Prof. Dr.
Marc Aubreville
I’ve received my Ph.D (Dr.-Ing.) from Friedrich-Alexander-Universität Erlangen-Nürnberg and my M. Sc. (Dipl.-Ing.) from Karlsruhe Institute of Technology. My vision is to utilize the power of artificial intelligence algorithms and scale them to clinical applications. With 10+ years of industry experience, I know what it takes to create a great medical product.
Publications:
2024
75.
Puget, Chloé; Ganz, Jonathan; Ostermaier, Julian; Conrad, Thomas; Parlak, Eda; Bertram, Christof A.; Kiupel, Matti; Breininger, Katharina; Aubreville, Marc; Klopfleisch, Robert
In: Veterinary Pathology, pp. 03009858241286806, 2024, ISSN: 0300-9858, 1544-2217.
Abstract | Links | BibTeX | Tags:
@article{puget_artificial_2024,
title = {Artificial intelligence can be trained to predict textitc-KIT -11 mutational status of canine mast cell tumors from hematoxylin and eosin-stained histological slides},
author = {Chloé Puget and Jonathan Ganz and Julian Ostermaier and Thomas Conrad and Eda Parlak and Christof A. Bertram and Matti Kiupel and Katharina Breininger and Marc Aubreville and Robert Klopfleisch},
url = {https://journals.sagepub.com/doi/10.1177/03009858241286806},
doi = {10.1177/03009858241286806},
issn = {0300-9858, 1544-2217},
year = {2024},
date = {2024-10-01},
urldate = {2024-10-21},
journal = {Veterinary Pathology},
pages = {03009858241286806},
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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
74.
Aubreville, Marc; Ganz, Jonathan; Ammeling, Jonas; Rosbach, Emely; Gehrke, Thomas; Scherzad, Agmal; Hackenberg, Stephan; Goncalves, Miguel
Prediction of tumor board procedural recommendations using large language models Journal Article
In: European Archives of Oto-Rhino-Laryngology, 2024, ISSN: 1434-4726.
Abstract | Links | BibTeX | Tags:
@article{aubreville_prediction_2024,
title = {Prediction of tumor board procedural recommendations using large language models},
author = {Marc Aubreville and Jonathan Ganz and Jonas Ammeling and Emely Rosbach and Thomas Gehrke and Agmal Scherzad and Stephan Hackenberg and Miguel Goncalves},
url = {https://doi.org/10.1007/s00405-024-08947-9},
doi = {10.1007/s00405-024-08947-9},
issn = {1434-4726},
year = {2024},
date = {2024-09-01},
journal = {European Archives of Oto-Rhino-Laryngology},
abstract = {Multidisciplinary tumor boards are meetings where a team of medical specialists, including medical oncologists, radiation oncologists, radiologists, surgeons, and pathologists, collaborate to determine the best treatment plan for cancer patients. While decision-making in this context is logistically and cost-intensive, it has a significant positive effect on overall cancer survival.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Multidisciplinary tumor boards are meetings where a team of medical specialists, including medical oncologists, radiation oncologists, radiologists, surgeons, and pathologists, collaborate to determine the best treatment plan for cancer patients. While decision-making in this context is logistically and cost-intensive, it has a significant positive effect on overall cancer survival.
73.
Ganz, Jonathan; Ammeling, Jonas; Jabari, Samir; Breininger, Katharina; Aubreville, Marc
Re-identification from histopathology images Journal Article
In: Medical Image Analysis, pp. 103335, 2024, ISSN: 13618415.
@article{ganz_re-identification_2024,
title = {Re-identification from histopathology images},
author = {Jonathan Ganz and Jonas Ammeling and Samir Jabari and Katharina Breininger and Marc Aubreville},
url = {https://linkinghub.elsevier.com/retrieve/pii/S1361841524002603},
doi = {10.1016/j.media.2024.103335},
issn = {13618415},
year = {2024},
date = {2024-09-01},
urldate = {2024-09-20},
journal = {Medical Image Analysis},
pages = {103335},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
72.
Aubreville, Marc; Ganz, Jonathan; Ammeling, Jonas; Kaltenecker, Christopher; Bertram, Christof A.
Model-based Cleaning of the QUILT-1M Pathology Dataset for Text-Conditional Image Synthesis Proceedings Article
In: Medical Imaging with Deep Learning, Paris, France, 2024.
Abstract | Links | BibTeX | Tags:
@inproceedings{aubreville_model-based_2024,
title = {Model-based Cleaning of the QUILT-1M Pathology Dataset for Text-Conditional Image Synthesis},
author = {Marc Aubreville and Jonathan Ganz and Jonas Ammeling and Christopher Kaltenecker and Christof A. Bertram},
url = {https://openreview.net/forum?id=m7wYKrUjzV},
year = {2024},
date = {2024-07-01},
booktitle = {Medical Imaging with Deep Learning},
address = {Paris, France},
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.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
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.
71.
Fick, Rutger H. J.; Bertram, Christof A.; Aubreville, Marc
Improving CNN-Based Mitosis Detection through Rescanning Annotated Glass Slides and Atypical Mitosis Subtyping Proceedings Article
In: Medical Imaging with Deep Learning 2024, 2024.
@inproceedings{fick_improving_2024,
title = {Improving CNN-Based Mitosis Detection through Rescanning Annotated Glass Slides and Atypical Mitosis Subtyping},
author = {Rutger H. J. Fick and Christof A. Bertram and Marc Aubreville},
url = {https://openreview.net/forum?id=00gWBAAbMI},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
booktitle = {Medical Imaging with Deep Learning 2024},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
70.
Glahn, Imaine; Haghofer, Andreas; Donovan, Taryn A.; Degasperi, Brigitte; Bartel, Alexander; Kreilmeier-Berger, Theresa; Hyndman, Philip S.; Janout, Hannah; Assenmacher, Charles-Antoine; Bartenschlager, Florian; Bolfa, Pompei; Dark, Michael J.; Klang, Andrea; Klopfleisch, Robert; Merz, Sophie; Richter, Barbara; Schulman, F. Yvonne; Ganz, Jonathan; Scharinger, Josef; Aubreville, Marc; Winkler, Stephan M.; Bertram, Christof A.
Automated Nuclear Morphometry: A Deep Learning Approach for Prognostication in Canine Pulmonary Carcinoma to Enhance Reproducibility Journal Article
In: Veterinary Sciences, vol. 11, no. 6, pp. 278, 2024, ISSN: 2306-7381.
Abstract | Links | BibTeX | Tags:
@article{glahn_automated_2024,
title = {Automated Nuclear Morphometry: A Deep Learning Approach for Prognostication in Canine Pulmonary Carcinoma to Enhance Reproducibility},
author = {Imaine Glahn and Andreas Haghofer and Taryn A. Donovan and Brigitte Degasperi and Alexander Bartel and Theresa Kreilmeier-Berger and Philip S. Hyndman and Hannah Janout and Charles-Antoine Assenmacher and Florian Bartenschlager and Pompei Bolfa and Michael J. Dark and Andrea Klang and Robert Klopfleisch and Sophie Merz and Barbara Richter and F. Yvonne Schulman and Jonathan Ganz and Josef Scharinger and Marc Aubreville and Stephan M. Winkler and Christof A. Bertram},
url = {https://www.mdpi.com/2306-7381/11/6/278},
doi = {10.3390/vetsci11060278},
issn = {2306-7381},
year = {2024},
date = {2024-06-01},
urldate = {2024-06-19},
journal = {Veterinary Sciences},
volume = {11},
number = {6},
pages = {278},
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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
69.
Krügel, Sebastian; Ammeling, Jonas; Aubreville, Marc; Fritz, Alexis; Kießig, Angelika; Uhl, Matthias
Perceived responsibility in AI-supported medicine Journal Article
In: AI & SOCIETY, 2024, ISSN: 0951-5666, 1435-5655.
Abstract | Links | BibTeX | Tags:
@article{krugel_perceived_2024,
title = {Perceived responsibility in AI-supported medicine},
author = {Sebastian Krügel and Jonas Ammeling and Marc Aubreville and Alexis Fritz and Angelika Kießig and Matthias Uhl},
url = {https://link.springer.com/10.1007/s00146-024-01972-6},
doi = {10.1007/s00146-024-01972-6},
issn = {0951-5666, 1435-5655},
year = {2024},
date = {2024-05-01},
urldate = {2024-05-23},
journal = {AI & SOCIETY},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
68.
Frenken, Ann‐Kathrin; Sievert, Matti; Panuganti, Bharat; Aubreville, Marc; Meyer, Till; Scherzad, Agmal; Gehrke, Thomas; Scheich, Matthias; Hackenberg, Stephan; Goncalves, Miguel
Feasibility of Optical Biopsy During Endoscopic Sinus Surgery With Confocal Laser Endomicroscopy: A Pilot Study Journal Article
In: The Laryngoscope, pp. lary.31503, 2024, ISSN: 0023-852X, 1531-4995.
Abstract | Links | BibTeX | Tags:
@article{frenken_feasibility_2024,
title = {Feasibility of Optical Biopsy During Endoscopic Sinus Surgery With Confocal Laser Endomicroscopy: A Pilot Study},
author = {Ann‐Kathrin Frenken and Matti Sievert and Bharat Panuganti and Marc Aubreville and Till Meyer and Agmal Scherzad and Thomas Gehrke and Matthias Scheich and Stephan Hackenberg and Miguel Goncalves},
url = {https://onlinelibrary.wiley.com/doi/10.1002/lary.31503},
doi = {10.1002/lary.31503},
issn = {0023-852X, 1531-4995},
year = {2024},
date = {2024-05-01},
urldate = {2024-05-19},
journal = {The Laryngoscope},
pages = {lary.31503},
abstract = {Objective
Confocal laser endomicroscopy (CLE) is an optical imaging technique that allows in vivo, real‐time, microscope‐like assessment of superficial lesions. Although there is substantial data on CLE use in the upper GI tract, there is limited information regarding its application in the nasal cavity and paranasal sinuses. This study aims to assess the feasibility and diagnostic metrics of CLE in the nasal cavity and paranasal sinuses regarding differentiation between healthy/benign and malignant tissue. These structures show, however, a wider variety of frequent and concomitant benign and malignant pathologies, which could pose an increased challenge for optical biopsy by CLE.
Methods
We performed CLE on a case series of six patients with various findings in the nose (three chronic rhinosinusitis, adenocarcinoma, meningoenzephalozele, esthesionneuroblastoma). Forty‐two sequences (3792 images) from various structures in the nasal cavity and/or paranasal sinuses were acquired. Biopsies were taken at corresponding locations and analyzed in hematoxylin and eosin staining as a standard of reference. Three independent examiners blinded to the histopathology assessed the sequences.
