Functions

• Full Professor at University of Applied Sciences and Arts Western Switzerland (HES-SO)
  Head of PARAM research group at iCoSys
• Lecturer at University of Fribourg
  Senior member of DIVA research group

Domains of Expertise

• Pattern Recognition, Machine Learning, Deep Learning
• Document Image Analysis, Natural Language Processing, Histopathological Image Analysis
• Graph Matching, Graph Edit Distance, Geometric Deep Learning

Teaching

• Deep Learning (Master, HES-SO, English)
• Pattern Recognition (Master, UniFR, English)
• Machine Learning (Bachelor, HEIA-FR, French)
• Algorithms and Data Structures (Bachelor, HEIA-FR, German)
• Software Skills Lab (Bachelor, UniFR, English)

Research Projects

Projects as principal investigator:

• Combining image and graph-based neural networks for handwriting recognition
  2024-2027, Swiss National Science Foundation
• The AI-powered travel insurance system
  2024-2027, Innosuisse
• Automated HVAC-concept audit and optimisation using AI
  2023-2026, Innosuisse
• Character segmentation of on-line handwriting
  2021-2022, Google Research
• Transcription and integration of old index cards
  2020-2022, Swiss National Library
• Automatic classification of job advertisements
  2020-2021, SECO
• Towards graph-based keyword spotting in historical Vietnamese steles
  2020-2021, Hasler Foundation
• Automatic translation from Swiss German to High German
  2018-2022, Swisscom
• Automatic handwriting recognition for tax form validator
  2017-2023, TAINA Technology
• The visible digital library
  2017-2019, Hasler Foundation
• Automatic handwriting recognition and writer identification based on the Kinematic Theory
  2014-2015, Swiss National Science Foundation
• Bootstrapping handwriting recognition systems for historical documents
  2012-2013, Swiss National Science Foundation

Selected projects as project partner:

• Smart learning in the digital era
  2022-2025, Innosuisse
• Trans-omic approach to colorectal cancer: an integrative computational and clinical perspective
  2020-2024, Swiss National Science Foundation
• Automatic handwriting recognition for 16th century letters
  2021-2023, Hasler Foundation
• 3D kinematics for remote patient monitoring
  2019-2020, H2020 (Attract)
• Mobilité douce des seniors
  2019-2020, HES-SO
• A combined budding/T-cell score in pT1 and stage II colorectal cancer
  2019-2022, Rising Tide Foundation
• Linking bio-based industry value chains across the Alpine region
  2018-2020, Interreg

Research Community

• Member of the governing board of the International Association for Pattern Recognition (IAPR)
• Member of the governing board of the International Graphonomics Society (IGS)
• Chair of the IAPR-TC11: Reading Systems
• Chair of the Swiss Association for Pattern Recognition
• Organizing Chair of the 4th International Workshop on Historical Document Imaging and Processing (HIP’2017) in Kyoto, Japan
• General Chair of the 16th International Conference on Document Analysis and Recognition (ICDAR’2021) in Lausanne, Switzerland
• Organizing Chair of the Workshop on Scaling-up Document Image Understanding (ScalDoc’2023) in San José, California, USA

Publications

2023

• A. L. Frei, A. Khan, P. Zens, A. Lugli, I. Zlobec, and A. Fischer, “GammaFocus: An image augmentation method to focus model attention for classification,” in Medical Imaging with Deep Learning, short paper track, 2023.
  [Bibtex]

  @inproceedings{frei2023gammafocus,
  title = {GammaFocus: An image augmentation method to focus model attention for classification},
  author = {Ana Leni Frei and Amjad Khan and Philipp Zens and Alessandro Lugli and Inti Zlobec and Andreas Fischer},
  booktitle = {Medical Imaging with Deep Learning, short paper track},
  year = {2023},
  url = {https://openreview.net/forum?id=MCAgRjgh6v},
  abstract = {In histopathology, histologic elements are not randomly located across an image but organize into structured patterns. In this regard, classification tasks or feature extraction from histology images may require context information to increase performance. In this work, we explore the importance of keeping context information for a cell classification task on Hematoxylin and Eosin (H&E) scanned whole slide images (WSI) in colorectal cancer. We show that to differentiate normal from malignant epithelial cells, the environment around the cell plays a critical role. We propose here an image augmentation based on gamma variations to guide deep learning models to focus on the object of interest while keeping context information. This augmentation method yielded more specific models and helped to increase the model performance (weighted F1 score with/without gamma augmentation respectively, PanNuke: 99.49 vs 99.37 and TCGA: 91.38 vs. 89.12, p<0.05). }
  }

