (HR) of 3.38. The algorithm achieved values comparable to manual morphometry. In contrast, the pathologists' estimates of anisokaryosis resulted in HR values ranging from 0.86 to 34.8, with slight inter-observer reproducibility (k = 0.204). Other conventional tests had no significant prognostic value in our study cohort. Fully automated morphometry promises a time-efficient and reproducible assessment of NP with a high prognostic value. Further refinement of the algorithm, particularly to address undersegmentation, and application to a larger study population are required.}, language = {en} } @article{GanzAmmelingJabarietal.2024, author = {Ganz, Jonathan and Ammeling, Jonas and Jabari, Samir and Breininger, Katharina and Aubreville, Marc}, title = {Re-identification from histopathology images}, volume = {2025}, pages = {103335}, journal = {Medical Image Analysis}, number = {99}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1361-8423}, doi = {https://doi.org/10.1016/j.media.2024.103335}, year = {2024}, abstract = {In numerous studies, deep learning algorithms have proven their potential for the analysis of histopathology images, for example, for revealing the subtypes of tumors or the primary origin of metastases. These models require large datasets for training, which must be anonymized to prevent possible patient identity leaks. This study demonstrates that even relatively simple deep learning algorithms can re-identify patients in large histopathology datasets with substantial accuracy. In addition, we compared a comprehensive set of state-of-the-art whole slide image classifiers and feature extractors for the given task. We evaluated our algorithms on two TCIA datasets including lung squamous cell carcinoma (LSCC) and lung adenocarcinoma (LUAD). We also demonstrate the algorithm's performance on an in-house dataset of meningioma tissue. We predicted the source patient of a slide with 𝐹1 scores of up to 80.1\% and 77.19\% on the LSCC and LUAD datasets, respectively, and with 77.09\% on our meningioma dataset. Based on our findings, we formulated a risk assessment scheme to estimate the risk to the patient's privacy prior to publication.}, language = {en} } @article{AubrevilleGanzAmmelingetal.2024, author = {Aubreville, Marc and Ganz, Jonathan and Ammeling, Jonas and Rosbach, Emely and Gehrke, Thomas and Scherzad, Agmal and Hackenberg, Stephan and Goncalves, Miguel}, title = {Prediction of tumor board procedural recommendations using large language models}, volume = {282}, journal = {European Archives of Oto-Rhino-Laryngology}, number = {3}, publisher = {Springer}, address = {Berlin}, issn = {1434-4726}, doi = {https://doi.org/10.1007/s00405-024-08947-9}, pages = {1619 -- 1629}, year = {2024}, language = {en} } @article{FrenkenSievertPanugantietal.2024, author = {Frenken, Ann-Kathrin and Sievert, Matti and Panuganti, Bharat and Aubreville, Marc and Meyer, Till and Scherzad, Agmal and Gehrke, Thomas and Scheich, Matthias and Hackenberg, Stephan and Goncalves, Miguel}, title = {Feasibility of Optical Biopsy During Endoscopic Sinus Surgery With Confocal Laser Endomicroscopy: A Pilot Study}, volume = {134}, journal = {The Laryngoscope}, number = {10}, publisher = {Wiley}, address = {Malden}, issn = {1531-4995}, doi = {https://doi.org/10.1002/lary.31503}, pages = {4217 -- 4224}, year = {2024}, language = {en} } @inproceedings{AmmelingHeckerGanzetal.2024, author = {Ammeling, Jonas and Hecker, Moritz and Ganz, Jonathan and Donovan, Taryn and Klopfleisch, Robert and Bertram, Christof and Breininger, Katharina and Aubreville, Marc}, title = {Automated Mitotic Index Calculation via Deep Learning and Immunohistochemistry}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2024: Proceedings, German Conference on Medical Image Computing, Erlangen, March 10-12, 2024}, editor = {Maier, Andreas and Deserno, Thomas Martin and Handels, Heinz and Maier-Hein, Klaus H. and Palm, Christoph and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-44037-4}, doi = {https://doi.org/10.1007/978-3-658-44037-4_37}, pages = {123 -- 128}, year = {2024}, language = {en} } @article{SievertAubrevilleMuelleretal.2024, author = {Sievert, Matti and Aubreville, Marc and Mueller, Sarina Katrin and Eckstein, Markus and Breininger, Katharina and Iro, Heinrich and Goncalves, Miguel}, title = {Diagnosis of malignancy in oropharyngeal confocal laser endomicroscopy using GPT 4.0 with vision}, volume = {281}, journal = {European Archives of Oto-Rhino-Laryngology}, number = {4}, publisher = {Springer}, address = {Berlin}, issn = {1434-4726}, doi = {https://doi.org/10.1007/s00405-024-08476-5}, pages = {2115 -- 2122}, year = {2024}, language = {en} } @article{Aubreville2024, author = {Aubreville, Marc}, title = {Developing Robust AI Applications for Clinical Use: The Special Case of Pathology}, volume = {3}, journal = {Trillium Pathology}, number = {1}, publisher = {Trillium GmbH Medizinischer Fachverlag}, address = {Grafrath}, issn = {2752-1915}, doi = {https://doi.org/10.47184/tp.2024.01.04}, pages = {20 -- 22}, year = {2024}, language = {en} } @inproceedings{PerniasRampasRichteretal.2024, author = {Pernias, Pablo and Rampas, Dominic and Richter, Mats Leon and Pal, Christopher and Aubreville, Marc}, title = {W{\"u}rstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models}, booktitle = {The Twelfth International Conference on Learning Representations (ICLR 2024)}, publisher = {OpenReview}, url = {https://openreview.net/forum?id=gU58d5QeGv}, year = {2024}, language = {en} } @inproceedings{EisenmannReinkeWeruetal.2023, author = {Eisenmann, Matthias and Reinke, Annika and Weru, Vivienn and Tizabi, Minu Dietlinde and Isensee, Fabian and Adler, Tim J. and Ali, Sharib and Andrearczyk, Vincent and Aubreville, Marc and Baid, Ujjwal and Bakas, Spyridon and Balu, Niranjan and Bano, Sophia and Bernal, Jorge and Bodenstedt, Sebastian and Casella, Alessandro and Cheplygina, Veronika and Daum, Marie and De Bruijne, Marleen and Depeursinge, Adrien and Dorent, Reuben and Egger, Jan and Ellis, David G. and Engelhardt, Sandy and Ganz, Melanie and Ghatwary, Noha M. and Girard, Gabriel and Godau, Patrick and Gupta, Anubha and Hansen, Lasse and Harada, Kanako and Heinrich, Mattias and Heller, Nicholas and Hering, Alessa and Huaulm{\´e}, Arnoud and Jannin, Pierre and Kavur, A. Emre and Kodym, Oldrich and Kozubek, Michal and Li, Jianning and Li, Hongwei and Ma, Jun and Mart{\´i}n-Isla, Carlos and Menze, Bjoern H. and Noble, Alison and Oreiller, Valentin and Padoy, Nicolas and Pati, Sarthak and Payette, Kelly and R{\"a}dsch, Tim and Rafael-Pati{\~n}o, Jonathan and Bawa, Vivek Singh and Speidel, Stefanie and Sudre, Carole H. and Van Wijnen, Kimberlin M. H. and Wagner, M. and Wei, D. and Yamlahi, Amine and Yap, Moi Hoon and Yuan, C. and Zenk, Maximilian and Zia, A. and Zimmerer, David and Aydogan, Dogu Baran and Bhattarai, B. and Bloch, Louise and Br{\"u}ngel, Raphael and Cho, J. and Choi, C. and Dou, Q. and Ezhov, Ivan and Friedrich, Christoph M. and Fuller, C. and Gaire, Rebati Raman and Galdran, Adrian and Garc{\´i}a-Faura, {\´A}lvaro and Grammatikopoulou, Maria and Hong, S. and Jahanifar, Mostafa and Jang, I. and Kadkhodamohammadi, Abdolrahim and Kang, I. and Kofler, Florian and Kondo, Satoshi and Kuijf, Hugo Jaco and Li, M. and Luu, M. and Martinčič, Tomaz and Morais, P. and Naser, M. A. and Oliveira, B. and Owen, D. and Pang, S. and Park, Jinah and Park, S. and Płotka, S. and Puybareau, {\´E}lodie and Rajpoot, Nasir M. and Ryu, K. and Saeed, N. and Shephard, Adam and Shi, P. and Štepec, Dejan and Subedi, Ronast and Tochon, Guillaume and Torres, Helena R. and Urien, H{\´e}l{\`e}ne and Vila{\c{c}}a, Jo{\~a}o L. and Wahid, Kareem A. and Wang, H. and Wang, J. and Wang, L. and Wang, Xiyue and Wiestler, Benedikt and Wodzinski, Marek and Xia, F. and Xie, J. and Xiong, Z. and Yang, Sen and Yang, Y. and Zhao, Z. and Maier-Hein, Klaus H. and J{\"a}ger, Paul F. and Kopp-Schneider, Annette and Maier-Hein, Lena}, title = {Why is the Winner the Best?