Results
Healthy and inflamed mucosa could be distinguished from malignant lesions with an accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of 84.1%, 85.4%, 83.1%, 72.5%, and 92.1%, respectively, with a substantial agreement between raters (Fleiss
κ = 0.62).
Conclusion
This technique shows, despite its limitations, potential as an adjunctive imaging technique during sinus surgery; however, the creation of a scoring system based on reproducible and defined characteristics in a larger more diverse population should be the focus of further research to improve its diagnostic value and clinical utility.
Level of Evidence
NA
Laryngoscope
, 2024},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Objective
Confocal laser endomicroscopy (CLE) is an optical imaging technique that allows in vivo, real‐time, microscope‐like assessment of superficial lesions. Although there is substantial data on CLE use in the upper GI tract, there is limited information regarding its application in the nasal cavity and paranasal sinuses. This study aims to assess the feasibility and diagnostic metrics of CLE in the nasal cavity and paranasal sinuses regarding differentiation between healthy/benign and malignant tissue. These structures show, however, a wider variety of frequent and concomitant benign and malignant pathologies, which could pose an increased challenge for optical biopsy by CLE.
Methods
We performed CLE on a case series of six patients with various findings in the nose (three chronic rhinosinusitis, adenocarcinoma, meningoenzephalozele, esthesionneuroblastoma). Forty‐two sequences (3792 images) from various structures in the nasal cavity and/or paranasal sinuses were acquired. Biopsies were taken at corresponding locations and analyzed in hematoxylin and eosin staining as a standard of reference. Three independent examiners blinded to the histopathology assessed the sequences.
Results
Healthy and inflamed mucosa could be distinguished from malignant lesions with an accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of 84.1%, 85.4%, 83.1%, 72.5%, and 92.1%, respectively, with a substantial agreement between raters (Fleiss
κ = 0.62).
Conclusion
This technique shows, despite its limitations, potential as an adjunctive imaging technique during sinus surgery; however, the creation of a scoring system based on reproducible and defined characteristics in a larger more diverse population should be the focus of further research to improve its diagnostic value and clinical utility.
Level of Evidence
NA
Laryngoscope
, 2024
Confocal laser endomicroscopy (CLE) is an optical imaging technique that allows in vivo, real‐time, microscope‐like assessment of superficial lesions. Although there is substantial data on CLE use in the upper GI tract, there is limited information regarding its application in the nasal cavity and paranasal sinuses. This study aims to assess the feasibility and diagnostic metrics of CLE in the nasal cavity and paranasal sinuses regarding differentiation between healthy/benign and malignant tissue. These structures show, however, a wider variety of frequent and concomitant benign and malignant pathologies, which could pose an increased challenge for optical biopsy by CLE.
Methods
We performed CLE on a case series of six patients with various findings in the nose (three chronic rhinosinusitis, adenocarcinoma, meningoenzephalozele, esthesionneuroblastoma). Forty‐two sequences (3792 images) from various structures in the nasal cavity and/or paranasal sinuses were acquired. Biopsies were taken at corresponding locations and analyzed in hematoxylin and eosin staining as a standard of reference. Three independent examiners blinded to the histopathology assessed the sequences.
Results
Healthy and inflamed mucosa could be distinguished from malignant lesions with an accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of 84.1%, 85.4%, 83.1%, 72.5%, and 92.1%, respectively, with a substantial agreement between raters (Fleiss
κ = 0.62).
Conclusion
This technique shows, despite its limitations, potential as an adjunctive imaging technique during sinus surgery; however, the creation of a scoring system based on reproducible and defined characteristics in a larger more diverse population should be the focus of further research to improve its diagnostic value and clinical utility.
Level of Evidence
NA
Laryngoscope
, 2024
67.
Pernias, Pablo; Rampas, Dominic; Richter, Mats L.; Pal, Christopher; Aubreville, Marc
Würstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models Proceedings Article
In: The Twelfth International Conference on Learning Representations, Vienna, Austria, 2024, (arXiv:2306.00637 [cs]).
Abstract | Links | BibTeX | Tags:
@inproceedings{pernias_wurstchen_2024,
title = {Würstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models},
author = {Pablo Pernias and Dominic Rampas and Mats L. Richter and Christopher Pal and Marc Aubreville},
url = {https://openreview.net/forum?id=gU58d5QeGv},
year = {2024},
date = {2024-05-01},
urldate = {2023-07-03},
booktitle = {The Twelfth International Conference on Learning Representations},
address = {Vienna, Austria},
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.},
note = {arXiv:2306.00637 [cs]},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
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.
66.
Aubreville, Marc; Stathonikos, Nikolas; Donovan, Taryn A.; Klopfleisch, Robert; Ammeling, Jonas; Ganz, Jonathan; Wilm, Frauke; Veta, Mitko; Jabari, Samir; Eckstein, Markus; Annuscheit, Jonas; Krumnow, Christian; Bozaba, Engin; Çayır, Sercan; Gu, Hongyan; Chen, Xiang ‘Anthony’; Jahanifar, Mostafa; Shephard, Adam; Kondo, Satoshi; Kasai, Satoshi; Kotte, Sujatha; Saipradeep, V. G.; Lafarge, Maxime W.; Koelzer, Viktor H.; Wang, Ziyue; Zhang, Yongbing; Yang, Sen; Wang, Xiyue; Breininger, Katharina; Bertram, Christof A.
In: Medical Image Analysis, vol. 94, pp. 103155, 2024, ISSN: 13618415.
@article{aubreville_domain_2024,
title = {Domain generalization across tumor types, laboratories, and species — Insights from the 2022 edition of the Mitosis Domain Generalization Challenge},
author = {Marc Aubreville and Nikolas Stathonikos and Taryn A. Donovan and Robert Klopfleisch and Jonas Ammeling and Jonathan Ganz and Frauke Wilm and Mitko Veta and Samir Jabari and Markus Eckstein and Jonas Annuscheit and Christian Krumnow and Engin Bozaba and Sercan Çayır and Hongyan Gu and Xiang ‘Anthony’ Chen and Mostafa Jahanifar and Adam Shephard and Satoshi Kondo and Satoshi Kasai and Sujatha Kotte and V. G. Saipradeep and Maxime W. Lafarge and Viktor H. Koelzer and Ziyue Wang and Yongbing Zhang and Sen Yang and Xiyue Wang and Katharina Breininger and Christof A. Bertram},
url = {https://linkinghub.elsevier.com/retrieve/pii/S136184152400080X},
doi = {10.1016/j.media.2024.103155},
issn = {13618415},
year = {2024},
date = {2024-05-01},
urldate = {2024-03-27},
journal = {Medical Image Analysis},
volume = {94},
pages = {103155},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
65.
Krügel, Sebastian; Ammeling, Jonas; Aubreville, Marc; Fritz, Alexis; Kießig, Angelika; Uhl, Matthias
Perceived responsibility in AI-supported medicine Journal Article
In: AI & SOCIETY, 2024, ISSN: 0951-5666, 1435-5655.
Abstract | Links | BibTeX | Tags:
@article{krugel_perceived_2024-1,
title = {Perceived responsibility in AI-supported medicine},
author = {Sebastian Krügel and Jonas Ammeling and Marc Aubreville and Alexis Fritz and Angelika Kießig and Matthias Uhl},
url = {https://link.springer.com/10.1007/s00146-024-01972-6},
doi = {10.1007/s00146-024-01972-6},
issn = {0951-5666, 1435-5655},
year = {2024},
date = {2024-05-01},
urldate = {2024-05-01},
journal = {AI & SOCIETY},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
64.
Oetter, Nicolai; Pröll, Jonas; Sievert, Matti; Goncalves, Miguel; Rohde, Maximilian; Nobis, Christopher-Philipp; Knipfer, Christian; Aubreville, Marc; Pan, Zhaoya; Breininger, Katharina; Maier, Andreas; Kesting, Marco; Stelzle, Florian
Oral mucosa – an examination map for confocal laser endomicroscopy within the oral cavity: an experimental clinical study Journal Article
In: Clinical Oral Investigations, vol. 28, no. 5, pp. 266, 2024, ISSN: 1436-3771.
Abstract | Links | BibTeX | Tags:
@article{oetter_oral_2024,
title = {Oral mucosa - an examination map for confocal laser endomicroscopy within the oral cavity: an experimental clinical study},
author = {Nicolai Oetter and Jonas Pröll and Matti Sievert and Miguel Goncalves and Maximilian Rohde and Christopher-Philipp Nobis and Christian Knipfer and Marc Aubreville and Zhaoya Pan and Katharina Breininger and Andreas Maier and Marco Kesting and Florian Stelzle},
url = {https://link.springer.com/10.1007/s00784-024-05664-9},
doi = {10.1007/s00784-024-05664-9},
issn = {1436-3771},
year = {2024},
date = {2024-04-01},
urldate = {2024-04-24},
journal = {Clinical Oral Investigations},
volume = {28},
number = {5},
pages = {266},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
63.
Sievert, Matti; Aubreville, Marc; Mueller, Sarina Katrin; Eckstein, Markus; Breininger, Katharina; Iro, Heinrich; Goncalves, Miguel
Diagnosis of malignancy in oropharyngeal confocal laser endomicroscopy using GPT 4.0 with vision Journal Article
In: European Archives of Oto-Rhino-Laryngology, 2024, ISSN: 0937-4477, 1434-4726.
@article{sievert_diagnosis_2024,
title = {Diagnosis of malignancy in oropharyngeal confocal laser endomicroscopy using GPT 4.0 with vision},
author = {Matti Sievert and Marc Aubreville and Sarina Katrin Mueller and Markus Eckstein and Katharina Breininger and Heinrich Iro and Miguel Goncalves},
url = {https://link.springer.com/10.1007/s00405-024-08476-5},
doi = {10.1007/s00405-024-08476-5},
issn = {0937-4477, 1434-4726},
year = {2024},
date = {2024-02-01},
urldate = {2024-02-21},
journal = {European Archives of Oto-Rhino-Laryngology},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
62.
Aubreville, Marc; Pan, Zhaoya; Sievert, Matti; Ammeling, Jonas; Ganz, Jonathan; Oetter, Nicolai; Stelzle, Florian; Frenken, Ann-Kathrin; Breininger, Katharina; Goncalves, Miguel
Few Shot Learning for the Classification of Confocal Laser Endomicroscopy Images of Head and Neck Tumors Book Section
In: Maier, Andreas; Deserno, Thomas M.; Handels, Heinz; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2024, pp. 143–148, Springer Fachmedien Wiesbaden, Wiesbaden, 2024, ISBN: 9783658440367 9783658440374.