• A. L. Frei, A. Khan, L. Studer, P. Zens, A. Lugli, A. Fischer, and I. Zlobec, "Local and global feature aggregation for accurate epithelial cell classification using graph attention mechanisms in histopathology images," in Medical Imaging with Deep Learning, short paper track, 2023.
  [Bibtex]

  @inproceedings{frei2023local,
  url = {https://openreview.net/forum?id=HlkroJOY-J},
  title = {Local and global feature aggregation for accurate epithelial cell classification using graph attention mechanisms in histopathology images},
  author = {Frei, Ana Leni and Khan, Amjad and Studer, Linda and Zens, Philipp and Lugli, Alessandro and Fischer, Andreas and Zlobec, Inti},
  booktitle = {Medical Imaging with Deep Learning, short paper track},
  year = {2023},
  abstract = {In digital pathology, cell-level tissue analyses are widely used to better understand tissue composition and structure. Publicly available datasets and models for cell detection and classification in colorectal cancer exist but lack the differentiation of normal and malignant epithelial cells that are important to perform prior to any downstream cell-based analysis. This classification task is particularly difficult due to the high intra-class variability of neoplastic cells. To tackle this, we present here a new method that uses graph-based node classification to take advantage of both local cell features and global tissue architecture to perform accurate epithelial cell classification. The proposed method demonstrated excellent performance on F1 score (PanNuke: 1.0, TCGA: 0.98) and performed significantly better than conventional computer vision methods (PanNuke: 0.99, TCGA: 0.92).}
  }

• M. Jungo, B. Wolf, A. Maksai, C. Musat, and A. Fischer, "Character Queries: A Transformer-Based Approach to On-line Handwritten Character Segmentation," in Document Analysis and Recognition - ICDAR 2023, Cham, 2023, p. 98–114.
  [Bibtex]

  @InProceedings{jungo23character,
  url = {https://doi.org/10.48550/arXiv.2309.03072},
  doi = {10.48550/arXiv.2309.03072},
  author = {Jungo, Michael and Wolf, Beat and Maksai, Andrii and Musat, Claudiu and Fischer, Andreas},
  editor = {Fink, Gernot A. and Jain, Rajiv and Kise, Koichi and Zanibbi, Richard},
  title = {Character Queries: A Transformer-Based Approach to On-line Handwritten Character Segmentation},
  booktitle = {Document Analysis and Recognition - ICDAR 2023},
  year = {2023},
  publisher = {Springer Nature Switzerland},
  address = {Cham},
  pages = {98--114},
  abstract = {On-line handwritten character segmentation is often associated with handwriting recognition and even though recognition models include mechanisms to locate relevant positions during the recognition process, it is typically insufficient to produce a precise segmentation. Decoupling the segmentation from the recognition unlocks the potential to further utilize the result of the recognition. We specifically focus on the scenario where the transcription is known beforehand, in which case the character segmentation becomes an assignment problem between sampling points of the stylus trajectory and characters in the text. Inspired by the k-means clustering algorithm, we view it from the perspective of cluster assignment and present a Transformer-based architecture where each cluster is formed based on a learned character query in the Transformer decoder block. In order to assess the quality of our approach, we create character segmentation ground truths for two popular on-line handwriting datasets, IAM-OnDB and HANDS-VNOnDB, and evaluate multiple methods on them, demonstrating that our approach achieves the overall best results.},
  isbn = {978-3-031-41676-7}
  }

• R. Plamondon, A. Bensalah, K. Lebel, R. Salameh, G. Séguin de Broin, C. O’Reilly, M. Begon, O. Desbiens, Y. Beloufa, A. Guy, D. and Berio, F. F. Leymarie, S. Boyoguéno-Bidias, A. Fischer, Z. Zhang, M. Morin, D. Alamargot, C. Rémi, N. Faci, R. Fortin, M. Simard, and C. Bazinet, "Lognormality: An Open Window on Neuromotor Control," in Graphonomics in Human Body Movement. Bridging Research and Practice from Motor Control to Handwriting Analysis and Recognition, Cham, 2023, p. 205–258.
  [Bibtex]