}, booktitle = {Proceedings: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition}, publisher = {IEEE}, address = {Los Alamitos}, isbn = {979-8-3503-0129-8}, issn = {2575-7075}, doi = {https://doi.org/10.1109/CVPR52729.2023.01911}, pages = {19955 -- 19967}, year = {2023}, language = {en} } @article{WilmIhlingMehesetal.2023, author = {Wilm, Frauke and Ihling, Christian and M{\´e}hes, G{\´a}bor and Terracciano, Luigi and Puget, Chlo{\´e} and Klopfleisch, Robert and Sch{\"u}ffler, Peter and Aubreville, Marc and Maier, Andreas and Mrowiec, Thomas and Breininger, Katharina}, title = {Pan-tumor T-lymphocyte detection using deep neural networks: Recommendations for transfer learning in immunohistochemistry}, volume = {2023}, pages = {100301}, journal = {Journal of Pathology Informatics}, number = {14}, publisher = {Elsevier}, address = {Amsterdam}, issn = {2153-3539}, doi = {https://doi.org/10.1016/j.jpi.2023.100301}, year = {2023}, abstract = {The success of immuno-oncology treatments promises long-term cancer remission for an increasing number of patients. The response to checkpoint inhibitor drugs has shown a correlation with the presence of immune cells in the tumor and tumor microenvironment. An in-depth understanding of the spatial localization of immune cells is therefore critical for understanding the tumor's immune landscape and predicting drug response. Computer-aided systems are well suited for efficiently quantifying immune cells in their spatial context. Conventional image analysis approaches are often based on color features and therefore require a high level of manual interaction. More robust image analysis methods based on deep learning are expected to decrease this reliance on human interaction and improve the reproducibility of immune cell scoring. However, these methods require sufficient training data and previous work has reported low robustness of these algorithms when they are tested on out-of-distribution data from different pathology labs or samples from different organs. In this work, we used a new image analysis pipeline to explicitly evaluate the robustness of marker-labeled lymphocyte quantification algorithms depending on the number of training samples before and after being transferred to a new tumor indication. For these experiments, we adapted the RetinaNet architecture for the task of T-lymphocyte detection and employed transfer learning to bridge the domain gap between tumor indications and reduce the annotation costs for unseen domains. On our test set, we achieved human-level performance for almost all tumor indications with an average precision of 0.74 in-domain and 0.72-0.74 cross-domain. From our results, we derive recommendations for model development regarding annotation extent, training sample selection, and label extraction for the development of robust algorithms for immune cell scoring. By extending the task of marker-labeled lymphocyte quantification to a multi-class detection task, the pre-requisite for subsequent analyses, e.g., distinguishing lymphocytes in the tumor stroma from tumor-infiltrating lymphocytes, is met.}, language = {en} } @article{FragosoGarciaWilmBertrametal.2023, author = {Fragoso-Garcia, Marco and Wilm, Frauke and Bertram, Christof and Merz, Sophie and Schmidt, Anja and Donovan, Taryn and Fuchs-Baumgartinger, Andrea and Bartel, Alexander and Marzahl, Christian and Diehl, Laura and Puget, Chloe and Maier, Andreas and Aubreville, Marc and Breininger, Katharina and Klopfleisch, Robert}, title = {Automated diagnosis of 7 canine skin tumors using machine learning on H\&E-stained whole slide images}, volume = {60}, journal = {Veterinary Pathology}, number = {6}, publisher = {Sage}, address = {London}, issn = {0300-9858}, doi = {https://doi.org/10.1177/03009858231189205}, pages = {865 -- 875}, year = {2023}, abstract = {Microscopic evaluation of hematoxylin and eosin-stained slides is still the diagnostic gold standard for a variety of diseases, including neoplasms. Nevertheless, intra- and interrater variability are well documented among pathologists. So far, computer assistance via automated image analysis has shown potential to support pathologists in improving accuracy and reproducibility of quantitative tasks. In this proof of principle study, we describe a machine-learning-based algorithm for the automated diagnosis of 7 of the most common canine skin tumors: trichoblastoma, squamous cell carcinoma, peripheral nerve sheath tumor, melanoma, histiocytoma, mast cell tumor, and plasmacytoma. We selected, digitized, and annotated 350 hematoxylin and eosin-stained slides (50 per tumor type) to create a database divided into training, n = 245 whole-slide images (WSIs), validation ( n = 35 WSIs), and test sets ( n = 70 WSIs). Full annotations included the 7 tumor classes and 6 normal skin structures. The data set was used to train a convolutional neural network (CNN) for the automatic segmentation of tumor and nontumor classes. Subsequently, the detected tumor regions were classified patch-wise into 1 of the 7 tumor classes. A majority of patches-approach led to a tumor classification accuracy of the network on the slide-level of 95\% (133/140 WSIs), with a patch-level precision of 85\%. The same 140 WSIs were provided to 6 experienced pathologists for diagnosis, who achieved a similar slide-level accuracy of 98\% (137/140 correct majority votes). Our results highlight the feasibility of artificial intelligence-based methods as a support tool in diagnostic oncologic pathology with future applications in other species and tumor types.}, language = {en} } @inproceedings{AmmelingMangerKwakaetal.2023, author = {Ammeling, Jonas and Manger, Carina and Kwaka, Elias and Kr{\"u}gel, Sebastian and Uhl, Matthias and Kießig, Angelika and Fritz, Alexis and Ganz, Jonathan and Riener, Andreas and Bertram, Christof and Breininger, Katharina and Aubreville, Marc}, title = {Appealing but Potentially Biasing - Investigation of the Visual Representation of Segmentation Predictions by AI Recommender Systems for Medical Decision Making}, booktitle = {Mensch und Computer 2023: Building Bridges: Tagungsband (Proceedings)}, editor = {Stolze, Markus and Loch, Frieder and Baldauf, Matthias and Alt, Florian and Schneegass, Christina and Kosch, Thomas and Hirzle, Teresa and Sadeghian, Shadan and Draxler, Fiona and Bektas, Kenan and Lohan, Katrin and Knierim, Pascal}, publisher = {ACM}, address = {New York}, isbn = {979-8-4007-0771-1}, doi = {https://doi.org/10.1145/3603555.3608561}, pages = {330 -- 335}, year = {2023}, language = {en} } @article{KruegelAmmelingAubrevilleetal.2024, author = {Kr{\"u}gel, Sebastian and Ammeling, Jonas and Aubreville, Marc and Fritz, Alexis and Kießig, Angelika and Uhl, Matthias}, title = {Perceived responsibility in AI-supported medicine}, volume = {40}, journal = {AI \& Society: Journal of Knowledge, Culture and Communication}, publisher = {Springer}, address = {London}, issn = {1435-5655}, doi = {https://doi.org/10.1007/s00146-024-01972-6}, pages = {1485 -- 1495}, year = {2024}, abstract = {In a representative vignette study in Germany with 1,653 respondents, we investigated laypeople's attribution of moral responsibility in collaborative medical diagnosis. Specifically, we compare people's judgments in a setting in which physicians are supported by an AI-based recommender system to a setting in which they are supported by a human colleague. It turns out that people tend to attribute moral responsibility to the artificial agent, although this is traditionally considered a category mistake in normative ethics. This tendency is stronger when people believe that AI may become conscious at some point. In consequence, less responsibility is attributed to human agents in settings with hybrid diagnostic teams than in settings with human-only diagnostic teams. Our findings may have implications for behavior exhibited in contexts of collaborative medical decision making with AI-based as opposed to human recommenders because less responsibility is attributed to agents who have the mental capacity to care about outcomes.