@incollection{maier_few_2024,
title = {Few Shot Learning for the Classification of Confocal Laser Endomicroscopy Images of Head and Neck Tumors},
author = {Marc Aubreville and Zhaoya Pan and Matti Sievert and Jonas Ammeling and Jonathan Ganz and Nicolai Oetter and Florian Stelzle and Ann-Kathrin Frenken and Katharina Breininger and Miguel Goncalves},
editor = {Andreas Maier and Thomas M. Deserno and Heinz Handels and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-44037-4_42},
doi = {10.1007/978-3-658-44037-4_42},
isbn = {9783658440367 9783658440374},
year = {2024},
date = {2024-01-01},
urldate = {2024-02-21},
booktitle = {Bildverarbeitung für die Medizin 2024},
pages = {143–148},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
61.
Ammeling, Jonas; Hecker, Moritz; Ganz, Jonathan; Donovan, Taryn A.; Klopfleisch, Robert; Bertram, Christof A.; Breininger, Katharina; Aubreville, Marc
Automated Mitotic Index Calculation via Deep Learning and Immunohistochemistry Book Section
In: Maier, Andreas; Deserno, Thomas M.; Handels, Heinz; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2024, pp. 123–128, Springer Fachmedien Wiesbaden, Wiesbaden, 2024, ISBN: 9783658440367 9783658440374.
@incollection{maier_automated_2024,
title = {Automated Mitotic Index Calculation via Deep Learning and Immunohistochemistry},
author = {Jonas Ammeling and Moritz Hecker and Jonathan Ganz and Taryn A. Donovan and Robert Klopfleisch and Christof A. Bertram and Katharina Breininger and Marc Aubreville},
editor = {Andreas Maier and Thomas M. Deserno and Heinz Handels and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-44037-4_37},
doi = {10.1007/978-3-658-44037-4_37},
isbn = {9783658440367 9783658440374},
year = {2024},
date = {2024-01-01},
urldate = {2024-02-21},
booktitle = {Bildverarbeitung für die Medizin 2024},
pages = {123–128},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
60.
Ganz, Jonathan; Puget, Chloé; Ammeling, Jonas; Parlak, Eda; Kiupel, Matti; Bertram, Christof A.; Breininger, Katharina; Klopfleisch, Robert; Aubreville, Marc
Assessment of Scanner Domain Shifts in Deep Multiple Instance Learning Book Section
In: Maier, Andreas; Deserno, Thomas M.; Handels, Heinz; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2024, pp. 137–142, Springer Fachmedien Wiesbaden, Wiesbaden, 2024, ISBN: 9783658440367 9783658440374.
@incollection{maier_assessment_2024,
title = {Assessment of Scanner Domain Shifts in Deep Multiple Instance Learning},
author = {Jonathan Ganz and Chloé Puget and Jonas Ammeling and Eda Parlak and Matti Kiupel and Christof A. Bertram and Katharina Breininger and Robert Klopfleisch and Marc Aubreville},
editor = {Andreas Maier and Thomas M. Deserno and Heinz Handels and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-44037-4_41},
doi = {10.1007/978-3-658-44037-4_41},
isbn = {9783658440367 9783658440374},
year = {2024},
date = {2024-01-01},
urldate = {2024-02-21},
booktitle = {Bildverarbeitung für die Medizin 2024},
pages = {137–142},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
59.
Aubreville, Marc
Developing Robust AI Applications for Clinical Use: The Special Case of Pathology Journal Article
In: Trillium Pathology, vol. TP 1/2004, 2024.
@article{aubreville_developing_2024,
title = {Developing Robust AI Applications for Clinical Use: The Special Case of Pathology},
author = {Marc Aubreville},
doi = {https://doi.org/10.47184/tp.2024.01.04},
year = {2024},
date = {2024-01-01},
journal = {Trillium Pathology},
volume = {TP 1/2004},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
58.
Qiu, Jingna; Aubreville, Marc; Wilm, Frauke; Öttl, Mathias; Utz, Jonas; Schlereth, Maja; Breininger, Katharina
Leveraging image captions for selective whole slide image annotation Proceedings Article
In: Proceedings of MICCAI, 2024.
Abstract | Links | BibTeX | Tags:
@inproceedings{qiu_leveraging_2024,
title = {Leveraging image captions for selective whole slide image annotation},
author = {Jingna Qiu and Marc Aubreville and Frauke Wilm and Mathias Öttl and Jonas Utz and Maja Schlereth and Katharina Breininger},
url = {https://arxiv.org/abs/2407.06363},
doi = {10.48550/ARXIV.2407.06363},
year = {2024},
date = {2024-01-01},
urldate = {2024-09-26},
booktitle = {Proceedings of MICCAI},
abstract = {Acquiring annotations for whole slide images (WSIs)-based deep learning tasks, such as creating tissue segmentation masks or detecting mitotic figures, is a laborious process due to the extensive image size and the significant manual work involved in the annotation. This paper focuses on identifying and annotating specific image regions that optimize model training, given a limited annotation budget. While random sampling helps capture data variance by collecting annotation regions throughout the WSIs, insufficient data curation may result in an inadequate representation of minority classes. Recent studies proposed diversity sampling to select a set of regions that maximally represent unique characteristics of the WSIs. This is done by pretraining on unlabeled data through self-supervised learning and then clustering all regions in the latent space. However, establishing the optimal number of clusters can be difficult and not all clusters are task-relevant. This paper presents prototype sampling, a new method for annotation region selection. It discovers regions exhibiting typical characteristics of each task-specific class. The process entails recognizing class prototypes from extensive histopathology image-caption databases and detecting unlabeled image regions that resemble these prototypes. Our results show that prototype sampling is more effective than random and diversity sampling in identifying annotation regions with valuable training information, resulting in improved model performance in semantic segmentation and mitotic figure detection tasks. Code is available at https://github.com/DeepMicroscopy/Prototype-sampling.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Acquiring annotations for whole slide images (WSIs)-based deep learning tasks, such as creating tissue segmentation masks or detecting mitotic figures, is a laborious process due to the extensive image size and the significant manual work involved in the annotation. This paper focuses on identifying and annotating specific image regions that optimize model training, given a limited annotation budget. While random sampling helps capture data variance by collecting annotation regions throughout the WSIs, insufficient data curation may result in an inadequate representation of minority classes. Recent studies proposed diversity sampling to select a set of regions that maximally represent unique characteristics of the WSIs. This is done by pretraining on unlabeled data through self-supervised learning and then clustering all regions in the latent space. However, establishing the optimal number of clusters can be difficult and not all clusters are task-relevant. This paper presents prototype sampling, a new method for annotation region selection. It discovers regions exhibiting typical characteristics of each task-specific class. The process entails recognizing class prototypes from extensive histopathology image-caption databases and detecting unlabeled image regions that resemble these prototypes. Our results show that prototype sampling is more effective than random and diversity sampling in identifying annotation regions with valuable training information, resulting in improved model performance in semantic segmentation and mitotic figure detection tasks. Code is available at https://github.com/DeepMicroscopy/Prototype-sampling.
2023
57.
Ammeling, Jonas; Manger, Carina; Kwaka, Elias; Krügel, Sebastian; Uhl, Matthias; Kießig, Angelika; Fritz, Alexis; Ganz, Jonathan; Riener, Andreas; Bertram, Christof A.; Breininger, Katharina; Aubreville, Marc
Appealing but Potentially Biasing – Investigation of the Visual Representation of Segmentation Predictions by AI Recommender Systems for Medical Decision Making Proceedings Article
In: MuC ’23: Mensch und Computer 2023, pp. 330–335, ACM, Rapperswil Switzerland, 2023, ISBN: 9798400707711.
@inproceedings{ammeling_appealing_2023,
title = {Appealing but Potentially Biasing - Investigation of the Visual Representation of Segmentation Predictions by AI Recommender Systems for Medical Decision Making},
author = {Jonas Ammeling and Carina Manger and Elias Kwaka and Sebastian Krügel and Matthias Uhl and Angelika Kießig and Alexis Fritz and Jonathan Ganz and Andreas Riener and Christof A. Bertram and Katharina Breininger and Marc Aubreville},
url = {https://dl.acm.org/doi/10.1145/3603555.3608561},
doi = {10.1145/3603555.3608561},
isbn = {9798400707711},
year = {2023},
date = {2023-09-01},
urldate = {2023-09-04},
booktitle = {MuC '23: Mensch und Computer 2023},
pages = {330–335},
publisher = {ACM},
address = {Rapperswil Switzerland},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
56.
Fragoso-Garcia, Marco; Wilm, Frauke; Bertram, Christof A.; Merz, Sophie; Schmidt, Anja; Donovan, Taryn; Fuchs-Baumgartinger, Andrea; Bartel, Alexander; Marzahl, Christian; Diehl, Laura; Puget, Chloe; Maier, Andreas; Aubreville, Marc; Breininger, Katharina; Klopfleisch, Robert
Automated diagnosis of 7 canine skin tumors using machine learning on H&E-stained whole slide images Journal Article
In: Veterinary Pathology, pp. 03009858231189205, 2023, ISSN: 0300-9858, 1544-2217.
Abstract | Links | BibTeX | Tags:
@article{fragoso-garcia_automated_2023,
title = {Automated diagnosis of 7 canine skin tumors using machine learning on H&E-stained whole slide images},
author = {Marco Fragoso-Garcia and Frauke Wilm and Christof A. Bertram and Sophie Merz and Anja Schmidt and Taryn Donovan and Andrea Fuchs-Baumgartinger and Alexander Bartel and Christian Marzahl and Laura Diehl and Chloe Puget and Andreas Maier and Marc Aubreville and Katharina Breininger and Robert Klopfleisch},
url = {http://journals.sagepub.com/doi/10.1177/03009858231189205},
doi = {10.1177/03009858231189205},
issn = {0300-9858, 1544-2217},
year = {2023},
date = {2023-07-01},
urldate = {2023-08-07},
journal = {Veterinary Pathology},
pages = {03009858231189205},
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},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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
55.
Hirling, Dominik; Tasnadi, Ervin; Caicedo, Juan; Caroprese, Maria V.; Sjögren, Rickard; Aubreville, Marc; Koos, Krisztian; Horvath, Peter
Segmentation metric misinterpretations in bioimage analysis Journal Article
In: Nature Methods, 2023, ISSN: 1548-7091, 1548-7105.
