@unpublished{RosbachAmmelingKruegeletal.2024, author = {Rosbach, Emely and Ammeling, Jonas and Kr{\"u}gel, Sebastian and Kießig, Angelika and Fritz, Alexis and Ganz, Jonathan and Puget, Chlo{\´e} and Donovan, Taryn and Klang, Andrea and K{\"o}ller, Maximilian C. and Bolfa, Pompei and Tecilla, Marco and Denk, Daniela and Kiupel, Matti and Paraschou, Georgios and Kok, Mun Keong and Haake, Alexander F. H. and de Krijger, Ronald R. and Sonnen, Andreas F.-P. and Kasantikul, Tanit and Dorrestein, Gerry M. and Smedley, Rebecca C. and Stathonikos, Nikolas and Uhl, Matthias and Bertram, Christof and Riener, Andreas and Aubreville, Marc}, title = {"When TwoWrongs Don't Make a Right" - Examining Confirmation Bias and the Role of Time Pressure During Human-AI Collaboration in Computational Pathology}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2411.01007}, year = {2024}, language = {en} } @unpublished{RosbachGanzAmmelingetal.2024, author = {Rosbach, Emely and Ganz, Jonathan and Ammeling, Jonas and Riener, Andreas and Aubreville, Marc}, title = {Automation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in Computational Pathology}, doi = {https://doi.org/10.48550/arXiv.2411.00998}, year = {2024}, language = {en} } @unpublished{AubrevilleStathonikosDonovanetal.2023, author = {Aubreville, Marc and Stathonikos, Nikolas and Donovan, Taryn and Klopfleisch, Robert and Ganz, Jonathan and Ammeling, Jonas and Wilm, Frauke and Veta, Mitko and Jabari, Samir and Eckstein, Markus and Annuscheit, Jonas and Krumnow, Christian and Bozaba, Engin and Cayir, Sercan and Gu, Hongyan and Chen, Xiang and Jahanifar, Mostafa and Shephard, Adam and Kondo, Satoshi and Kasai, Satoshi and Kotte, Sujatha and Saipradeep, Vangala and Lafarge, Maxime W. and Koelzer, Viktor H. and Wang, Ziyue and Zhang, Yongbing and Yang, Sen and Wang, Xiyue and Breininger, Katharina and Bertram, Christof}, title = {Domain generalization across tumor types, laboratories, and species - Insights from the 2022 edition of the Mitosis Domain Generalization Challenge}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2309.15589}, year = {2023}, abstract = {Recognition of mitotic figures in histologic tumor specimens is highly relevant to patient outcome assessment. This task is challenging for algorithms and human experts alike, with deterioration of algorithmic performance under shifts in image representations. Considerable covariate shifts occur when assessment is performed on different tumor types, images are acquired using different digitization devices, or specimens are produced in different laboratories. This observation motivated the inception of the 2022 challenge on MItosis Domain Generalization (MIDOG 2022). The challenge provided annotated histologic tumor images from six different domains and evaluated the algorithmic approaches for mitotic figure detection provided by nine challenge participants on ten independent domains. Ground truth for mitotic figure detection was established in two ways: a three-expert consensus and an independent, immunohistochemistry-assisted set of labels. This work represents an overview of the challenge tasks, the algorithmic strategies employed by the participants, and potential factors contributing to their success. With an F1 score of 0.764 for the top-performing team, we summarize that domain generalization across various tumor domains is possible with today's deep learning-based recognition pipelines. When assessed against the immunohistochemistry-assisted reference standard, all methods resulted in reduced recall scores, but with only minor changes in the order of participants in the ranking.}, language = {en} } @unpublished{HaghoferParlakBarteletal.2023, author = {Haghofer, Andreas and Parlak, Eda and Bartel, Alexander and Donovan, Taryn and Assenmacher, Charles-Antoine and Bolfa, Pompei and Dark, Michael and Fuchs-Baumgartinger, Andrea and Klang, Andrea and J{\"a}ger, Kathrin 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 Kiupel, Matti and Bertram, Christof}, title = {Nuclear Morphometry using a Deep Learning-based Algorithm has Prognostic Relevance for Canine Cutaneous Mast Cell Tumors}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2309.15031}, year = {2023}, abstract = {Variation in nuclear size and shape is an important criterion of malignancy for many tumor types; however, categorical estimates by pathologists have poor reproducibility. Measurements of nuclear characteristics (morphometry) can improve reproducibility, but manual methods are time consuming. In this study, we evaluated fully automated morphometry using a deep learning-based algorithm in 96 canine cutaneous mast cell tumors with information on patient survival. Algorithmic morphometry was compared with karyomegaly estimates by 11 pathologists, manual nuclear morphometry of 12 cells by 9 pathologists, and the mitotic count as a benchmark. The prognostic value of automated morphometry was high with an area under the ROC curve regarding the tumor-specific survival of 0.943 (95\% CI: 0.889 - 0.996) for the