}, language = {en} } @inproceedings{AmmelingWilmGanzetal.2023, author = {Ammeling, Jonas and Wilm, Frauke and Ganz, Jonathan and Breininger, Katharina and Aubreville, Marc}, title = {Reference Algorithms for the Mitosis Domain Generalization (MIDOG) 2022 Challenge}, booktitle = {Mitosis Domain Generalization and Diabetic Retinopathy Analysis}, editor = {Sheng, Bin and Aubreville, Marc}, publisher = {Springer}, address = {Cham}, isbn = {978-3-031-33658-4}, doi = {https://doi.org/10.1007/978-3-031-33658-4_19}, pages = {201 -- 205}, year = {2023}, language = {en} } @article{SievertAubrevilleGostianetal.2022, author = {Sievert, Matti and Aubreville, Marc and Gostian, Antoniu-Oreste and Mantsopoulos, Konstantinos and Koch, Michael and M{\"u}ller, Sarina K. and Eckstein, Markus and Rupp, Robin and Stelzle, Florian and Oetter, Nicolai and Maier, Andreas and Iro, Heinrich and Goncalves, Miguel}, title = {Validity of tissue homogeneity in confocal laser endomicroscopy on the diagnosis of laryngeal and hypopharyngeal squamous cell carcinoma}, volume = {279}, journal = {European Archives of Oto-Rhino-Laryngology and Head \& Neck}, number = {8}, publisher = {Springer Nature}, address = {Cham}, issn = {1434-4726}, doi = {https://doi.org/10.1007/s00405-022-07304-y}, pages = {4147 -- 4156}, year = {2022}, abstract = {Purpose Confocal laser endomicroscopy (CLE) allows imaging of the laryngeal mucosa in a thousand-fold magnification. This study analyzes differences in tissue homogeneity between healthy mucosa and squamous cell carcinoma (SCC) via CLE. Materials and methods We included five SCC patients with planned total laryngectomy in this study between October 2020 and February 2021. We captured CLE scans of the tumor and healthy mucosa. Analysis of image homogeneity to diagnose SCC was performed by measuring the signal intensity in four regions of interest (ROI) in each frame in a total of 60 sequences. Each sequence was assigned to the corresponding histological pattern, derived from hematoxylin and eosin staining. In addition, we recorded the subjective evaluation of seven investigators regarding tissue homogeneity. Results Out of 3600 images, 1620 (45\%) correlated with benign mucosa and 1980 (55\%) with SCC. ROIs of benign mucosa and SCC had a mean and standard deviation (SD) of signal intensity of, respectively, 232.1 ± 3.34 and 467.3 ± 9.72 (P < 0.001). The mean SD between the four different ROIs was 39.1 ± 1.03 for benign and 101.5 ± 2.6 for SCC frames (P < 0.001). In addition, homogeneity yielded a sensitivity and specificity of 81.8\% and 86.2\%, respectively, regarding the investigator-dependent analysis. Conclusions SCC shows a significant tissue inhomogeneity in comparison to the healthy epithelium. The results support this feature's importance in identifying malignant mucosa areas during CLE examination. However, the examiner-dependent evaluation emphasizes that homogeneity is a sub-criterion that must be considered in a broad context.}, language = {en} } @article{SievertMantsopoulosMuelleretal.2021, author = {Sievert, Matti and Mantsopoulos, Konstantinos and M{\"u}ller, Sarina K. and Rupp, Robin and Eckstein, Markus and Stelzle, Florian and Oetter, Nicolai and Maier, Andreas and Aubreville, Marc and Iro, Heinrich and Goncalves, Miguel}, title = {Validation of a classification and scoring system for the diagnosis of laryngeal and pharyngeal squamous cell carcinomas by confocal laser endomicroscopy}, volume = {2022}, journal = {Brazilian Journal of Otorhinolaryngology}, number = {88, S4}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1808-8694}, doi = {https://doi.org/10.1016/j.bjorl.2021.06.002}, pages = {26 -- 32}, year = {2021}, abstract = {Introduction Confocal laser endomicroscopy is an optical imaging technique that allows in vivo, real-time, microscope-like images of the upper aerodigestive tract's mucosa. The assessment of morphological tissue characteristics for the correct differentiation between healthy and malignant suspected mucosa requires strict evaluation criteria. Objective This study aims to validate an eight-point score for the correct assessment of malignancy. Methods We performed confocal laser endomicroscopy between March and October 2020 in 13 patients. 197 sequences (11.820 images) originated from the marginal area of pharyngeal and laryngeal carcinomas. Specimens were taken at corresponding locations and analyzed in H\&E staining as a standard of reference. A total of six examiners evaluated the sequences based on a scoring system; they were blinded to the histopathological examination. The primary endpoints are sensitivity, specificity, and accuracy. Secondary endpoints are interrater reliability and receiver operator characteristics. Results Healthy mucosa showed epithelium with uniform size and shape with distinct cytoplasmic membranes and regular vessel architecture. Confocal laser endomicroscopy of malignant cells demonstrated a disorganized arrangement of variable cellular morphology. We calculated an accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of 83.2\%, 81.3\%, 85.5\%, 86.7\%, and 79.7\%, respectively, with a κ-value of 0.64, and an area under the curve of 0.86. Conclusion The results confirm that this scoring system is applicable in the laryngeal and pharyngeal mucosa to classify benign and malignant tissue. A scoring system based on defined and reproducible characteristics can help translate this experimental method to broad clinical practice in head and neck diagnosis.}, language = {en} } @article{SievertEcksteinMantsopoulosetal.2021, author = {Sievert, Matti and Eckstein, Markus and Mantsopoulos, Konstantinos and M{\"u}ller, Sarina K. and Stelzle, Florian and Aubreville, Marc and Oetter, Nicolai and Maier, Andreas and Iro, Heinrich and Goncalves, Miguel}, title = {Impact of intraepithelial capillary loops and atypical vessels in confocal laser endomicroscopy for the diagnosis of laryngeal and hypopharyngeal squamous cell carcinoma}, volume = {279}, journal = {European Archives of Oto-Rhino-Laryngology and Head \& Neck}, number = {4}, publisher = {Springer Nature}, address = {Cham}, issn = {1434-4726}, doi = {https://doi.org/10.1007/s00405-021-06954-8}, pages = {2029 -- 2037}, year = {2021}, abstract = {Purpose Confocal laser endomicroscopy (CLE) allows surface imaging of the laryngeal and pharyngeal mucosa in vivo at a thousand-fold magnification. This study aims to compare irregular blood vessels and intraepithelial capillary loops in healthy mucosa and squamous cell carcinoma (SCC) via CLE. Materials and methods We included ten patients with confirmed SCC and planned total laryngectomy in this study between March 2020 and February 2021. CLE images of these patients were collected and compared with the corresponding histology in hematoxylin and eosin staining. We analyzed the characteristic endomicroscopic patterns of blood vessels and intraepithelial capillary loops for the diagnosis of SCC. Results In a total of 54 sequences, we identified 243 blood vessels which were analyzed regarding structure, diameter, and Fluorescein leakage, confirming that irregular, corkscrew-like vessels (24.4\% vs. 1.3\%; P < .001), dilated intraepithelial capillary loops (90.8\% vs. 28.7\%; P < .001), and increased capillary leakage (40.7\% vs. 2.5\%; P < .001), are significantly more frequently detected in SCC compared to the healthy epithelium. We defined a vessel diameter of 30 μm in capillary loops as a cut-off value, obtaining a sensitivity, specificity, PPV, and NPV and accuracy of 90.6\%, 71.3\%, 57.4\%, 94.7\%, and 77.1\%, respectively, for the detection of malignancy based solely on capillary architecture. Conclusion Capillaries within malignant lesions are fundamentally different from those in healthy mucosa regions. The capillary architecture is a significant feature aiding the identification of malignant mucosa areas during in-vivo, real-time CLE examination.}, language = {en} } @inproceedings{AubrevilleGanzAmmelingetal.2023, author = {Aubreville, Marc and Ganz, Jonathan and Ammeling, Jonas and Donovan, Taryn and Fick, Rutger H. J. and Breininger, Katharina and Bertram, Christof}, title = {Deep Learning-based Subtyping of Atypical and Normal Mitoses using a Hierarchical Anchor-free Object Detector}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2023: Proceedings, German Workshop on Medical Image Computing, Braunschweig, July 2-4, 2023}, editor = {Deserno, Thomas Martin and Handels, Heinz and Maier, Andreas and Maier-Hein, Klaus H. and Palm, Christoph and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-41657-7}, doi = {https://doi.org/10.1007/978-3-658-41657-7_40}, pages = {189 -- 195}, year = {2023}, language = {en} } @inproceedings{LausserBertramKlopfleischetal.2023, author = {Lausser, Ludwig and Bertram, Christof and Klopfleisch, Robert and Aubreville, Marc}, title = {Limits of Human Expert Ensembles in Mitosis Multi-expert Ground Truth Generation}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2023: Proceedings, German Workshop on Medical Image Computing, Braunschweig, July 2-4, 2023}, editor = {Deserno, Thomas Martin and Handels, Heinz and Maier, Andreas and Maier-Hein, Klaus H. and Palm, Christoph