Abstract | Links | BibTeX | Tags:
@article{hirling_segmentation_2023,
title = {Segmentation metric misinterpretations in bioimage analysis},
author = {Dominik Hirling and Ervin Tasnadi and Juan Caicedo and Maria V. Caroprese and Rickard Sjögren and Marc Aubreville and Krisztian Koos and Peter Horvath},
url = {https://www.nature.com/articles/s41592-023-01942-8},
doi = {10.1038/s41592-023-01942-8},
issn = {1548-7091, 1548-7105},
year = {2023},
date = {2023-07-01},
urldate = {2023-07-28},
journal = {Nature Methods},
abstract = {Abstract
Quantitative evaluation of image segmentation algorithms is crucial in the field of bioimage analysis. The most common assessment scores, however, are often misinterpreted and multiple definitions coexist with the same name. Here we present the ambiguities of evaluation metrics for segmentation algorithms and show how these misinterpretations can alter leaderboards of influential competitions. We also propose guidelines for how the currently existing problems could be tackled.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Abstract
Quantitative evaluation of image segmentation algorithms is crucial in the field of bioimage analysis. The most common assessment scores, however, are often misinterpreted and multiple definitions coexist with the same name. Here we present the ambiguities of evaluation metrics for segmentation algorithms and show how these misinterpretations can alter leaderboards of influential competitions. We also propose guidelines for how the currently existing problems could be tackled.
Quantitative evaluation of image segmentation algorithms is crucial in the field of bioimage analysis. The most common assessment scores, however, are often misinterpreted and multiple definitions coexist with the same name. Here we present the ambiguities of evaluation metrics for segmentation algorithms and show how these misinterpretations can alter leaderboards of influential competitions. We also propose guidelines for how the currently existing problems could be tackled.
54.
Aubreville, Marc; Wilm, Frauke; Stathonikos, Nikolas; Breininger, Katharina; Donovan, Taryn A.; Jabari, Samir; Veta, Mitko; Ganz, Jonathan; Ammeling, Jonas; Diest, Paul J. Van; Klopfleisch, Robert; Bertram, Christof A.
A comprehensive multi-domain dataset for mitotic figure detection Journal Article
In: Scientific Data, vol. 10, no. 1, pp. 484, 2023, ISSN: 2052-4463.
Abstract | Links | BibTeX | Tags:
@article{aubreville_comprehensive_2023,
title = {A comprehensive multi-domain dataset for mitotic figure detection},
author = {Marc Aubreville and Frauke Wilm and Nikolas Stathonikos and Katharina Breininger and Taryn A. Donovan and Samir Jabari and Mitko Veta and Jonathan Ganz and Jonas Ammeling and Paul J. Van Diest and Robert Klopfleisch and Christof A. Bertram},
url = {https://www.nature.com/articles/s41597-023-02327-4},
doi = {10.1038/s41597-023-02327-4},
issn = {2052-4463},
year = {2023},
date = {2023-07-01},
urldate = {2023-07-26},
journal = {Scientific Data},
volume = {10},
number = {1},
pages = {484},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
53.
Sheng, Bin; Aubreville, Marc (Ed.)
Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-33657-7 978-3-031-33658-4.
@book{sheng_mitosis_2023,
title = {Mitosis Domain Generalization and Diabetic Retinopathy Analysis: MICCAI Challenges MIDOG 2022 and DRAC 2022, Held in Conjunction with MICCAI 2022, Singapore, September 18–22, 2022, Proceedings},
editor = {Bin Sheng and Marc Aubreville},
url = {https://link.springer.com/10.1007/978-3-031-33658-4},
doi = {10.1007/978-3-031-33658-4},
isbn = {978-3-031-33657-7 978-3-031-33658-4},
year = {2023},
date = {2023-05-01},
urldate = {2023-07-01},
volume = {13597},
publisher = {Springer Nature Switzerland},
address = {Cham},
series = {Lecture Notes in Computer Science},
keywords = {},
pubstate = {published},
tppubtype = {book}
}
52.
Rampas, Dominic; Pernias, Pablo; Aubreville, Marc
A Novel Sampling Scheme for Text- and Image-Conditional Image Synthesis in Quantized Latent Spaces Miscellaneous
2023, (arXiv:2211.07292 [cs]).
Abstract | Links | BibTeX | Tags:
@misc{rampas_novel_2023,
title = {A Novel Sampling Scheme for Text- and Image-Conditional Image Synthesis in Quantized Latent Spaces},
author = {Dominic Rampas and Pablo Pernias and Marc Aubreville},
url = {http://arxiv.org/abs/2211.07292},
year = {2023},
date = {2023-05-01},
urldate = {2023-07-03},
abstract = {Recent advancements in the domain of text-to-image synthesis have culminated in a multitude of enhancements pertaining to quality, fidelity, and diversity. Contemporary techniques enable the generation of highly intricate visuals which rapidly approach near-photorealistic quality. Nevertheless, as progress is achieved, the complexity of these methodologies increases, consequently intensifying the comprehension barrier between individuals within the field and those external to it. In an endeavor to mitigate this disparity, we propose a streamlined approach for text-to-image generation, which encompasses both the training paradigm and the sampling process. Despite its remarkable simplicity, our method yields aesthetically pleasing images with few sampling iterations, allows for intriguing ways for conditioning the model, and imparts advantages absent in state-of-the-art techniques. To demonstrate the efficacy of this approach in achieving outcomes comparable to existing works, we have trained a one-billion parameter text-conditional model, which we refer to as "Paella". In the interest of fostering future exploration in this field, we have made our source code and models publicly accessible for the research community.},
note = {arXiv:2211.07292 [cs]},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Recent advancements in the domain of text-to-image synthesis have culminated in a multitude of enhancements pertaining to quality, fidelity, and diversity. Contemporary techniques enable the generation of highly intricate visuals which rapidly approach near-photorealistic quality. Nevertheless, as progress is achieved, the complexity of these methodologies increases, consequently intensifying the comprehension barrier between individuals within the field and those external to it. In an endeavor to mitigate this disparity, we propose a streamlined approach for text-to-image generation, which encompasses both the training paradigm and the sampling process. Despite its remarkable simplicity, our method yields aesthetically pleasing images with few sampling iterations, allows for intriguing ways for conditioning the model, and imparts advantages absent in state-of-the-art techniques. To demonstrate the efficacy of this approach in achieving outcomes comparable to existing works, we have trained a one-billion parameter text-conditional model, which we refer to as “Paella”. In the interest of fostering future exploration in this field, we have made our source code and models publicly accessible for the research community.
51.
Ammeling, Jonas; Schmidt, Lars-Henning; Ganz, Jonathan; Niedermair, Tanja; Brochhausen-Delius, Christoph; Schulz, Christian; Breininger, Katharina; Aubreville, Marc
Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays Book Section
In: Deserno, Thomas M.; Handels, Heinz; Maier, Andreas; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2023, pp. 220–225, Springer Fachmedien Wiesbaden, Wiesbaden, 2023, ISBN: 978-3-658-41656-0 978-3-658-41657-7, (Series Title: Informatik aktuell).
@incollection{deserno_attention-based_2023,
title = {Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays},
author = {Jonas Ammeling and Lars-Henning Schmidt and Jonathan Ganz and Tanja Niedermair and Christoph Brochhausen-Delius and Christian Schulz and Katharina Breininger and Marc Aubreville},
editor = {Thomas M. Deserno and Heinz Handels and Andreas Maier and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-41657-7_48},
doi = {10.1007/978-3-658-41657-7_48},
isbn = {978-3-658-41656-0 978-3-658-41657-7},
year = {2023},
date = {2023-02-01},
urldate = {2023-06-30},
booktitle = {Bildverarbeitung für die Medizin 2023},
pages = {220–225},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
note = {Series Title: Informatik aktuell},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
50.
Aubreville, Marc; Ganz, Jonathan; Ammeling, Jonas; Donovan, Taryn A.; Fick, Rutger Hj.; Breininger, Katharina; Bertram, Christof A.
Deep Learning-based Subtyping of Atypical and Normal Mitoses using a Hierarchical Anchor-free Object Detector Book Section
In: Deserno, Thomas M.; Handels, Heinz; Maier, Andreas; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2023, pp. 189–195, Springer Fachmedien Wiesbaden, Wiesbaden, 2023, ISBN: 978-3-658-41656-0 978-3-658-41657-7, (Series Title: Informatik aktuell).
@incollection{deserno_deep_2023,
title = {Deep Learning-based Subtyping of Atypical and Normal Mitoses using a Hierarchical Anchor-free Object Detector},
author = {Marc Aubreville and Jonathan Ganz and Jonas Ammeling and Taryn A. Donovan and Rutger Hj. Fick and Katharina Breininger and Christof A. Bertram},
editor = {Thomas M. Deserno and Heinz Handels and Andreas Maier and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-41657-7_40},
doi = {10.1007/978-3-658-41657-7_40},
isbn = {978-3-658-41656-0 978-3-658-41657-7},
year = {2023},
date = {2023-02-01},
urldate = {2023-06-30},
booktitle = {Bildverarbeitung für die Medizin 2023},
pages = {189–195},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
note = {Series Title: Informatik aktuell},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
49.
Ganz, Jonathan; Lipnik, Karoline; Ammeling, Jonas; Richter, Barbara; Puget, Chloé; Parlak, Eda; Diehl, Laura; Klopfleisch, Robert; Donovan, Taryn A.; Kiupel, Matti; Bertram, Christof A.; Breininger, Katharina; Aubreville, Marc
Deep Learning-based Automatic Assessment of AgNOR-scores in Histopathology Images Book Section
In: Deserno, Thomas M.; Handels, Heinz; Maier, Andreas; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2023, pp. 226–231, Springer Fachmedien Wiesbaden, Wiesbaden, 2023, ISBN: 978-3-658-41656-0 978-3-658-41657-7, (Series Title: Informatik aktuell).
@incollection{deserno_deep_2023-1,
title = {Deep Learning-based Automatic Assessment of AgNOR-scores in Histopathology Images},
author = {Jonathan Ganz and Karoline Lipnik and Jonas Ammeling and Barbara Richter and Chloé Puget and Eda Parlak and Laura Diehl and Robert Klopfleisch and Taryn A. Donovan and Matti Kiupel and Christof A. Bertram and Katharina Breininger and Marc Aubreville},
editor = {Thomas M. Deserno and Heinz Handels and Andreas Maier and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-41657-7_49},
doi = {10.1007/978-3-658-41657-7_49},
isbn = {978-3-658-41656-0 978-3-658-41657-7},
year = {2023},
date = {2023-02-01},
urldate = {2023-06-30},
booktitle = {Bildverarbeitung für die Medizin 2023},
pages = {226–231},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
note = {Series Title: Informatik aktuell},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
48.