  @inproceedings{plamondon2023lognormality,
  doi = {10.1007/978-3-031-45461-5_15},
  url = {https://doi.org/10.1007/978-3-031-45461-5_15},
  title = {Lognormality: An Open Window on Neuromotor Control},
  author = {Plamondon, R{\'e}jean and Bensalah, Asma and Lebel, Karina and Salameh, Romeo and S{\'e}guin de Broin, Guillaume and O’Reilly, Christian and Begon, Mickael and Desbiens, Olivier and Beloufa, Youssef and Guy, Aymeric and and Berio, Daniel and Leymarie, Frederic Fol and Boyogu{\'e}no-Bidias, Simon-Pierre and Fischer, Andreas and Zhang, Zigeng and Morin, Marie-France and Alamargot, Denis and R{\'e}mi, C{\'e}line and Faci, Nadir and Fortin, Rapha{\"e}lle and Simard, Marie-No{\"e}lle and Bazinet, Caroline},
  editor = {Parziale, Antonio and Diaz, Moises and Melo, Filipe},
  booktitle = {Graphonomics in Human Body Movement. Bridging Research and Practice from Motor Control to Handwriting Analysis and Recognition},
  pages = {205--258},
  year = {2023},
  address = {Cham},
  publisher = {Springer Nature Switzerland},
  abstract = {This invited special session of IGS 2023 presents the works carried out at Laboratoire Scribens and some of its collaborating laboratories. It summarises the 17 talks presented in the colloquium {\#}611 entitled « La lognormalit{\'e}: une fen{\^e}tre ouverte sur le contr{\^o}le neuromoteur» (Lognormality: a window opened on neuromotor control), at the 2023 conference of the Association Francophone pour le Savoir (ACFAS) on May 10, 2023. These talks covered a wide range of subjects related to the Kinematic Theory, including key elements of the theory, some gesture analysis algorithms that have emerged from it, and its application to various fields, particularly in biomedical engineering and human-machine interaction.},
  isbn = {978-3-031-45461-5}
  }

• A. Scius-Bertrand, P. Ströbel, M. Volk, T. Hodel, and A. Fischer, "The Bullinger Dataset: A Writer Adaptation Challenge," in Document Analysis and Recognition - ICDAR 2023, Cham, 2023, p. 397–410.
  [Bibtex]

  @inproceedings{scius2023bullinger,
  doi = {10.1007/978-3-031-41676-7_23},
  url = {https://doi.org/10.1007/978-3-031-41676-7_23},
  title = {The Bullinger Dataset: A Writer Adaptation Challenge},
  author = {Scius-Bertrand, Anna and Str{\"o}bel, Phillip and Volk, Martin and Hodel, Tobias and Fischer, Andreas},
  editor = {Fink, Gernot A. and Jain, Rajiv and Kise, Koichi and Zanibbi, Richard},
  booktitle = {Document Analysis and Recognition - ICDAR 2023},
  pages = {397--410},
  year = {2023},
  address = {Cham},
  publisher = {Springer Nature Switzerland},
  abstract = {One of the main challenges of automatically transcribing large collections of handwritten letters is to cope with the high variability of writing styles present in the collection. In particular, the writing styles of non-frequent writers, who have contributed only few letters, are often missing in the annotated learning samples used for training handwriting recognition systems. In this paper, we introduce the Bullinger dataset for writer adaptation, which is based on the Heinrich Bullinger letter collection from the 16th century, using a subset of 3,622 annotated letters (about 1.2 million words) from 306 writers. We provide baseline results for handwriting recognition with modern recognizers, before and after the application of standard techniques for supervised adaptation of frequent writers and self-supervised adaptation of non-frequent writers.},
  isbn = {978-3-031-41676-7}
  }

• A. Scius-Bertrand, C. Rémi, E. Biabiany, J. Nagau, and A. Fischer, "Towards Visuo-Structural Handwriting Evaluation Based on Graph Matching," in Graphonomics in Human Body Movement. Bridging Research and Practice from Motor Control to Handwriting Analysis and Recognition, Cham, 2023, p. 75–88.
  [Bibtex]

  @inproceedings{scius2023towards,
  doi = {10.1007/978-3-031-45461-5_6},
  url = {https://doi.org/10.1007/978-3-031-45461-5_6},
  author= {Scius-Bertrand, Anna and R{\'e}mi, C{\'e}line and Biabiany, Emmanuel and Nagau, Jimmy and Fischer, Andreas},
  editor = {Parziale, Antonio and Diaz, Moises and Melo, Filipe},
  title = {Towards Visuo-Structural Handwriting Evaluation Based on Graph Matching},
  booktitle = {Graphonomics in Human Body Movement. Bridging Research and Practice from Motor Control to Handwriting Analysis and Recognition},
  year = {2023},
  publisher = {Springer Nature Switzerland},
  address = {Cham},
  pages = {75--88},
  abstract = {Judging the quality of handwriting based on visuo-structural criteria is fundamental for teachers when accompanying children who are learning to write. Automatic methods for quality assessment can support teachers when dealing with a large number of handwritings, in order to identify children who are having difficulties. In this paper, we investigate the potential of graph-based handwriting representation and graph matching to capture visuo-structural features and determine the legibility of cursive handwriting. On a comprehensive dataset of words written by children aged from 3 to 11 years, we compare the judgment of human experts with a graph-based analysis, both with respect to classification and clustering. The results are promising and highlight the potential of graph-based methods for handwriting evaluation.},
  isbn = {978-3-031-45461-5}
  }