and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-41657-7}, doi = {https://doi.org/10.1007/978-3-658-41657-7_27}, pages = {116 -- 121}, year = {2023}, language = {en} } @article{AubrevilleStathonikosBertrametal.2022, author = {Aubreville, Marc and Stathonikos, Nikolas and Bertram, Christof and Klopfleisch, Robert and Hoeve, Natalie ter and Ciompi, Francesco and Wilm, Frauke and Marzahl, Christian and Donovan, Taryn and Maier, Andreas and Breen, Jack and Ravikumar, Nishant and Chung, Youjin and Park, Jinah and Nateghi, Ramin and Pourakpour, Fattaneh and Fick, Rutger H. J. and Ben Hadj, Saima and Jahanifar, Mostafa and Shepard, Adam and Dexl, Jakob and Wittenberg, Thomas and Kondo, Satoshi and Lafarge, Maxime W. and Kolezer, Viktor H. and Liang, Jingtang and Wang, Yubo and Long, Xi and Liu, Jingxin and Razavi, Salar and Khademi, April and Yang, Sen and Wang, Xiyue and Erber, Ramona and Klang, Andrea and Lipnik, Karoline and Bolfa, Pompei and Dark, Michael and Wasinger, Gabriel and Veta, Mitko and Breininger, Katharina}, title = {Mitosis domain generalization in histopathology images — The MIDOG challenge}, volume = {2023}, pages = {102699}, journal = {Medical Image Analysis}, number = {84}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1361-8415}, doi = {https://doi.org/10.1016/j.media.2022.102699}, year = {2022}, language = {en} } @article{PanBreiningerAubrevilleetal.2022, author = {Pan, Zhaoya and Breininger, Katharina and Aubreville, Marc and Stelzle, Florian and Oetter, Nicolai and Maier, Andreas and Mantsopoulos, Konstantinos and Iro, Heinrich and Goncalves, Miguel and Sievert, Matti}, title = {Defining a baseline identification of artifacts in confocal laser endomicroscopy in head and neck cancer imaging}, volume = {44}, pages = {103779}, journal = {American Journal of Otolaryngology}, number = {2}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1532-818X}, doi = {https://doi.org/10.1016/j.amjoto.2022.103779}, year = {2022}, language = {en} } @inproceedings{TheelkeWilmMarzahletal.2021, author = {Theelke, Luisa and Wilm, Frauke and Marzahl, Christian and Bertram, Christof and Klopfleisch, Robert and Maier, Andreas and Aubreville, Marc and Breininger, Katharina}, title = {Iterative Cross-Scanner Registration for Whole Slide Images}, booktitle = {2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)}, publisher = {IEEE}, address = {Piscataway}, isbn = {978-1-6654-0191-3}, issn = {2473-9944}, doi = {https://doi.org/10.1109/ICCVW54120.2021.00071}, pages = {582 -- 590}, year = {2021}, language = {en} } @inproceedings{AubrevilleBertramStathonikosetal.2021, author = {Aubreville, Marc and Bertram, Christof and Stathonikos, Nikolas and ter Hoeve, Natalie and Ciompi, Francesco and Klopfleisch, Robert and Veta, Mitko and Donovan, Taryn and Marzahl, Christian and Wilm, Frauke and Breininger, Katharina and Maier, Andreas}, title = {Quantifying the Scanner-Induced Domain Gap in Mitosis Detection}, booktitle = {MIDL: Medical Imaging with Deep Learning 2021}, publisher = {MIDL Foundation}, address = {Nijmegen}, url = {https://2021.midl.io/papers/i6}, year = {2021}, language = {en} } @inproceedings{WilmMarzahlBreiningeretal.2022, author = {Wilm, Frauke and Marzahl, Christian and Breininger, Katharina and Aubreville, Marc}, title = {Domain Adversarial RetinaNet as a Reference Algorithm for the MItosis DOmain Generalization Challenge}, booktitle = {Biomedical Image Registration, Domain Generalisation and Out-of-Distribution Analysis : MICCAI 2021 Challenges}, editor = {Aubreville, Marc and Zimmerer, David and Heinrich, Mattias}, publisher = {Springer}, address = {Cham}, isbn = {978-3-030-97281-3}, doi = {https://doi.org/10.1007/978-3-030-97281-3_1}, pages = {5 -- 13}, year = {2022}, language = {en} } @article{BertramMarzahlBarteletal.2022, author = {Bertram, Christof and Marzahl, Christian and Bartel, Alexander and Stayt, Jason and Bonsembiante, Federico and Beeler-Marfisi, Janet and Barton, Ann K. and Brocca, Ginevra and Gelain, Maria Elena and Gl{\"a}sel, Agnes C. and du Preez, Kelly and Weiler, Kristina and Weissenbacher-Lang, Christiane and Breininger, Katharina and Aubreville, Marc and Maier, Andreas and Klopfleisch, Robert and Hill, Jenny}, title = {Cytologic scoring of equine exercise-induced pulmonary hemorrhage: Performance of human experts and a deep learning-based algorithm}, volume = {60}, journal = {Veterinary Pathology}, number = {1}, publisher = {Sage}, address = {London}, issn = {1544-2217}, doi = {https://doi.org/10.1177/03009858221137582}, pages = {75 -- 85}, year = {2022}, abstract = {Exercise-induced pulmonary hemorrhage (EIPH) is a relevant respiratory disease in sport horses, which can be diagnosed by examination of bronchoalveolar lavage fluid (BALF) cells using the total hemosiderin score (THS). The aim of this study was to evaluate the diagnostic accuracy and reproducibility of annotators and to validate a deep learning-based algorithm for the THS. Digitized cytological specimens stained for iron were prepared from 52 equine BALF samples. Ten annotators produced a THS for each slide according to published methods. The reference methods for comparing annotator's and algorithmic performance included a ground truth dataset, the mean annotators' THSs, and chemical iron measurements. Results of the study showed that annotators had marked interobserver variability of the THS, which was mostly due to a systematic error between annotators in grading the intracytoplasmatic hemosiderin content of individual macrophages. Regarding overall measurement error between the annotators, 87.7\% of the variance could be reduced by using standardized grades based on the ground truth. The algorithm was highly consistent with the ground truth in assigning hemosiderin grades. Compared with the ground truth THS, annotators had an accuracy of diagnosing EIPH (THS of < or ≥ 75) of 75.7\%, whereas, the algorithm had an accuracy of 92.3\% with no relevant differences in correlation with chemical iron measurements. The results show that deep learning-based algorithms are useful for improving reproducibility and routine applicability of the THS. For THS by experts, a diagnostic uncertainty interval of 40 to 110 is proposed. THSs within this interval have insufficient reproducibility regarding the EIPH diagnosis.}, language = {en} } @article{MarzahlAubrevilleBertrametal.2021, author = {Marzahl, Christian and Aubreville, Marc and Bertram, Christof and Maier, Jennifer and Bergler, Christian and Kr{\"o}ger, Christine and Voigt, J{\"o}rn and Breininger, Katharina and Klopfleisch, Robert and Maier, Andreas}, title = {EXACT: a collaboration toolset for algorithm-aided annotation of images with annotation version control}, volume = {11}, pages = {4343}, journal = {Scientific Reports}, publisher = {Springer Nature}, address = {London}, issn = {2045-2322}, doi = {https://doi.org/10.1038/s41598-021-83827-4}, year = {2021}, abstract = {In many research areas, scientific progress is accelerated by multidisciplinary access to image data and their interdisciplinary annotation. However, keeping track of these annotations to ensure a high-quality multi-purpose data set is a challenging and labour intensive task. We developed the open-source online platform EXACT (EXpert Algorithm Collaboration Tool) that enables the collaborative interdisciplinary analysis of images from different domains online and offline. EXACT supports multi-gigapixel medical whole slide images as well as image series with thousands of images. The software utilises a flexible plugin system that can be adapted to diverse applications such as counting mitotic figures with a screening mode, finding false annotations on a novel validation view, or using the latest deep learning image analysis technologies. This is combined with a version control system which makes it possible to keep track of changes in the data sets and, for example, to link the results of deep learning experiments to specific data set versions. EXACT is freely available and has already been successfully applied to a broad range of annotation tasks, including highly diverse applications like deep learning supported cytology scoring, interdisciplinary multi-centre whole slide image tumour annotation, and highly specialised whale sound spectroscopy clustering.