Lausser, Ludwig M.; Bertram, Christof A.; Klopfleisch, Robert; Aubreville, Marc
Limits of Human Expert Ensembles in Mitosis Multi-expert Ground Truth Generation Book Section
In: Deserno, Thomas M.; Handels, Heinz; Maier, Andreas; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2023, pp. 116–121, Springer Fachmedien Wiesbaden, Wiesbaden, 2023, ISBN: 978-3-658-41656-0 978-3-658-41657-7, (Series Title: Informatik aktuell).
@incollection{deserno_limits_2023,
title = {Limits of Human Expert Ensembles in Mitosis Multi-expert Ground Truth Generation},
author = {Ludwig M. Lausser and Christof A. Bertram and Robert Klopfleisch and Marc Aubreville},
editor = {Thomas M. Deserno and Heinz Handels and Andreas Maier and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-41657-7_27},
doi = {10.1007/978-3-658-41657-7_27},
isbn = {978-3-658-41656-0 978-3-658-41657-7},
year = {2023},
date = {2023-02-01},
urldate = {2023-06-30},
booktitle = {Bildverarbeitung für die Medizin 2023},
pages = {116–121},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
note = {Series Title: Informatik aktuell},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
47.
Wilm, Frauke; Fragoso, Marco; Bertram, Christof A.; Stathonikos, Nikolas; Öttl, Mathias; Qiu, Jingna; Klopfleisch, Robert; Maier, Andreas; Breininger, Katharina; Aubreville, Marc
Multi-scanner Canine Cutaneous Squamous Cell Carcinoma Histopathology Dataset Book Section
In: Deserno, Thomas M.; Handels, Heinz; Maier, Andreas; Maier-Hein, Klaus; Palm, Christoph; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2023, pp. 206–211, Springer Fachmedien Wiesbaden, Wiesbaden, 2023, ISBN: 978-3-658-41656-0 978-3-658-41657-7, (Series Title: Informatik aktuell).
@incollection{deserno_multi-scanner_2023,
title = {Multi-scanner Canine Cutaneous Squamous Cell Carcinoma Histopathology Dataset},
author = {Frauke Wilm and Marco Fragoso and Christof A. Bertram and Nikolas Stathonikos and Mathias Öttl and Jingna Qiu and Robert Klopfleisch and Andreas Maier and Katharina Breininger and Marc Aubreville},
editor = {Thomas M. Deserno and Heinz Handels and Andreas Maier and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff},
url = {https://link.springer.com/10.1007/978-3-658-41657-7_46},
doi = {10.1007/978-3-658-41657-7_46},
isbn = {978-3-658-41656-0 978-3-658-41657-7},
year = {2023},
date = {2023-02-01},
urldate = {2023-06-30},
booktitle = {Bildverarbeitung für die Medizin 2023},
pages = {206–211},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
note = {Series Title: Informatik aktuell},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
46.
Wilm, Frauke; Ihling, Christian; Méhes, Gábor; Terracciano, Luigi; Puget, Chloé; Klopfleisch, Robert; Schüffler, Peter; Aubreville, Marc; Maier, Andreas; Mrowiec, Thomas; Breininger, Katharina
Pan-tumor T-lymphocyte detection using deep neural networks: Recommendations for transfer learning in immunohistochemistry Journal Article
In: Journal of Pathology Informatics, vol. 14, pp. 100301, 2023, ISSN: 21533539.
@article{wilm_pan-tumor_2023,
title = {Pan-tumor T-lymphocyte detection using deep neural networks: Recommendations for transfer learning in immunohistochemistry},
author = {Frauke Wilm and Christian Ihling and Gábor Méhes and Luigi Terracciano and Chloé Puget and Robert Klopfleisch and Peter Schüffler and Marc Aubreville and Andreas Maier and Thomas Mrowiec and Katharina Breininger},
url = {https://linkinghub.elsevier.com/retrieve/pii/S2153353923001153},
doi = {10.1016/j.jpi.2023.100301},
issn = {21533539},
year = {2023},
date = {2023-02-01},
urldate = {2023-07-01},
journal = {Journal of Pathology Informatics},
volume = {14},
pages = {100301},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
45.
Qiu, Jingna; Wilm, Frauke; Öttl, Mathias; Schlereth, Maja; Liu, Chang; Heimann, Tobias; Aubreville, Marc; Breininger, Katharina
Adaptive Region Selection for Active Learning in Whole Slide Image Semantic Segmentation Book Section
In: Greenspan, Hayit; Madabhushi, Anant; Mousavi, Parvin; Salcudean, Septimiu; Duncan, James; Syeda-Mahmood, Tanveer; Taylor, Russell (Ed.): Medical Image Computing and Computer Assisted Intervention – MICCAI 2023, vol. 14221, pp. 90–100, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-43894-3 978-3-031-43895-0, (Series Title: Lecture Notes in Computer Science).
@incollection{greenspan_adaptive_2023,
title = {Adaptive Region Selection for Active Learning in Whole Slide Image Semantic Segmentation},
author = {Jingna Qiu and Frauke Wilm and Mathias Öttl and Maja Schlereth and Chang Liu and Tobias Heimann and Marc Aubreville and Katharina Breininger},
editor = {Hayit Greenspan and Anant Madabhushi and Parvin Mousavi and Septimiu Salcudean and James Duncan and Tanveer Syeda-Mahmood and Russell Taylor},
url = {https://link.springer.com/10.1007/978-3-031-43895-0_9},
doi = {10.1007/978-3-031-43895-0_9},
isbn = {978-3-031-43894-3 978-3-031-43895-0},
year = {2023},
date = {2023-01-01},
urldate = {2023-11-20},
booktitle = {Medical Image Computing and Computer Assisted Intervention – MICCAI 2023},
volume = {14221},
pages = {90–100},
publisher = {Springer Nature Switzerland},
address = {Cham},
note = {Series Title: Lecture Notes in Computer Science},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
44.
Aubreville, Marc; Stathonikos, Nikolas; Bertram, Christof A.; Klopfleisch, Robert; Hoeve, Natalie Ter; Ciompi, Francesco; Wilm, Frauke; Marzahl, Christian; Donovan, Taryn A.; Maier, Andreas; Breen, Jack; Ravikumar, Nishant; Chung, Youjin; Park, Jinah; Nateghi, Ramin; Pourakpour, Fattaneh; Fick, Rutger H. J.; Hadj, Saima Ben; Jahanifar, Mostafa; Shephard, Adam; Dexl, Jakob; Wittenberg, Thomas; Kondo, Satoshi; Lafarge, Maxime W.; Koelzer, Viktor H.; Liang, Jingtang; Wang, Yubo; Long, Xi; Liu, Jingxin; Razavi, Salar; Khademi, April; Yang, Sen; Wang, Xiyue; Erber, Ramona; Klang, Andrea; Lipnik, Karoline; Bolfa, Pompei; Dark, Michael J.; Wasinger, Gabriel; Veta, Mitko; Breininger, Katharina
Mitosis domain generalization in histopathology images — The MIDOG challenge Journal Article
In: Medical Image Analysis, vol. 84, pp. 102699, 2023, ISSN: 13618415.
@article{aubreville_mitosis_2023,
title = {Mitosis domain generalization in histopathology images — The MIDOG challenge},
author = {Marc Aubreville and Nikolas Stathonikos and Christof A. Bertram and Robert Klopfleisch and Natalie Ter Hoeve and Francesco Ciompi and Frauke Wilm and Christian Marzahl and Taryn A. Donovan and Andreas Maier and Jack Breen and Nishant Ravikumar and Youjin Chung and Jinah Park and Ramin Nateghi and Fattaneh Pourakpour and Rutger H. J. Fick and Saima Ben Hadj and Mostafa Jahanifar and Adam Shephard and Jakob Dexl and Thomas Wittenberg and Satoshi Kondo and Maxime W. Lafarge and Viktor H. Koelzer and Jingtang Liang and Yubo Wang and Xi Long and Jingxin Liu and Salar Razavi and April Khademi and Sen Yang and Xiyue Wang and Ramona Erber and Andrea Klang and Karoline Lipnik and Pompei Bolfa and Michael J. Dark and Gabriel Wasinger and Mitko Veta and Katharina Breininger},
url = {https://linkinghub.elsevier.com/retrieve/pii/S1361841522003279},
doi = {10.1016/j.media.2022.102699},
issn = {13618415},
year = {2023},
date = {2023-01-01},
urldate = {2023-05-19},
journal = {Medical Image Analysis},
volume = {84},
pages = {102699},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
43.
Pan, Zhaoya; Breininger, Katharina; Aubreville, Marc; Stelzle, Florian; Oetter, Nicolai; Maier, Andreas; Mantsopoulos, Konstantinos; Iro, Heinrich; Goncalves, Miguel; Sievert, Matti
Defining a baseline identification of artifacts in confocal laser endomicroscopy in head and neck cancer imaging Journal Article
In: American Journal of Otolaryngology, vol. 44, no. 2, pp. 103779, 2023, ISSN: 01960709.
@article{pan_defining_2023,
title = {Defining a baseline identification of artifacts in confocal laser endomicroscopy in head and neck cancer imaging},
author = {Zhaoya Pan and Katharina Breininger and Marc Aubreville and Florian Stelzle and Nicolai Oetter and Andreas Maier and Konstantinos Mantsopoulos and Heinrich Iro and Miguel Goncalves and Matti Sievert},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0196070922004069},
doi = {10.1016/j.amjoto.2022.103779},
issn = {01960709},
year = {2023},
date = {2023-01-01},
urldate = {2023-07-01},
journal = {American Journal of Otolaryngology},
volume = {44},
number = {2},
pages = {103779},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
42.
Ammeling, Jonas; Wilm, Frauke; Ganz, Jonathan; Breininger, Katharina; Aubreville, Marc
Reference Algorithms for the Mitosis Domain Generalization (MIDOG) 2022 Challenge Book Section
In: Sheng, Bin; Aubreville, Marc (Ed.): Mitosis Domain Generalization and Diabetic Retinopathy Analysis, vol. 13597, pp. 201–205, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-33657-7 978-3-031-33658-4, (Series Title: Lecture Notes in Computer Science).