standard deviation (SD) of nuclear area, which was higher than manual morphometry of all pathologists combined (0.868, 95\% CI: 0.737 - 0.991) and the mitotic count (0.885, 95\% CI: 0.765 - 1.00). At the proposed thresholds, the hazard ratio for algorithmic morphometry (SD of nuclear area ≥9.0μm2) was 18.3 (95\% CI: 5.0 - 67.1), for manual morphometry (SD of nuclear area ≥10.9μm2) 9.0 (95\% CI: 6.0 - 13.4), for karyomegaly estimates 7.6 (95\% CI: 5.7 - 10.1), and for the mitotic count 30.5 (95\% CI: 7.8 - 118.0). Inter-rater reproducibility for karyomegaly estimates was fair (κ = 0.226) with highly variable sensitivity/specificity values for the individual pathologists. Reproducibility for manual morphometry (SD of nuclear area) was good (ICC = 0.654). This study supports the use of algorithmic morphometry as a prognostic test to overcome the limitations of estimates and manual measurements.}, language = {en} } @unpublished{GanzMarzahlAmmelingetal.2024, author = {Ganz, Jonathan and Marzahl, Christian and Ammeling, Jonas and Richter, Barbara and Puget, Chlo{\´e} and Denk, Daniela and Demeter, Elena A. and Tabaran, Flaviu A. and Wasinger, Gabriel and Lipnik, Karoline and Tecilla, Marco and Valentine, Matthew J. and Dark, Michael and Abele, Niklas and Bolfa, Pompei and Erber, Ramona and Klopfleisch, Robert and Merz, Sophie and Donovan, Taryn and Jabari, Samir and Bertram, Christof and Breininger, Katharina and Aubreville, Marc}, title = {On the Value of PHH3 for Mitotic Figure Detection on H\&E-stained Images}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2406.19899}, year = {2024}, abstract = {The count of mitotic figures (MFs) observed in hematoxylin and eosin (H\&E)-stained slides is an important prognostic marker as it is a measure for tumor cell proliferation. However, the identification of MFs has a known low inter-rater agreement. Deep learning algorithms can standardize this task, but they require large amounts of annotated data for training and validation. Furthermore, label noise introduced during the annotation process may impede the algorithm's performance. Unlike H\&E, the mitosis-specific antibody phospho-histone H3 (PHH3) specifically highlights MFs. Counting MFs on slides stained against PHH3 leads to higher agreement among raters and has therefore recently been used as a ground truth for the annotation of MFs in H\&E. However, as PHH3 facilitates the recognition of cells indistinguishable from H\&E stain alone, the use of this ground truth could potentially introduce noise into the H\&E-related dataset, impacting model performance. This study analyzes the impact of PHH3-assisted MF annotation on inter-rater reliability and object level agreement through an extensive multi-rater experiment. We found that the annotators' object-level agreement increased when using PHH3-assisted labeling. Subsequently, MF detectors were evaluated on the resulting datasets to investigate the influence of PHH3-assisted labeling on the models' performance. Additionally, a novel dual-stain MF detector was developed to investigate the interpretation-shift of PHH3-assisted labels used in H\&E, which clearly outperformed single-stain detectors. However, the PHH3-assisted labels did not have a positive effect on solely H\&E-based models. The high performance of our dual-input detector reveals an information mismatch between the H\&E and PHH3-stained images as the cause of this effect.}, language = {en} } @unpublished{GanzAmmelingRosbachetal.2024, 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? Benchmarking Foundation Models Against End-to-End Training for Mitotic Figure Classification}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2412.06365}, year = {2024}, abstract = {Foundation models (FMs), i.e., models trained on a vast amount of typically unlabeled data, have become popular and available recently for the domain of histopathology. The key idea is to extract semantically rich vectors from any input patch, allowing for the use of simple subsequent classification networks potentially reducing the required amounts of labeled data, and increasing domain robustness. In this work, we investigate to which degree this also holds for mitotic figure classification. Utilizing two popular public mitotic figure datasets, we compared linear probing of five publicly available FMs against models trained on ImageNet and a simple ResNet50 end-to-end-trained baseline. We found that the end-to-end-trained baseline outperformed all FM-based classifiers, regardless of the amount of data provided. Additionally, we did not observe the FM-based classifiers to be more robust against domain shifts, rendering both of the above assumptions incorrect.