}, language = {en} } @article{SievertAubrevilleOetteretal.2021, author = {Sievert, Matti and Aubreville, Marc and Oetter, Nicolai and Stelzle, Florian and Maier, Andreas and Mantsopoulos, Konstantinos and Iro, Heinrich and Goncalves, Miguel}, title = {Konfokale Laser-Endomikroskopie des Kopf-Hals-Plattenepithelkarzinoms: eine systematische {\"U}bersicht}, volume = {100}, journal = {Laryngo-Rhino-Otologie}, number = {11}, publisher = {Thieme}, address = {Stuttgart}, issn = {0935-8943}, doi = {https://doi.org/10.1055/a-1339-1635}, pages = {875 -- 881}, year = {2021}, language = {de} } @article{SievertStelzleAubrevilleetal.2021, author = {Sievert, Matti and Stelzle, Florian and Aubreville, Marc and M{\"u}ller, Sarina K. and Eckstein, Markus and Oetter, Nicolai and Maier, Andreas and Mantsopoulos, Konstantinos and Iro, Heinrich and Goncalves, Miguel}, title = {Intraoperative free margins assessment of oropharyngeal squamous cell carcinoma with confocal laser endomicroscopy: a pilot study}, volume = {278}, journal = {European Archives of Oto-Rhino-Laryngology and Head \& Neck}, number = {11}, publisher = {Springer Nature}, address = {Cham}, issn = {1434-4726}, doi = {https://doi.org/10.1007/s00405-021-06659-y}, pages = {4433 -- 4439}, year = {2021}, abstract = {Purpose This pilot study aimed to assess the feasibility of intraoperative assessment of safe margins with confocal laser endomicroscopy (CLE) during oropharyngeal squamous cell carcinoma (OPSCC) surgery. Methods We included five consecutive patients confirmed OPSCC and planned tumor resection in September and October 2020. Healthy appearing mucosa in the marginal zone, and the tumor margin, were examined with CLE and biopsy during tumor resection. A total of 12,809 CLE frames were correlated with the gold standard of hematoxylin and eosin staining. Three head and neck surgeons and one pathologist were asked to identify carcinoma in a sample of 169 representative images, blinded to the histological results. Results Healthy mucosa showed epithelium with uniform size and shape with distinct cytoplasmic membranes and regular vessel architecture. CLE optical biopsy of OPSCC demonstrated a disorganized arrangement of variable cellular morphology. We calculated an accuracy, sensitivity, specificity, PPV, and NPV of 86\%, 90\%, 79\%, 88\%, and 82\%, respectively, with inter-rater reliability and κ-value of 0.60. Conclusion CLE can be easily integrated into the intraoperative setting, generate real-time, in-vivo microscopic images of the oropharynx for evaluation and demarcation of cancer. It can eventually contribute to a less radical approach by enabling a more precise evaluation of the cancer margin.}, language = {en} } @article{MeutenMooreDonovanetal.2021, author = {Meuten, Donald J. and Moore, Frances M. and Donovan, Taryn and Bertram, Christof and Klopfleisch, Robert and Foster, Robert A. and Smedley, Rebecca C. and Dark, Michael and Milovancev, Milan and Stromberg, Paul and Williams, Bruce H. and Aubreville, Marc and Avallone, Giancarlo and Bolfa, Pompei and Cullen, John and Dennis, Michelle M. and Goldschmidt, Michael and Luong, Richard and Miller, Andrew D. and Miller, Margaret A. and Munday, John S. and Roccabianca, Paola and Salas, Elisa N. and Schulman, F. Yvonne and Laufer-Amorim, Renee and Asakawa, Midori G. and Craig, Linden and Dervisis, Nick and Esplin, D. Glen and George, Jeanne W. and Hauck, Marlene and Kagawa, Yumiko and Kiupel, Matti and Linder, Keith and Meichner, Kristina and Marconato, Laura and Oblak, Michelle L. and Santos, Renato L. and Simpson, R. Mark and Tvedten, Harold and Whitley, Derick}, title = {International Guidelines for Veterinary Tumor Pathology: A Call to Action}, volume = {58}, journal = {Veterinary Pathology}, number = {5}, publisher = {Sage}, address = {London}, issn = {1544-2217}, doi = {https://doi.org/10.1177/03009858211013712}, pages = {766 -- 794}, year = {2021}, language = {en} } @inproceedings{WilmBertramMarzahletal.2021, author = {Wilm, Frauke and Bertram, Christof and Marzahl, Christian and Bartel, Alexander and Donovan, Taryn and Assenmacher, Charles-Antoine and Becker, Kathrin and Bennett, Mark and Corner, Sarah M. and Cossic, Brieuc and Denk, Daniela and Dettwiler, Martina and Garcia Gonzalez, Beatriz and Gurtner, Corinne and Heier, Annabelle and Lehmbecker, Annika and Merz, Sophie and Plog, Stephanie and Schmidt, Anja and Sebastian, Franziska and Smedley, Rebecca C. and Tecilla, Marco and Thaiwong, Tuddow and Breininger, Katharina and Kiupel, Matti and Maier, Andreas and Klopfleisch, Robert and Aubreville, Marc}, title = {Influence of inter-annotator variability on automatic mitotic figure assessment}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2021}, publisher = {Springer}, address = {Wiesbaden}, isbn = {978-3-658-33198-6}, doi = {https://doi.org/10.1007/978-3-658-33198-6_56}, pages = {241 -- 246}, year = {2021}, language = {en} } @inproceedings{BertramDonovanTecillaetal.2021, author = {Bertram, Christof and Donovan, Taryn and Tecilla, Marco and Bartenschlager, Florian and Fragoso-Garcia, Marco and Wilm, Frauke and Marzahl, Christian and Breininger, Katharina and Maier, Andreas and Klopfleisch, Robert and Aubreville, Marc}, title = {Dataset on bi- and multi-nucleated tumor cells in canine cutaneous mast cell tumors}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2021: Proceedings, German Workshop on Medical Image Computing, Regensburg, March 7-9, 2021}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-33197-9}, issn = {1431-472X}, doi = {https://doi.org/10.1007/978-3-658-33198-6_33}, pages = {134 -- 139}, year = {2021}, language = {en} } @inproceedings{MarzahlBertramWilmetal.2021, author = {Marzahl, Christian and Bertram, Christof and Wilm, Frauke and Voigt, J{\"o}rn and Barton, Ann K. and Klopfleisch, Robert and Breininger, Katharina and Maier, Andreas and Aubreville, Marc}, title = {Cell detection for asthma on partially annotated whole slide images}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2021: Proceedings, German Workshop on Medical Image Computing, Regensburg, March 7-9, 2021}, subtitle = {learning to be EXACT}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-33197-9}, issn = {1431-472X}, doi = {https://doi.org/10.1007/978-3-658-33198-6_36}, pages = {147 -- 152}, year = {2021}, language = {en} } @article{AmmelingGanzWilmetal.2025, author = {Ammeling, Jonas and Ganz, Jonathan and Wilm, Frauke and Breininger, Katharina and Aubreville, Marc}, title = {Investigation of Class Separability within Object Detection Models in Histopathology}, volume = {44}, journal = {IEEE Transactions on Medical Imaging}, number = {8}, publisher = {IEEE}, address = {New York}, issn = {0278-0062}, doi = {https://doi.org/10.1109/TMI.2025.3560134}, pages = {3162 -- 3174}, year = {2025}, language = {en} } @article{OetterProellSievertetal.2024, author = {Oetter, Nicolai and Pr{\"o}ll, Jonas and Sievert, Matti and Goncalves, Miguel and Rohde, Maximilian and Nobis, Christopher-Philipp and Knipfer, Christian and Aubreville, Marc and Pan, Zhaoya and Breininger, Katharina and Maier, Andreas and Kesting, Marco and Stelzle, Florian}, title = {Oral mucosa - an examination map for confocal laser endomicroscopy within the oral cavity: an experimental clinical study}, volume = {28}, pages = {266}, journal = {Clinical Oral Investigations}, number = {5}, publisher = {Springer}, address = {Berlin}, issn = {1436-3771}, doi = {https://doi.org/10.1007/s00784-024-05664-9}, year = {2024}, abstract = {Objectives Confocal laser endomicroscopy (CLE) is an optical method that enables microscopic visualization of oral mucosa. Previous studies have shown that it is possible to differentiate between physiological and malignant oral mucosa. However, differences in mucosal architecture were not taken into account. The objective was to map the different oral mucosal morphologies and to establish a "CLE map" of physiological mucosa as baseline for further application of this powerful technology. Materials and methods The CLE database consisted of 27 patients. The following spots were examined: (1) upper lip (intraoral) (2) alveolar ridge (3) lateral tongue (4) floor of the mouth (5) hard palate (6) intercalary line. All sequences were examined by two CLE experts for morphological differences and video quality. Results Analysis revealed clear differences in image quality and possibility of depicting tissue morphologies between the various localizations of oral mucosa: imaging of the alveolar ridge and hard palate showed visually most discriminative tissue morphology. Labial mucosa was also visualized well using CLE. Here, typical morphological features such as uniform cells with regular intercellular gaps and vessels could be clearly depicted. Image generation and evaluation was particularly difficult in the area of the buccal mucosa, the lateral tongue and the floor of the mouth. Conclusion A physiological "CLE map" for the entire oral cavity could be created for the first time. Clinical relevance This will make it possible to take into account the existing physiological morphological features when differentiating between normal mucosa and oral squamous cell carcinoma in future work.