@incollection{sheng_reference_2023,
title = {Reference Algorithms for the Mitosis Domain Generalization (MIDOG) 2022 Challenge},
author = {Jonas Ammeling and Frauke Wilm and Jonathan Ganz and Katharina Breininger and Marc Aubreville},
editor = {Bin Sheng and Marc Aubreville},
url = {https://link.springer.com/10.1007/978-3-031-33658-4_19},
doi = {10.1007/978-3-031-33658-4_19},
isbn = {978-3-031-33657-7 978-3-031-33658-4},
year = {2023},
date = {2023-01-01},
urldate = {2023-07-02},
booktitle = {Mitosis Domain Generalization and Diabetic Retinopathy Analysis},
volume = {13597},
pages = {201–205},
publisher = {Springer Nature Switzerland},
address = {Cham},
note = {Series Title: Lecture Notes in Computer Science},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
2022
41.
Sievert, Matti; Mantsopoulos, Konstantinos; Mueller, Sarina K.; Rupp, Robin; Eckstein, Markus; Stelzle, Florian; Oetter, Nicolai; Maier, Andreas; Aubreville, Marc; Iro, Heinrich; Goncalves, Miguel
In: Brazilian Journal of Otorhinolaryngology, vol. 88, pp. S26–S32, 2022, ISSN: 18088694.
@article{sievert_validation_2022,
title = {Validation of a classification and scoring system for the diagnosis of laryngeal and pharyngeal squamous cell carcinomas by confocal laser endomicroscopy},
author = {Matti Sievert and Konstantinos Mantsopoulos and Sarina K. Mueller and Robin Rupp and Markus Eckstein and Florian Stelzle and Nicolai Oetter and Andreas Maier and Marc Aubreville and Heinrich Iro and Miguel Goncalves},
url = {https://linkinghub.elsevier.com/retrieve/pii/S1808869421001245},
doi = {10.1016/j.bjorl.2021.06.002},
issn = {18088694},
year = {2022},
date = {2022-11-01},
urldate = {2023-06-30},
journal = {Brazilian Journal of Otorhinolaryngology},
volume = {88},
pages = {S26–S32},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
40.
Wilm, Frauke; Fragoso, Marco; Marzahl, Christian; Qiu, Jingna; Puget, Chloé; Diehl, Laura; Bertram, Christof A.; Klopfleisch, Robert; Maier, Andreas; Breininger, Katharina; Aubreville, Marc
Pan-tumor CAnine cuTaneous Cancer Histology (CATCH) dataset Journal Article
In: Scientific Data, vol. 9, no. 1, pp. 588, 2022, ISSN: 2052-4463.
Abstract | Links | BibTeX | Tags:
@article{wilm_pan-tumor_2022,
title = {Pan-tumor CAnine cuTaneous Cancer Histology (CATCH) dataset},
author = {Frauke Wilm and Marco Fragoso and Christian Marzahl and Jingna Qiu and Chloé Puget and Laura Diehl and Christof A. Bertram and Robert Klopfleisch and Andreas Maier and Katharina Breininger and Marc Aubreville},
url = {https://www.nature.com/articles/s41597-022-01692-w},
doi = {10.1038/s41597-022-01692-w},
issn = {2052-4463},
year = {2022},
date = {2022-09-01},
urldate = {2023-06-30},
journal = {Scientific Data},
volume = {9},
number = {1},
pages = {588},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
39.
Marzahl, Christian; Hill, Jenny; Stayt, Jason; Bienzle, Dorothee; Welker, Lutz; Wilm, Frauke; Voigt, Jörn; Aubreville, Marc; Maier, Andreas; Klopfleisch, Robert; Breininger, Katharina; Bertram, Christof A.
Inter-species cell detection – datasets on pulmonary hemosiderophages in equine, human and feline specimens Journal Article
In: Scientific Data, vol. 9, no. 1, pp. 269, 2022, ISSN: 2052-4463.
Abstract | Links | BibTeX | Tags:
@article{marzahl_inter-species_2022,
title = {Inter-species cell detection - datasets on pulmonary hemosiderophages in equine, human and feline specimens},
author = {Christian Marzahl and Jenny Hill and Jason Stayt and Dorothee Bienzle and Lutz Welker and Frauke Wilm and Jörn Voigt and Marc Aubreville and Andreas Maier and Robert Klopfleisch and Katharina Breininger and Christof A. Bertram},
url = {https://www.nature.com/articles/s41597-022-01389-0},
doi = {10.1038/s41597-022-01389-0},
issn = {2052-4463},
year = {2022},
date = {2022-06-01},
urldate = {2023-07-01},
journal = {Scientific Data},
volume = {9},
number = {1},
pages = {269},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
38.
Sievert, Matti; Eckstein, Markus; Mantsopoulos, Konstantinos; Mueller, Sarina K.; Stelzle, Florian; Aubreville, Marc; Oetter, Nicolai; Maier, Andreas; Iro, Heinrich; Goncalves, Miguel
In: European Archives of Oto-Rhino-Laryngology, vol. 279, no. 4, pp. 2029–2037, 2022, ISSN: 0937-4477, 1434-4726.
Abstract | Links | BibTeX | Tags:
@article{sievert_impact_2022,
title = {Impact of intraepithelial capillary loops and atypical vessels in confocal laser endomicroscopy for the diagnosis of laryngeal and hypopharyngeal squamous cell carcinoma},
author = {Matti Sievert and Markus Eckstein and Konstantinos Mantsopoulos and Sarina K. Mueller and Florian Stelzle and Marc Aubreville and Nicolai Oetter and Andreas Maier and Heinrich Iro and Miguel Goncalves},
url = {https://link.springer.com/10.1007/s00405-021-06954-8},
doi = {10.1007/s00405-021-06954-8},
issn = {0937-4477, 1434-4726},
year = {2022},
date = {2022-04-01},
urldate = {2023-06-30},
journal = {European Archives of Oto-Rhino-Laryngology},
volume = {279},
number = {4},
pages = {2029–2037},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
37.
Bertram, Christof A.; Aubreville, Marc; Donovan, Taryn A.; Bartel, Alexander; Wilm, Frauke; Marzahl, Christian; Assenmacher, Charles-Antoine; Becker, Kathrin; Bennett, Mark; Corner, Sarah; Cossic, Brieuc; Denk, Daniela; Dettwiler, Martina; Gonzalez, Beatriz Garcia; Gurtner, Corinne; Haverkamp, Ann-Kathrin; Heier, Annabelle; Lehmbecker, Annika; Merz, Sophie; Noland, Erica L.; Plog, Stephanie; Schmidt, Anja; Sebastian, Franziska; Sledge, Dodd G.; Smedley, Rebecca C.; Tecilla, Marco; Thaiwong, Tuddow; Fuchs-Baumgartinger, Andrea; Meuten, Donald J.; Breininger, Katharina; Kiupel, Matti; Maier, Andreas; Klopfleisch, Robert
Computer-assisted mitotic count using a deep learning–based algorithm improves interobserver reproducibility and accuracy Journal Article
In: Veterinary Pathology, vol. 59, no. 2, pp. 211–226, 2022, ISSN: 0300-9858, 1544-2217.
Abstract | Links | BibTeX | Tags:
@article{bertram_computer-assisted_2022,
title = {Computer-assisted mitotic count using a deep learning–based algorithm improves interobserver reproducibility and accuracy},
author = {Christof A. Bertram and Marc Aubreville and Taryn A. Donovan and Alexander Bartel and Frauke Wilm and Christian Marzahl and Charles-Antoine Assenmacher and Kathrin Becker and Mark Bennett and Sarah Corner and Brieuc Cossic and Daniela Denk and Martina Dettwiler and Beatriz Garcia Gonzalez and Corinne Gurtner and Ann-Kathrin Haverkamp and Annabelle Heier and Annika Lehmbecker and Sophie Merz and Erica L. Noland and Stephanie Plog and Anja Schmidt and Franziska Sebastian and Dodd G. Sledge and Rebecca C. Smedley and Marco Tecilla and Tuddow Thaiwong and Andrea Fuchs-Baumgartinger and Donald J. Meuten and Katharina Breininger and Matti Kiupel and Andreas Maier and Robert Klopfleisch},
url = {http://journals.sagepub.com/doi/10.1177/03009858211067478},
doi = {10.1177/03009858211067478},
issn = {0300-9858, 1544-2217},
year = {2022},
date = {2022-03-01},
urldate = {2023-06-30},
journal = {Veterinary Pathology},
volume = {59},
number = {2},
pages = {211–226},
abstract = {The mitotic count (MC) is an important histological parameter for prognostication of malignant neoplasms. However, it has inter- and intraobserver discrepancies due to difficulties in selecting the region of interest (MC-ROI) and in identifying or classifying mitotic figures (MFs). Recent progress in the field of artificial intelligence has allowed the development of high-performance algorithms that may improve standardization of the MC. As algorithmic predictions are not flawless, computer-assisted review by pathologists may ensure reliability. In the present study, we compared partial (MC-ROI preselection) and full (additional visualization of MF candidates and display of algorithmic confidence values) computer-assisted MC analysis to the routine (unaided) MC analysis by 23 pathologists for whole-slide images of 50 canine cutaneous mast cell tumors (ccMCTs). Algorithmic predictions aimed to assist pathologists in detecting mitotic hotspot locations, reducing omission of MFs, and improving classification against imposters. The interobserver consistency for the MC significantly increased with computer assistance (interobserver correlation coefficient},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The mitotic count (MC) is an important histological parameter for prognostication of malignant neoplasms. However, it has inter- and intraobserver discrepancies due to difficulties in selecting the region of interest (MC-ROI) and in identifying or classifying mitotic figures (MFs). Recent progress in the field of artificial intelligence has allowed the development of high-performance algorithms that may improve standardization of the MC. As algorithmic predictions are not flawless, computer-assisted review by pathologists may ensure reliability. In the present study, we compared partial (MC-ROI preselection) and full (additional visualization of MF candidates and display of algorithmic confidence values) computer-assisted MC analysis to the routine (unaided) MC analysis by 23 pathologists for whole-slide images of 50 canine cutaneous mast cell tumors (ccMCTs). Algorithmic predictions aimed to assist pathologists in detecting mitotic hotspot locations, reducing omission of MFs, and improving classification against imposters. The interobserver consistency for the MC significantly increased with computer assistance (interobserver correlation coefficient
36.
Ganz, Jonathan; Bertram, Christof A.; Klopfleisch, Robert; Jabari, Samir; Breininger, Katharina; Aubreville, Marc
Classification of visibility in multi-stain microscopy images Proceedings Article
In: Medical Imaging with Deep Learning 2022, Zurich, 2022.