}, language = {en} } @unpublished{PugetGanzOstermaieretal.2024, author = {Puget, Chlo{\´e} and Ganz, Jonathan and Ostermaier, Julian and Konrad, Thomas and Parlak, Eda and Bertram, Christof and Kiupel, Matti and Breininger, Katharina and Aubreville, Marc and Klopfleisch, Robert}, title = {Deep Learning model predicts the c-Kit-11 mutational status of canine cutaneous mast cell tumors by HE stained histological slides}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2401.06169}, year = {2024}, abstract = {Numerous prognostic factors are currently assessed histopathologically in biopsies of canine mast cell tumors to evaluate clinical behavior. In addition, PCR analysis of the c-Kit exon 11 mutational status is often performed to evaluate the potential success of a tyrosine kinase inhibitor therapy. This project aimed at training deep learning models (DLMs) to identify the c-Kit-11 mutational status of MCTs solely based on morphology without additional molecular analysis. HE slides of 195 mutated and 173 non-mutated tumors were stained consecutively in two different laboratories and scanned with three different slide scanners. This resulted in six different datasets (stain-scanner variations) of whole slide images. DLMs were trained with single and mixed datasets and their performances was assessed under scanner and staining domain shifts. The DLMs correctly classified HE slides according to their c-Kit 11 mutation status in, on average, 87\% of cases for the best-suited stain-scanner variant. A relevant performance drop could be observed when the stain-scanner combination of the training and test dataset differed. Multi-variant datasets improved the average accuracy but did not reach the maximum accuracy of algorithms trained and tested on the same stain-scanner variant. In summary, DLM-assisted morphological examination of MCTs can predict c-Kit-exon 11 mutational status of MCTs with high accuracy. However, the recognition performance is impeded by a change of scanner or staining protocol. Larger data sets with higher numbers of scans originating from different laboratories and scanners may lead to more robust DLMs to identify c-Kit mutations in HE slides.}, language = {en} } @unpublished{GanzMarzahlAmmelingetal.2024, author = {Ganz, Jonathan and Marzahl, Christian and Ammeling, Jonas and Rosbach, Emely and Richter, Barbara and Puget, Chlo{\´e} and Denk, Daniela and Demeter, Elena A. and Tabaran, Flaviu A. and Wasinger, Gabriel and Lipnik, Karoline and Tecilla, Marco and Valentine, Matthew J. and Dark, Michael and Abele, Niklas and Bolfa, Pompei and Erber, Ramona and Klopfleisch, Robert and Merz, Sophie and Donovan, Taryn and Jabari, Samir and Bertram, Christof and Breininger, Katharina and Aubreville, Marc}, title = {Information Mismatch in PHH3-Assisted Mitosis Annotation Leads to Interpretation Shifts in H\&E Slide Analysis}, titleParent = {Research Square}, publisher = {Research Square}, address = {Durham}, doi = {https://doi.org/10.21203/rs.3.rs-4900505/v1}, year = {2024}, abstract = {The count of mitotic figures (MFs) observed in hematoxylin and eosin (H\&E)-stained slides is an important prognostic marker, as it is a measure for tumor cell proliferation. However, the identification of MFs has a known low inter-rater agreement. In a computer-aided setting, deep learning algorithms can help to mitigate this, but they require large amounts of annotated data for training and validation. Furthermore, label noise introduced during the annotation process may impede the algorithms' performance. Unlike H\&E, where identification of MFs is based mainly on morphological features, the mitosis-specific antibody phospho-histone H3 (PHH3) specifically highlights MFs. Counting MFs on slides stained against PHH3 leads to higher agreement among raters and has therefore recently been used as a ground truth for the annotation of MFs in H\&E. However, as PHH3 facilitates the recognition of cells indistinguishable from H\&E staining alone, the use of this ground truth could potentially introduce an interpretation shift and even label noise into the H\&E-related dataset, impacting model performance. This study analyzes the impact of PHH3-assisted MF annotation on inter-rater reliability and object level agreement through an extensive multi-rater experiment. Subsequently, MF detectors, including a novel dual-stain detector, were evaluated on the resulting datasets to investigate the influence of PHH3-assisted labeling on the models' performance. We found that the annotators' object-level agreement significantly increased when using PHH3-assisted labeling (F1: 0.53 to 0.74). However, this enhancement in label consistency did not translate to improved performance for H\&E-based detectors, neither during the training phase nor the evaluation phase. Conversely, the dual-stain detector was able to benefit from the higher consistency. This reveals an information mismatch between the H\&E and PHH3-stained images as the cause of this effect, which renders PHH3-assisted annotations not well-aligned for use with H\&E-based detectors. Based on our findings, we propose an improved PHH3-assisted labeling procedure.