}, language = {en} } @article{AmmelingGanzRosbachetal.2026, author = {Ammeling, Jonas and Ganz, Jonathan and Rosbach, Emely and Lausser, Ludwig and Bertram, Christof and Breininger, Katharina and Aubreville, Marc}, title = {Benchmarking Foundation Models for Mitotic Figure Classification}, volume = {3}, pages = {2026:003}, journal = {Machine Learning for Biomedical Imaging}, number = {MELBA-BVM 2025 Special Issue}, publisher = {Melba editors}, address = {[s. l.]}, issn = {2766-905X}, doi = {https://doi.org/10.59275/j.melba.2026-a3eb}, pages = {38 -- 55}, year = {2026}, abstract = {The performance of deep learning models is known to scale with data quantity and diversity. In pathology, as in many other medical imaging domains, the availability of labeled images for a specific task is often limited. Self-supervised learning techniques have enabled the use of vast amounts of unlabeled data to train large-scale neural networks, i.e., foundation models, that can address the limited data problem by providing semantically rich feature vectors that can generalize well to new tasks with minimal training effort increasing model performance and robustness. In this work, we investigate the use of foundation models for mitotic figure classification. The mitotic count, which can be derived from this classification task, is an independent prognostic marker for specific tumors and part of certain tumor grading systems. In particular, we investigate the data scaling laws on multiple current foundation models and evaluate their robustness to unseen tumor domains. Next to the commonly used linear probing paradigm, we also adapt the models using low-rank adaptation (LoRA) of their attention mechanisms. We compare all models against end-to-end-trained baselines, both CNNs and Vision Transformers. Our results demonstrate that LoRA-adapted foundation models provide superior performance to those adapted with standard linear probing, reaching performance levels close to 100 \% data availability with only 10 \% of training data. Furthermore, LoRA-adaptation of the most recent foundation models almost closes the out-of-domain performance gap when evaluated on unseen tumor domains. However, full fine-tuning of traditional architectures still yields competitive performance.}, language = {en} } @inproceedings{RosbachGanzAmmelingetal.2025, author = {Rosbach, Emely and Ganz, Jonathan and Ammeling, Jonas and Riener, Andreas and Aubreville, Marc}, title = {Automation Bias in AI-assisted Medical Decision-making under Time Pressure in Computational Pathology}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2025: Proceedings, German Conference on Medical Image Computing, Regensburg March 09-11, 2025}, editor = {Palm, Christoph and Breininger, Katharina and Deserno, Thomas Martin and Handels, Heinz and Maier, Andreas and Maier-Hein, Klaus H. and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-47422-5}, doi = {https://doi.org/10.1007/978-3-658-47422-5_27}, pages = {129 -- 134}, year = {2025}, language = {en} } @inproceedings{GanzAmmelingRosbachetal.2025, author = {Ganz, Jonathan and Ammeling, Jonas and Rosbach, Emely and Lausser, Ludwig and Bertram, Christof and Breininger, Katharina and Aubreville, Marc}, title = {Is Self-supervision Enough?}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2025: Proceedings, German Conference on Medical Image Computing, Regensburg March 09-11, 2025}, subtitle = {Benchmarking Foundation Models Against End-to-end Training for Mitotic Figure Classification}, editor = {Palm, Christoph and Breininger, Katharina and Deserno, Thomas Martin and Handels, Heinz and Maier, Andreas and Maier-Hein, Klaus H. and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-47422-5}, doi = {https://doi.org/10.1007/978-3-658-47422-5_15}, pages = {63 -- 68}, year = {2025}, language = {en} } @unpublished{AmmelingGanzRosbachetal.2025, author = {Ammeling, Jonas and Ganz, Jonathan and Rosbach, Emely and Lausser, Ludwig and Bertram, Christof and Breininger, Katharina and Aubreville, Marc}, title = {Benchmarking Foundation Models for Mitotic Figure Classification}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2508.04441}, year = {2025}, language = {en} } @article{PugetGanzBertrametal.2025, author = {Puget, Chlo{\´e} and Ganz, Jonathan and Bertram, Christof and Conrad, Thomas and Baeblich, Malte and Voss, Anne and Landmann, Katharina and Haake, Alexander F. H. and Spree, Andreas and Hartung, Svenja and Aeschlimann, Leonore and Soto, Sara and de Brot, Simone and Dettwiler, Martina and Aupperle-Lellbach, Heike and Bolfa, Pompei and Bartel, Alexander and Kiupel, Matti and Breininger, Katharina and Aubreville, Marc and Klopfleisch, Robert}, title = {Artificial intelligence predicts c-KIT exon 11 genotype by phenotype in canine cutaneous mast cell tumors: Can human observers learn it?}, volume = {63}, journal = {Veterinary Pathology}, number = {2}, publisher = {Sage}, address = {London}, issn = {1544-2217}, doi = {https://doi.org/10.1177/03009858251380284}, pages = {369 -- 379}, year = {2025}, abstract = {Canine cutaneous mast cell tumors (ccMCTs) are frequent neoplasms with variable biological behaviors. Internal tandem duplication mutations in c-KIT exon 11 (c-KIT-11-ITD) are associated with poor prognosis but predict therapeutic response to tyrosine kinase inhibitors. In a previous work, deep learning algorithms managed to predict the presence of c-KIT-11-ITD on digitalized hematoxylin and eosin-stained histological slides (whole-slide images, WSIs) in up to 87\% of cases, suggesting the existence of morphological features characterizing ccMCTs carrying c-KIT-11-ITD. This 3-stage blinded study aimed to identify morphological features indicative of c-KIT-11-ITD and to evaluate the ability of human observers to learn this task. 17 untrained pathologists first classified 8 WSIs and 200 image patches (highly relevant for algorithmic classification) of ccMCTs as either positive or negative for c-KIT-11-ITD. Second, they self-trained to recognize c-KIT-11-ITD by looking at the same WSIs and patches correctly sorted. Third, pathologists classified 15 new WSIs and 200 new patches according to c-KIT-11-ITD status. In addition, participants reported microscopic features they considered relevant for their decision. Without training, participants correctly classified the c-KIT-11-ITD status of 63\%-88\% of WSIs and 43\%-55\% of patches. With self-training, 25\%-38\% of WSIs and 55\%-56\% of patches were correctly classified. High cellular pleomorphism, anisokaryosis, and sparse cytoplasmic granulation were commonly suggested as features associated with c-KIT-11-ITD-positive ccMCTs, none of which showed reliable predictivity in a follow-up study. The results indicate that transfer of algorithmic skills to the human observer is difficult. A c-KIT-11-ITD-specific morphological feature remains to be extracted from the artificial intelligence model.}, language = {en} } @article{RosbachAmmelingGanzetal.2026, author = {Rosbach, Emely and Ammeling, Jonas and Ganz, Jonathan and Bertram, Christof and Conrad, Thomas and Riener, Andreas and Aubreville, Marc}, title = {Stuck on Suggestions: Automation Bias, the Anchoring Effect, and the Factors That Shape Them in Computational Pathology}, volume = {3}, pages = {2026:007}, journal = {Machine Learning for Biomedical Imaging}, number = {MELBA-BVM 2025 Special Issue}, publisher = {Melba editors}, address = {[s. l.]