BibTeX | Tags:
@inproceedings{ganz_classification_2022,
title = {Classification of visibility in multi-stain microscopy images},
author = {Jonathan Ganz and Christof A. Bertram and Robert Klopfleisch and Samir Jabari and Katharina Breininger and Marc Aubreville},
year = {2022},
date = {2022-01-01},
booktitle = {Medical Imaging with Deep Learning 2022},
address = {Zurich},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
35.
Wilm, Frauke; Marzahl, Christian; Breininger, Katharina; Aubreville, Marc
Domain Adversarial RetinaNet as a Reference Algorithm for the MItosis DOmain Generalization Challenge Book Section
In: Aubreville, Marc; Zimmerer, David; Heinrich, Mattias (Ed.): Biomedical Image Registration, Domain Generalisation and Out-of-Distribution Analysis, vol. 13166, pp. 5–13, Springer International Publishing, Cham, 2022, ISBN: 978-3-030-97280-6 978-3-030-97281-3, (Series Title: Lecture Notes in Computer Science).
@incollection{aubreville_domain_2022,
title = {Domain Adversarial RetinaNet as a Reference Algorithm for the MItosis DOmain Generalization Challenge},
author = {Frauke Wilm and Christian Marzahl and Katharina Breininger and Marc Aubreville},
editor = {Marc Aubreville and David Zimmerer and Mattias Heinrich},
url = {https://link.springer.com/10.1007/978-3-030-97281-3_1},
doi = {10.1007/978-3-030-97281-3_1},
isbn = {978-3-030-97280-6 978-3-030-97281-3},
year = {2022},
date = {2022-01-01},
urldate = {2023-06-30},
booktitle = {Biomedical Image Registration, Domain Generalisation and Out-of-Distribution Analysis},
volume = {13166},
pages = {5–13},
publisher = {Springer International Publishing},
address = {Cham},
note = {Series Title: Lecture Notes in Computer Science},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
2021
34.
Sievert, Matti; Stelzle, Florian; Aubreville, Marc; Mueller, Sarina K.; Eckstein, Markus; Oetter, Nicolai; Maier, Andreas; Mantsopoulos, Konstantinos; Iro, Heinrich; Goncalves, Miguel
Intraoperative free margins assessment of oropharyngeal squamous cell carcinoma with confocal laser endomicroscopy: a pilot study Journal Article
In: European Archives of Oto-Rhino-Laryngology, vol. 278, no. 11, pp. 4433–4439, 2021, ISSN: 0937-4477, 1434-4726.
Abstract | Links | BibTeX | Tags:
@article{sievert_intraoperative_2021,
title = {Intraoperative free margins assessment of oropharyngeal squamous cell carcinoma with confocal laser endomicroscopy: a pilot study},
author = {Matti Sievert and Florian Stelzle and Marc Aubreville and Sarina K. Mueller and Markus Eckstein and Nicolai Oetter and Andreas Maier and Konstantinos Mantsopoulos and Heinrich Iro and Miguel Goncalves},
url = {https://link.springer.com/10.1007/s00405-021-06659-y},
doi = {10.1007/s00405-021-06659-y},
issn = {0937-4477, 1434-4726},
year = {2021},
date = {2021-11-01},
urldate = {2023-06-30},
journal = {European Archives of Oto-Rhino-Laryngology},
volume = {278},
number = {11},
pages = {4433–4439},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
33.
Theelke, Luisa; Wilm, Frauke; Marzahl, Christian; Bertram, Christof A.; Klopfleisch, Robert; Maier, Andreas; Aubreville, Marc; Breininger, Katharina
Iterative Cross-Scanner Registration for Whole Slide Images Proceedings Article
In: 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), pp. 582–590, IEEE, Montreal, BC, Canada, 2021, ISBN: 978-1-66540-191-3.
@inproceedings{theelke_iterative_2021,
title = {Iterative Cross-Scanner Registration for Whole Slide Images},
author = {Luisa Theelke and Frauke Wilm and Christian Marzahl and Christof A. Bertram and Robert Klopfleisch and Andreas Maier and Marc Aubreville and Katharina Breininger},
url = {https://ieeexplore.ieee.org/document/9607816/},
doi = {10.1109/ICCVW54120.2021.00071},
isbn = {978-1-66540-191-3},
year = {2021},
date = {2021-10-01},
urldate = {2023-06-30},
booktitle = {2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)},
pages = {582–590},
publisher = {IEEE},
address = {Montreal, BC, Canada},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
32.
Meuten, Donald J.; Moore, Frances M.; Donovan, Taryn A.; Bertram, Christof A.; Klopfleisch, Robert; Foster, Robert A.; Smedley, Rebecca C.; Dark, Michael J.; Milovancev, Milan; Stromberg, Paul; Williams, Bruce H.; Aubreville, Marc; Avallone, Giancarlo; Bolfa, Pompei; Cullen, John; Dennis, Michelle M.; Goldschmidt, Michael; Luong, Richard; Miller, Andrew D.; Miller, Margaret A.; Munday, John S.; Roccabianca, Paola; Salas, Elisa N.; Schulman, F. Yvonne; Laufer-Amorim, Renee; Asakawa, Midori G.; Craig, Linden; Dervisis, Nick; Esplin, D. Glen; George, Jeanne W.; Hauck, Marlene; Kagawa, Yumiko; Kiupel, Matti; Linder, Keith; Meichner, Kristina; Marconato, Laura; Oblak, Michelle L.; Santos, Renato L.; Simpson, R. Mark; Tvedten, Harold; Whitley, Derick
International Guidelines for Veterinary Tumor Pathology: A Call to Action Journal Article
In: Veterinary Pathology, vol. 58, no. 5, pp. 766–794, 2021, ISSN: 0300-9858, 1544-2217.
Abstract | Links | BibTeX | Tags:
@article{meuten_international_2021,
title = {International Guidelines for Veterinary Tumor Pathology: A Call to Action},
author = {Donald J. Meuten and Frances M. Moore and Taryn A. Donovan and Christof A. Bertram and Robert Klopfleisch and Robert A. Foster and Rebecca C. Smedley and Michael J. Dark and Milan Milovancev and Paul Stromberg and Bruce H. Williams and Marc Aubreville and Giancarlo Avallone and Pompei Bolfa and John Cullen and Michelle M. Dennis and Michael Goldschmidt and Richard Luong and Andrew D. Miller and Margaret A. Miller and John S. Munday and Paola Roccabianca and Elisa N. Salas and F. Yvonne Schulman and Renee Laufer-Amorim and Midori G. Asakawa and Linden Craig and Nick Dervisis and D. Glen Esplin and Jeanne W. George and Marlene Hauck and Yumiko Kagawa and Matti Kiupel and Keith Linder and Kristina Meichner and Laura Marconato and Michelle L. Oblak and Renato L. Santos and R. Mark Simpson and Harold Tvedten and Derick Whitley},
url = {http://journals.sagepub.com/doi/10.1177/03009858211013712},
doi = {10.1177/03009858211013712},
issn = {0300-9858, 1544-2217},
year = {2021},
date = {2021-09-01},
urldate = {2023-06-30},
journal = {Veterinary Pathology},
volume = {58},
number = {5},
pages = {766–794},
abstract = {Standardization of tumor assessment lays the foundation for validation of grading systems, permits reproducibility of oncologic studies among investigators, and increases confidence in the significance of study results. Currently, there is minimal methodological standardization for assessing tumors in veterinary medicine, with few attempts to validate published protocols and grading schemes. The current article attempts to address these shortcomings by providing standard guidelines for tumor assessment parameters and protocols for evaluating specific tumor types. More detailed information is available in the Supplemental Files, the intention of which is 2-fold: publication as part of this commentary, but more importantly, these will be available as “living documents” on a website ( www.vetcancerprotocols.org ), which will be updated as new information is presented in the peer-reviewed literature. Our hope is that veterinary pathologists will agree that this initiative is needed, and will contribute to and utilize this information for routine diagnostic work and oncologic studies. Journal editors and reviewers can utilize checklists to ensure publications include sufficient detail and standardized methods of tumor assessment. To maintain the relevance of the guidelines and protocols, it is critical that the information is periodically updated and revised as new studies are published and validated with the intent of providing a repository of this information. Our hope is that this initiative (a continuation of efforts published in this journal in 2011) will facilitate collaboration and reproducibility between pathologists and institutions, increase case numbers, and strengthen clinical research findings, thus ensuring continued progress in veterinary oncologic pathology and improving patient care.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Standardization of tumor assessment lays the foundation for validation of grading systems, permits reproducibility of oncologic studies among investigators, and increases confidence in the significance of study results. Currently, there is minimal methodological standardization for assessing tumors in veterinary medicine, with few attempts to validate published protocols and grading schemes. The current article attempts to address these shortcomings by providing standard guidelines for tumor assessment parameters and protocols for evaluating specific tumor types. More detailed information is available in the Supplemental Files, the intention of which is 2-fold: publication as part of this commentary, but more importantly, these will be available as “living documents” on a website ( www.vetcancerprotocols.org ), which will be updated as new information is presented in the peer-reviewed literature. Our hope is that veterinary pathologists will agree that this initiative is needed, and will contribute to and utilize this information for routine diagnostic work and oncologic studies. Journal editors and reviewers can utilize checklists to ensure publications include sufficient detail and standardized methods of tumor assessment. To maintain the relevance of the guidelines and protocols, it is critical that the information is periodically updated and revised as new studies are published and validated with the intent of providing a repository of this information. Our hope is that this initiative (a continuation of efforts published in this journal in 2011) will facilitate collaboration and reproducibility between pathologists and institutions, increase case numbers, and strengthen clinical research findings, thus ensuring continued progress in veterinary oncologic pathology and improving patient care.
31.
Sievert, Matti; Oetter, Nicolai; Aubreville, Marc; Stelzle, Florian; Maier, Andreas; Eckstein, Markus; Mantsopoulos, Konstantinos; Gostian, Antoniu-Oreste; Mueller, Sarina K; Koch, Michael; Agaimy, Abbas; Iro, Heinrich; Goncalves, Miguel
Feasibility of intraoperative assessment of safe surgical margins during laryngectomy with confocal laser endomicroscopy: A pilot study Journal Article
In: Auris Nasus Larynx, vol. 48, no. 4, pp. 764–769, 2021, ISSN: 03858146.