}, language = {en} } @unpublished{AmmelingHeckerGanzetal.2023, 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 Volume Corrected Mitotic Index Calculation Through Annotation-Free Deep Learning using Immunohistochemistry as Reference Standard}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2311.08949}, year = {2023}, abstract = {The volume-corrected mitotic index (M/V-Index) was shown to provide prognostic value in invasive breast carcinomas. However, despite its prognostic significance, it is not established as the standard method for assessing aggressive biological behaviour, due to the high additional workload associated with determining the epithelial proportion. In this work, we show that using a deep learning pipeline solely trained with an annotation-free, immunohistochemistry-based approach, provides accurate estimations of epithelial segmentation in canine breast carcinomas. We compare our automatic framework with the manually annotated M/V-Index in a study with three board-certified pathologists. Our results indicate that the deep learning-based pipeline shows expert-level performance, while providing time efficiency and reproducibility.}, language = {en} } @unpublished{WilmFragosoGarciaBertrametal.2022, author = {Wilm, Frauke and Fragoso-Garcia, Marco and Bertram, Christof and Stathonikos, Nikolas and {\"O}ttl, Mathias and Qiu, Jingna and Klopfleisch, Robert and Maier, Andreas and Aubreville, Marc and Breininger, Katharina}, title = {Mind the Gap: Scanner-induced domain shifts pose challenges for representation learning in histopathology}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2211.16141}, year = {2022}, language = {en} } @unpublished{AmmelingSchmidtGanzetal.2022, author = {Ammeling, Jonas and Schmidt, Lars-Henning and Ganz, Jonathan and Niedermair, Tanja and Brochhausen-Delius, Christoph and Schulz, Christian and Breininger, Katharina and Aubreville, Marc}, title = {Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2212.07724}, year = {2022}, abstract = {Attention-based multiple instance learning (AMIL) algorithms have proven to be successful in utilizing gigapixel whole-slide images (WSIs) for a variety of different computational pathology tasks such as outcome prediction and cancer subtyping problems. We extended an AMIL approach to the task of survival prediction by utilizing the classical Cox partial likelihood as a loss function, converting the AMIL model into a nonlinear proportional hazards model. We applied the model to tissue microarray (TMA) slides of 330 lung cancer patients. The results show that AMIL approaches can handle very small amounts of tissue from a TMA and reach similar C-index performance compared to established survival prediction methods trained with highly discriminative clinical factors such as age, cancer grade, and cancer stage.}, language = {en} } @unpublished{StoeveAubrevilleOetteretal.2018, author = {Stoeve, Maike and Aubreville, Marc and Oetter, Nicolai and Knipfer, Christian and Neumann, Helmut and Stelzle, Florian and Maier, Andreas}, title = {Motion Artifact Detection in Confocal Laser Endomicroscopy Images}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.1711.01117}, year = {2018}, abstract = {Confocal Laser Endomicroscopy (CLE), an optical imaging technique allowing non-invasive examination of the mucosa on a (sub)- cellular level, has proven to be a valuable diagnostic tool in gastroenterology and shows promising results in various anatomical regions including the oral cavity. Recently, the feasibility of automatic carcinoma detection for CLE images of sufficient quality was shown. However, in real world data sets a high amount of CLE images is corrupted by artifacts. Amongst the most prevalent artifact types are motion-induced image deteriorations. In the scope of this work, algorithmic approaches for the automatic detection of motion artifact-tainted image regions were developed. Hence, this work provides an important step towards clinical applicability of automatic carcinoma detection. Both, conventional machine learning and novel, deep learning-based approaches were assessed. The deep learning-based approach outperforms the conventional approaches, attaining an AUC of 0.90.}, language = {en} } @unpublished{AubrevillePanSievertetal.2023, author = {Aubreville, Marc and Pan, Zhaoya and Sievert, Matti and Ammeling, Jonas and Ganz, Jonathan and Oetter, Nicolai and Stelzle, Florian and Frenken, Ann-Kathrin and Breininger, Katharina and Goncalves, Miguel}, title = {Few Shot Learning for the Classification of Confocal Laser Endomicroscopy Images of Head and Neck Tumors}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2311.07216}, year = {2023}, abstract = {The surgical removal of head and neck tumors requires safe margins, which are usually confirmed intraoperatively by means of frozen sections. This method is, in itself, an oversampling procedure, which has a relatively low sensitivity compared to the definitive tissue analysis on paraffin-embedded sections. Confocal laser endomicroscopy (CLE) is an in-vivo imaging technique that has shown its potential in the live optical biopsy of tissue. An automated analysis of this notoriously difficult to interpret modality would help surgeons. However, the images of CLE show a wide variability