}, issn = {2766-905X}, doi = {https://doi.org/10.59275/j.melba.2026-87b1}, pages = {126 -- 147}, year = {2026}, abstract = {Artificial intelligence (AI)-driven clinical decision support systems (CDSS) hold promise to improve diagnostic accuracy and efficiency in computational pathology. However, collaboration between human experts and AI may give rise to cognitive biases, such as automation and anchoring bias, wherein users may be inclined to blindly adopt system recommendations or be disproportionately influenced by the presence of AI predictions, even when they are inaccurate. These biases may be exacerbated under time pressure, pervasive in routine pathology diagnostics, or shaped by individual user characteristics. To investigate these effects, we conducted a web-based experiment in which trained pathology experts (n = 28) estimated tumor cell percentages twice: once independently and once with the aid of an AI. A subset of the estimates in each condition was performed under time constraints. Our findings indicate that AI integration generally enhances diagnostic performance. However, it also introduced a 7\% automation bias rate, quantified as the number of accepted negative consultations, where a previously correct independent assessment gets overturned by inaccurate AI guidance. While time pressure did not increase the frequency of automation bias occurrence, it appeared to intensify its severity, as evidenced by a performance decline linked to increased automation reliance under cognitive load. A linear mixed-effects model (LMM) analysis, simulating weighted averaging, revealed a statistically significant positive coefficient for AI advice, indicating a moderate degree of anchoring on system output. This effect was further intensified under time pressure, suggesting that anchoring bias may become more pronounced when cognitive resources are limited. A secondary LMM evaluation assessing automation reliance, used as a proxy for both automation and anchoring bias, demonstrated that professional experience and self-efficacy were associated with reduced dependence on system support, whereas higher confidence during AI-assisted decision-making was linked to increased automation reliance. Together, these findings underscore the dual nature of AI integration in clinical workflows, offering performance benefits while also introducing risks of cognitive bias-driven diagnostic errors. As an initial investigation focused on a single medical specialty and diagnostic task, this study aims to lay the groundwork for future research to explore these phenomena across diverse clinical contexts, ultimately supporting the establishment of appropriate reliance on automated systems and the safe, effective integration of human-AI collaboration in medical decision-making.}, language = {en} } @article{PugetGanzOstermaieretal.2024, author = {Puget, Chlo{\´e} and Ganz, Jonathan and Ostermaier, Julian and Conrad, Thomas and Parlak, Eda and Bertram, Christof and Kiupel, Matti and Breininger, Katharina and Aubreville, Marc and Klopfleisch, Robert}, title = {Artificial intelligence can be trained to predict c-KIT-11 mutational status of canine mast cell tumors from hematoxylin and eosin-stained histological slides}, volume = {62}, journal = {Veterinary Pathology}, number = {2}, publisher = {Sage}, address = {London}, issn = {1544-2217}, doi = {https://doi.org/10.1177/03009858241286806}, pages = {152 -- 160}, year = {2024}, abstract = {Numerous prognostic factors are currently assessed histologically and immunohistochemically in canine mast cell tumors (MCTs) to evaluate clinical behavior. In addition, polymerase chain reaction (PCR) is often performed to detect internal tandem duplication (ITD) mutations in exon 11 of the c-KIT gene ( c-KIT-11-ITD) to predict the therapeutic response to tyrosine kinase inhibitors. This project aimed at training deep learning models (DLMs) to identify MCTs with c-KIT-11-ITD solely based on morphology. Hematoxylin and eosin (HE) stained slides of 368 cutaneous, subcutaneous, and mucocutaneous MCTs (195 with ITD and 173 without) were stained consecutively in 2 different laboratories and scanned with 3 different slide scanners. This resulted in 6 data sets (stain-scanner variations representing diagnostic institutions) of whole-slide images. DLMs were trained with single and mixed data sets and their performances were assessed under stain-scanner variations (domain shifts). The DLM correctly classified HE slides according to their c-KIT-11-ITD status in up to 87\% of cases with a 0.90 sensitivity and a 0.83 specificity. A relevant performance drop could be observed when the stain-scanner combination of training and test data set differed. Multi-institutional data sets improved the average accuracy but did not reach the maximum accuracy of algorithms trained and tested on the same stain-scanner variant (ie, intra-institutional). In summary, DLM-based morphological examination can predict c-KIT-11-ITD with high accuracy in canine MCTs in HE slides. However, staining protocol and scanner type influence accuracy. Larger data sets of scans from different laboratories and scanners may lead to more robust DLMs to identify c- KIT mutations in HE slides.}, language = {en} } @article{WilmFragosoGarciaMarzahletal.2022, author = {Wilm, Frauke and Fragoso-Garcia, Marco and Marzahl, Christian and Qiu, Jingna and Puget, Chlo{\´e} and Diehl, Laura and Bertram, Christof and Klopfleisch, Robert and Maier, Andreas and Breininger, Katharina and Aubreville, Marc}, title = {Pan-tumor CAnine cuTaneous Cancer Histology (CATCH) dataset}, volume = {9}, pages = {588}, journal = {Scientific Data}, publisher = {Springer}, address = {London}, issn = {2052-4463}, doi = {https://doi.org/10.1038/s41597-022-01692-w}, year = {2022}, abstract = {Due to morphological similarities, the differentiation of histologic sections of cutaneous tumors into individual subtypes can be challenging. Recently, deep learning-based approaches have proven their potential for supporting pathologists in this regard. However, many of these supervised algorithms require a large amount of annotated data for robust development. We present a publicly available dataset of 350 whole slide images of seven different canine cutaneous tumors complemented by 12,424 polygon annotations for 13 histologic classes, including seven cutaneous tumor subtypes. In inter-rater experiments, we show a high consistency of the provided labels, especially for tumor annotations. We further validate the dataset by training a deep neural network for the task of tissue segmentation and tumor subtype classification. We achieve a class-averaged Jaccard coefficient of 0.7047, and 0.9044 for tumor in particular. For classification, we achieve a slide-level accuracy of 0.9857. Since canine cutaneous tumors possess various histologic homologies to human tumors the added value of this dataset is not limited to veterinary pathology but extends to more general fields of application.}, language = {en} } @article{AubrevilleWilmStathonikosetal.2023, author = {Aubreville, Marc and Wilm, Frauke and Stathonikos, Nikolas and Breininger, Katharina and Donovan, Taryn and Jabari, Samir and Veta, Mitko and Ganz, Jonathan and Ammeling, Jonas and van Diest, Paul J and Klopfleisch, Robert and Bertram, Christof}, title = {A comprehensive multi-domain dataset for mitotic figure detection}, volume = {10}, pages = {484}, journal = {Scientific Data}, publisher = {Springer}, address = {London}, issn = {2052-4463}, doi = {https://doi.org/10.1038/s41597-023-02327-4}, year = {2023}, abstract = {The prognostic value of mitotic figures in tumor tissue is well-established for many tumor types and automating this task is of high research interest. However, especially deep learning-based methods face performance deterioration in the presence of domain shifts, which may arise from different tumor types, slide preparation and digitization devices. We introduce the MIDOG++ dataset, an extension of the MIDOG 2021 and 2022 challenge datasets. We provide region of interest images from 503 histological specimens of seven different tumor types with variable morphology with in total labels for 11,937 mitotic figures: breast carcinoma, lung carcinoma, lymphosarcoma, neuroendocrine tumor, cutaneous mast cell tumor, cutaneous melanoma, and (sub)cutaneous soft tissue sarcoma. The specimens were processed in several laboratories utilizing diverse scanners. We evaluated the extent of the domain shift by using state-of-the-art approaches, observing notable differences in single-domain training. In a leave-one-domain-out setting, generalizability improved considerably. This mitotic figure dataset is the first that incorporates a wide domain shift based on different tumor types, laboratories, whole slide image scanners, and species.