@article{sievert_feasibility_2021,
title = {Feasibility of intraoperative assessment of safe surgical margins during laryngectomy with confocal laser endomicroscopy: A pilot study},
author = {Matti Sievert and Nicolai Oetter and Marc Aubreville and Florian Stelzle and Andreas Maier and Markus Eckstein and Konstantinos Mantsopoulos and Antoniu-Oreste Gostian and Sarina K Mueller and Michael Koch and Abbas Agaimy and Heinrich Iro and Miguel Goncalves},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0385814621000237},
doi = {10.1016/j.anl.2021.01.005},
issn = {03858146},
year = {2021},
date = {2021-08-01},
urldate = {2023-06-30},
journal = {Auris Nasus Larynx},
volume = {48},
number = {4},
pages = {764–769},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
30.
Donovan, Taryn A.; Moore, Frances M.; Bertram, Christof A.; Luong, Richard; Bolfa, Pompei; Klopfleisch, Robert; Tvedten, Harold; Salas, Elisa N.; Whitley, Derick B.; Aubreville, Marc; Meuten, Donald J.
Mitotic Figures—Normal, Atypical, and Imposters: A Guide to Identification Journal Article
In: Veterinary Pathology, vol. 58, no. 2, pp. 243–257, 2021, ISSN: 0300-9858, 1544-2217.
Abstract | Links | BibTeX | Tags:
@article{donovan_mitotic_2021,
title = {Mitotic Figures—Normal, Atypical, and Imposters: A Guide to Identification},
author = {Taryn A. Donovan and Frances M. Moore and Christof A. Bertram and Richard Luong and Pompei Bolfa and Robert Klopfleisch and Harold Tvedten and Elisa N. Salas and Derick B. Whitley and Marc Aubreville and Donald J. Meuten},
url = {http://journals.sagepub.com/doi/10.1177/0300985820980049},
doi = {10.1177/0300985820980049},
issn = {0300-9858, 1544-2217},
year = {2021},
date = {2021-03-01},
urldate = {2023-06-30},
journal = {Veterinary Pathology},
volume = {58},
number = {2},
pages = {243–257},
abstract = {Counting mitotic figures (MF) in hematoxylin and eosin–stained histologic sections is an integral part of the diagnostic pathologist’s tumor evaluation. The mitotic count (MC) is used alone or as part of a grading scheme for assessment of prognosis and clinical decisions. Determining MCs is subjective, somewhat laborious, and has interobserver variation. Proposals for standardizing this parameter in the veterinary field are limited to terminology (use of the term MC) and area (MC is counted in an area measuring 2.37 mm
2
). Digital imaging techniques are now commonplace and widely used among veterinary pathologists, and field of view area can be easily calculated with digital imaging software. In addition to standardizing the methods of counting MF, the morphologic characteristics of MF and distinguishing atypical mitotic figures (AMF) versus mitotic-like figures (MLF) need to be defined. This article provides morphologic criteria for MF identification and for distinguishing normal phases of MF from AMF and MLF. Pertinent features of digital microscopy and application of computational pathology (CPATH) methods are discussed. Correct identification of MF will improve MC consistency, reproducibility, and accuracy obtained from manual (glass slide or whole-slide imaging) and CPATH approaches.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Counting mitotic figures (MF) in hematoxylin and eosin–stained histologic sections is an integral part of the diagnostic pathologist’s tumor evaluation. The mitotic count (MC) is used alone or as part of a grading scheme for assessment of prognosis and clinical decisions. Determining MCs is subjective, somewhat laborious, and has interobserver variation. Proposals for standardizing this parameter in the veterinary field are limited to terminology (use of the term MC) and area (MC is counted in an area measuring 2.37 mm
2
). Digital imaging techniques are now commonplace and widely used among veterinary pathologists, and field of view area can be easily calculated with digital imaging software. In addition to standardizing the methods of counting MF, the morphologic characteristics of MF and distinguishing atypical mitotic figures (AMF) versus mitotic-like figures (MLF) need to be defined. This article provides morphologic criteria for MF identification and for distinguishing normal phases of MF from AMF and MLF. Pertinent features of digital microscopy and application of computational pathology (CPATH) methods are discussed. Correct identification of MF will improve MC consistency, reproducibility, and accuracy obtained from manual (glass slide or whole-slide imaging) and CPATH approaches.
2
). Digital imaging techniques are now commonplace and widely used among veterinary pathologists, and field of view area can be easily calculated with digital imaging software. In addition to standardizing the methods of counting MF, the morphologic characteristics of MF and distinguishing atypical mitotic figures (AMF) versus mitotic-like figures (MLF) need to be defined. This article provides morphologic criteria for MF identification and for distinguishing normal phases of MF from AMF and MLF. Pertinent features of digital microscopy and application of computational pathology (CPATH) methods are discussed. Correct identification of MF will improve MC consistency, reproducibility, and accuracy obtained from manual (glass slide or whole-slide imaging) and CPATH approaches.
29.
Marzahl, Christian; Aubreville, Marc; Bertram, Christof A.; Maier, Jennifer; Bergler, Christian; Kröger, Christine; Voigt, Jörn; Breininger, Katharina; Klopfleisch, Robert; Maier, Andreas
EXACT: a collaboration toolset for algorithm-aided annotation of images with annotation version control Journal Article
In: Scientific Reports, vol. 11, no. 1, pp. 4343, 2021, ISSN: 2045-2322.
Abstract | Links | BibTeX | Tags:
@article{marzahl_exact_2021,
title = {EXACT: a collaboration toolset for algorithm-aided annotation of images with annotation version control},
author = {Christian Marzahl and Marc Aubreville and Christof A. Bertram and Jennifer Maier and Christian Bergler and Christine Kröger and Jörn Voigt and Katharina Breininger and Robert Klopfleisch and Andreas Maier},
url = {https://www.nature.com/articles/s41598-021-83827-4},
doi = {10.1038/s41598-021-83827-4},
issn = {2045-2322},
year = {2021},
date = {2021-02-01},
urldate = {2023-06-30},
journal = {Scientific Reports},
volume = {11},
number = {1},
pages = {4343},
abstract = {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.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
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.
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.
28.
Ganz, Jonathan; Kirsch, Tobias; Hoffmann, Lucas; Bertram, Christof A; Hoffmann, Christoph; Maier, Andreas; Breininger, Katharina; Blümcke, Ingmar; Jabari, Samir; Aubreville, Marc
Automatic and explainable grading of meningiomas from histopathology images Proceedings Article
In: pp. 69–80, PMLR, 2021, ISBN: 2640-3498.
@inproceedings{ganz_automatic_2021,
title = {Automatic and explainable grading of meningiomas from histopathology images},
author = {Jonathan Ganz and Tobias Kirsch and Lucas Hoffmann and Christof A Bertram and Christoph Hoffmann and Andreas Maier and Katharina Breininger and Ingmar Blümcke and Samir Jabari and Marc Aubreville},
url = {https://proceedings.mlr.press/v156/ganz21a/ganz21a.pdf},
isbn = {2640-3498},
year = {2021},
date = {2021-01-01},
pages = {69–80},
publisher = {PMLR},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
27.
Marzahl, Christian; Wilm, Frauke; Tharun, Lars; Perner, Sven; Bertram, Christof A; Kröger, Christine; Voigt, Jörn; Klopfleisch, Robert; Maier, Andreas; Aubreville, Marc
Robust quad-tree based registration on whole slide images Proceedings Article
In: pp. 181–190, PMLR, 2021, ISBN: 2640-3498.
BibTeX | Tags:
@inproceedings{marzahl_robust_2021,
title = {Robust quad-tree based registration on whole slide images},
author = {Christian Marzahl and Frauke Wilm and Lars Tharun and Sven Perner and Christof A Bertram and Christine Kröger and Jörn Voigt and Robert Klopfleisch and Andreas Maier and Marc Aubreville},
isbn = {2640-3498},
year = {2021},
date = {2021-01-01},
pages = {181–190},
publisher = {PMLR},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
26.
Wilm, Frauke; Bertram, Christof A.; Marzahl, Christian; Bartel, Alexander; Donovan, Taryn A.; Assenmacher, Charles-Antoine; Becker, Kathrin; Bennett, Mark; Corner, Sarah; Cossic, Brieuc; Denk, Daniela; Dettwiler, Martina; Gonzalez, Beatriz Garcia; Gurtner, Corinne; Heier, Annabelle; Lehmbecker, Annika; Merz, Sophie; Plog, Stephanie; Schmidt, Anja; Sebastian, Franziska; Smedley, Rebecca C.; Tecilla, Marco; Thaiwong, Tuddow; Breininger, Katharina; Kiupel, Matti; Maier, Andreas; Klopfleisch, Robert; Aubreville, Marc
Influence of Inter-Annotator Variability on Automatic Mitotic Figure Assessment Book Section
In: Palm, Christoph; Deserno, Thomas M.; Handels, Heinz; Maier, Andreas; Maier-Hein, Klaus; Tolxdorff, Thomas (Ed.): Bildverarbeitung für die Medizin 2021, pp. 241–246, Springer Fachmedien Wiesbaden, Wiesbaden, 2021, ISBN: 978-3-658-33197-9 978-3-658-33198-6, (Series Title: Informatik aktuell).
@incollection{palm_influence_2021,
title = {Influence of Inter-Annotator Variability on Automatic Mitotic Figure Assessment},
author = {Frauke Wilm and Christof A. Bertram and Christian Marzahl and Alexander Bartel and Taryn A. Donovan and Charles-Antoine Assenmacher and Kathrin Becker and Mark Bennett and Sarah Corner and Brieuc Cossic and Daniela Denk and Martina Dettwiler and Beatriz Garcia Gonzalez and Corinne Gurtner and Annabelle Heier and Annika Lehmbecker and Sophie Merz and Stephanie Plog and Anja Schmidt and Franziska Sebastian and Rebecca C. Smedley and Marco Tecilla and Tuddow Thaiwong and Katharina Breininger and Matti Kiupel and Andreas Maier and Robert Klopfleisch and Marc Aubreville},
editor = {Christoph Palm and Thomas M. Deserno and Heinz Handels and Andreas Maier and Klaus Maier-Hein and Thomas Tolxdorff},
url = {http://link.springer.com/10.1007/978-3-658-33198-6_56},
doi = {10.1007/978-3-658-33198-6_56},
isbn = {978-3-658-33197-9 978-3-658-33198-6},
year = {2021},
date = {2021-01-01},
urldate = {2023-06-30},
booktitle = {Bildverarbeitung für die Medizin 2021},
pages = {241–246},
publisher = {Springer Fachmedien Wiesbaden},
address = {Wiesbaden},
note = {Series Title: Informatik aktuell},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}