of patterns, caused both by individual factors but also, and most strongly, by the anatomical structures of the imaged tissue, making it a challenging pattern recognition task. In this work, we evaluate four popular few shot learning (FSL) methods towards their capability of generalizing to unseen anatomical domains in CLE images. We evaluate this on images of sinunasal tumors (SNT) from five patients and on images of the vocal folds (VF) from 11 patients using a cross-validation scheme. The best respective approach reached a median accuracy of 79.6\% on the rather homogeneous VF dataset, but only of 61.6\% for the highly diverse SNT dataset. Our results indicate that FSL on CLE images is viable, but strongly affected by the number of patients, as well as the diversity of anatomical patterns.}, language = {en} } @unpublished{AubrevilleKrappmannBertrametal.2017, author = {Aubreville, Marc and Krappmann, Maximilian and Bertram, Christof and Klopfleisch, Robert and Maier, Andreas}, title = {A Guided Spatial Transformer Network for Histology Cell Differentiation}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.1707.08525}, year = {2017}, language = {en} } @unpublished{PerniasRampasAubreville2023, author = {Pernias, Pablo and Rampas, Dominic and Aubreville, Marc}, title = {W{\"u}rstchen: Efficient Pretraining of Text-to-Image Models}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2306.00637}, year = {2023}, abstract = {We introduce Wuerstchen, a novel technique for text-to-image synthesis that unites competitive performance with unprecedented cost-effectiveness and ease of training on constrained hardware. Building on recent advancements in machine learning, our approach, which utilizes latent diffusion strategies at strong latent image compression rates, significantly reduces the computational burden, typically associated with state-of-the-art models, while preserving, if not enhancing, the quality of generated images. Wuerstchen achieves notable speed improvements at inference time, thereby rendering real-time applications more viable. One of the key advantages of our method lies in its modest training requirements of only 9,200 GPU hours, slashing the usual costs significantly without compromising the end performance. In a comparison against the state-of-the-art, we found the approach to yield strong competitiveness. This paper opens the door to a new line of research that prioritizes both performance and computational accessibility, hence democratizing the use of sophisticated AI technologies. Through Wuerstchen, we demonstrate a compelling stride forward in the realm of text-to-image synthesis, offering an innovative path to explore in future research.}, language = {en} } @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} } @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} } @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} } @unpublished{AubrevilleGanzAmmelingetal.2024, author = {Aubreville, Marc and Ganz, Jonathan and Ammeling, Jonas and Kaltenecker, Christopher and Bertram, Christof}, title = {Model-based Cleaning of the QUILT-1M Pathology Dataset for Text-Conditional Image Synthesis}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2404.07676}, year = {2024}, abstract = {The QUILT-1M dataset is the first openly available dataset containing images harvested from various online sources. While it provides a huge data variety, the image quality and composition is highly heterogeneous, impacting its utility for text-conditional image synthesis. We propose an automatic pipeline that provides predictions of the most common impurities within the images, e.g., visibility of narrators, desktop environment and pathology software, or text within the image. Additionally, we propose to use semantic alignment filtering of the image-text pairs. Our findings demonstrate that by rigorously filtering the dataset, there is a substantial enhancement of image fidelity in text-to-image tasks.}, language = {en} } @unpublished{MarzahlBertramAubrevilleetal.2020, author = {Marzahl, Christian and Bertram, Christof and Aubreville, Marc and Petrick, Anne and Weiler, Kristina and Gl{\"a}sel, Agnes C. and Fragoso-Garcia, Marco and Merz, Sophie and Bartenschlager, Florian and Hoppe, Judith and Langenhagen, Alina and Jasensky, Anne-Katherine and Voigt, J{\"o}rn and Klopfleisch, Robert and Maier, Andreas}, title = {Are Fast Labeling Methods Reliable? A Case Study of Computer-Aided Expert Annotations on Microscopy Slides}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2004.05838}, year = {2020}, language = {en} } @unpublished{BertramVetaMarzahletal.2020, author = {Bertram, Christof and Veta, Mitko and Marzahl, Christian and Stathonikos, Nikolas and Maier, Andreas and Klopfleisch, Robert and Aubreville, Marc}, title = {Are pathologist-defined labels reproducible? Comparison of the TUPAC16 mitotic figure dataset with an alternative set of labels}, publisher = {arXiv}, address = {Ithaca}, doi = {https://doi.org/10.48550/arXiv.2007.05351}, year = {2020}, language = {en} }