}, language = {en} } @article{AubrevilleBertramDonovanetal.2020, author = {Aubreville, Marc and Bertram, Christof and Donovan, Taryn and Marzahl, Christian and Maier, Andreas and Klopfleisch, Robert}, title = {A completely annotated whole slide image dataset of canine breast cancer to aid human breast cancer research}, volume = {7}, pages = {417}, journal = {Scientific data}, publisher = {Springer}, address = {London}, issn = {2052-4463}, doi = {https://doi.org/10.1038/s41597-020-00756-z}, year = {2020}, abstract = {Canine mammary carcinoma (CMC) has been used as a model to investigate the pathogenesis of human breast cancer and the same grading scheme is commonly used to assess tumor malignancy in both. One key component of this grading scheme is the density of mitotic figures (MF). Current publicly available datasets on human breast cancer only provide annotations for small subsets of whole slide images (WSIs). We present a novel dataset of 21 WSIs of CMC completely annotated for MF. For this, a pathologist screened all WSIs for potential MF and structures with a similar appearance. A second expert blindly assigned labels, and for non-matching labels, a third expert assigned the final labels. Additionally, we used machine learning to identify previously undetected MF. Finally, we performed representation learning and two-dimensional projection to further increase the consistency of the annotations. Our dataset consists of 13,907 MF and 36,379 hard negatives. We achieved a mean F1-score of 0.791 on the test set and of up to 0.696 on a human breast cancer dataset.}, language = {en} } @unpublished{IvanHirlingGrexaetal.2025, author = {Ivan, Zsanett Zsofia and Hirling, Dominik and Grexa, Istvan and Ammeling, Jonas and Micsik, Tamas and Dobra, Katalin and Kuthi, Levente and Sukosd, Farkas and Aubreville, Marc and Miczan, Vivien and Horvath, Peter}, title = {Subphase-Labeled Mitotic Dataset for AI-powered Cell Division Analysis}, publisher = {bioRxiv}, address = {Cold Spring Harbor}, doi = {https://doi.org/10.1101/2025.07.17.665280}, year = {2025}, abstract = {Mitosis detection represents a critical task in the field of digital pathology, as determination of the mitotic index (MI) plays an important role in the tumor grading and prognostic assessment of patients. Manual determination of MI is a labor-intensive and time-consuming task for practitioners with rather high interobserver variability, thus, automation has become a priority. There has been substantial progress towards creating robust mitosis detection algorithms in recent years, primarily driven by the Mitosis Domain Generalization (MIDOG) challenges. In parallel, there has been growing interest in the molecular characterization of mitosis with the goal of achieving a more comprehensive understanding of its underlying mechanisms in a subphase-specific manner. Here, 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 of the dataset. We envision a new perspective for domain generalization by improving the performance of models with subtyping mitotic cells into the 5 main stages of normal mitosis, complemented with an atypical mitotic class. We believe that our work broadens the horizon in digital pathology: subtyping information could provide useful help for mitosis detection, while also providing promising new directions in answering biological questions, such as molecular analysis of the subphases on a single cell level.}, language = {en} } @article{MarzahlHillStaytetal.2022, author = {Marzahl, Christian and Hill, Jenny and Stayt, Jason and Bienzle, Dorothee and Welker, Lutz and Wilm, Frauke and Voigt, J{\"o}rn and Aubreville, Marc and Maier, Andreas and Klopfleisch, Robert and Breininger, Katharina and Bertram, Christof}, title = {Inter-species cell detection - datasets on pulmonary hemosiderophages in equine, human and feline specimens}, volume = {9}, pages = {269}, journal = {Scientific Data}, publisher = {Springer}, address = {London}, issn = {2052-4463}, doi = {https://doi.org/10.1038/s41597-022-01389-0}, year = {2022}, abstract = {Pulmonary hemorrhage (P-Hem) occurs among multiple species and can have various causes. Cytology of bronchoalveolar lavage fluid (BALF) using a 5-tier scoring system of alveolar macrophages based on their hemosiderin content is considered the most sensitive diagnostic method. We introduce a novel, fully annotated multi-species P-Hem dataset, which consists of 74 cytology whole slide images (WSIs) with equine, feline and human samples. To create this high-quality and high-quantity dataset, we developed an annotation pipeline combining human expertise with deep learning and data visualisation techniques. We applied a deep learning-based object detection approach trained on 17 expertly annotated equine WSIs, to the remaining 39 equine, 12 human and 7 feline WSIs. The resulting annotations were semi-automatically screened for errors on multiple types of specialised annotation maps and finally reviewed by a trained pathologist. Our dataset contains a total of 297,383 hemosiderophages classified into five grades. It is one of the largest publicly available WSIs datasets with respect to the number of annotations, the scanned area and the number of species covered.}, language = {en} } @article{BertramAubrevilleMarzahletal.2019, author = {Bertram, Christof and Aubreville, Marc and Marzahl, Christian and Maier, Andreas and Klopfleisch, Robert}, title = {A large-scale dataset for mitotic figure assessment on whole slide images of canine cutaneous mast cell tumor}, volume = {6}, pages = {274}, journal = {Scientific data}, publisher = {Springer}, address = {London}, issn = {2052-4463}, doi = {https://doi.org/10.1038/s41597-019-0290-4}, year = {2019}, abstract = {We introduce a novel, large-scale dataset for microscopy cell annotations. The dataset includes 32 whole slide images (WSI) of canine cutaneous mast cell tumors, selected to include both low grade cases as well as high grade cases. The slides have been completely annotated for mitotic figures and we provide secondary annotations for neoplastic mast cells, inflammatory granulocytes, and mitotic figure look-alikes. Additionally to a blinded two-expert manual annotation with consensus, we provide an algorithm-aided dataset, where potentially missed mitotic figures were detected by a deep neural network and subsequently assessed by two human experts. We included 262,481 annotations in total, out of which 44,880 represent mitotic figures. For algorithmic validation, we used a customized RetinaNet approach, followed by a cell classification network. We find F1-Scores of 0.786 and 0.820 for the manually labelled and the algorithm-aided dataset, respectively. The dataset provides, for the first time, WSIs completely annotated for mitotic figures and thus enables assessment of mitosis detection algorithms on complete WSIs as well as region of interest detection algorithms.}, language = {en} } @article{IvanHirlingGrexaetal.2026, author = {Ivan, Zsanett Zsofia and Hirling, Dominik and Grexa, Istvan and Ammeling, Jonas and Molnar, Csaba and Micsik, Tamas and Dobra, Katalin and Kuthi, Levente and Sukosd, Farkas and Fillinger, Janos and Moldvay, Judit and Toth, Erika and Aubreville, Marc and Miczan, Vivien and Horvath, Peter}, title = {A Subphase-Labeled Mitotic Dataset for AI-powered Cell Division Analysis}, journal = {Scientific Data}, publisher = {Springer}, address = {London}, issn = {2052-4463}, doi = {https://doi.org/10.1038/s41597-026-07007-7}, pages = {17}, year = {2026}, 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.}, language = {en} } @inproceedings{BertramWeissDonovanetal.2025, author = {Bertram, Christof and Weiss, Viktoria and Donovan, Taryn and Banerjee, Sweta and Conrad, Thomas and Ammeling, Jonas and Klopfleisch, Robert and Kaltenecker, Christopher and Aubreville, Marc}, title = {Histologic Dataset of Normal and Atypical Mitotic Figures on Human Breast Cancer (AMi-Br)}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2025: Proceedings, German Conference on Medical Image Computing, Regensburg March 09-11, 2025}, editor = {Palm, Christoph and Breininger, Katharina and Deserno, Thomas Martin and Handels, Heinz and Maier, Andreas and Maier-Hein, Klaus H. and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-47422-5}, doi = {https://doi.org/10.1007/978-3-658-47422-5_25}, pages = {113 -- 118}, year = {2025}, language = {en} }