• A. Scius-Bertrand, M. Bui, and A. Fischer, "A Hybrid Deep Learning Approach to Keyword Spotting in Vietnamese Stele Images," Informatica, vol. 47, iss. 3, 2023.
  [Bibtex]

  @article{scius2023hybrid,
  doi = {10.31449/inf.v47i3.4785},
  url = {https://doi.org/10.31449/inf.v47i3.4785},
  year = {2023},
  author = {Scius-Bertrand, Anna and Bui, Marc and Fischer, Andreas},
  title = {A Hybrid Deep Learning Approach to Keyword Spotting in Vietnamese Stele Images},
  journal = {Informatica},
  volume = {47},
  number = {3},
  abstract = {In order to access the rich cultural heritage conveyed in Vietnamese steles, automatic reading of stone engravings would be a great support for historians, who are analyzing tens of thousands of stele images. Approaching the challenging problem with deep learning alone is difficult because the data-driven models require large representative datasets with expert human annotations, which are not available for the steles and costly to obtain. In this article, we present a hybrid approach to spot keywords in stele images that combines data-driven deep learning with knowledge-based structural modeling and matching of Chu Nom characters. The main advantage of the proposed method is that it is annotation-free, i.e. no human data annotation is required. In an experimental evaluation, we demonstrate that keywords can be successfully spotted with a mean average precision of more than 70% when a single engraving style is considered.}
  }

• P. Ströbel, T. Hodel, A. Fischer, A. Scius-Bertrand, B. Wolf, A. Janka, J. Widmer, P. Scheurer, and M. Volk, "Bullingers Briefwechsel zugänglich machen: Stand der Handschriftenerkennung," , 2023.
  [Bibtex]

  @article{strobel2023bullingers,
  doi = {10.5281/zenodo.7715357},
  url = {https://boris.unibe.ch/id/eprint/180287},
  title = {Bullingers Briefwechsel zug{\"a}nglich machen: Stand der Handschriftenerkennung},
  author = {Str{\"o}bel, Phillip and Hodel, Tobias and Fischer, Andreas and Scius-Bertrand, Anna and Wolf, Beat and Janka, Anna and Widmer, Jonas and Scheurer, Patricia and Volk, Martin},
  year = {2023},
  publisher = {University of Zurich},
  abstract = {Anhand des Briefwechsels Heinrich Bullingers (1504-1575), das rund 10'000 Briefe umfasst, demonstrieren wir den Stand der Forschung in automatisierter Handschriftenerkennung. Es finden sich mehr als hundert unterschiedliche Schreiberhände in den Briefen mit sehr unterschiedlicher Verteilung. Das Korpus ist zweisprachig (Latein/Deutsch) und teilweise findet der Sprachwechsel innerhalb von Abschnitten oder gar Sätzen statt. Auf Grund dieser Vielfalt eignet sich der Briefwechsel optimal als Testumgebung für entsprechende Algorithmen und ist aufschlussreiche für Forschungsprojekte und Erinnerungsinstitutionen mit ähnlichen Problemstellungen. Im Paper werden drei Verfahren gegeneinander gestellt und abgewogen. Im folgenden werde drei Ansätze an dem Korpus getestet, die Aufschlüsse zum Stand und möglichen Entwicklungen im Bereich der Handschriftenerkennung versprechen. Erstens wird mit Transkribus eine etablierte Plattform genutzt, die zwei Engines (HTR+ und PyLaia) anbietet. Zweitens wird mit Hilfe von Data Augmentation versucht die Erkennung mit der state-of-the-art Engine HTRFlor zu verbessern und drittens werden neue Transformer-basierte Modelle (TrOCR) eingesetzt.}
  }

• L. Studer, J. Bokhorst, I. Nagtegaal, I. Zlobec, H. Dawson, and A. Fischer, "Tumor Budding T-cell Graphs: Assessing the Need for Resection in pT1 Colorectal Cancer Patients," in Medical Imaging with Deep Learning, 2023.
  [Bibtex]

  @inproceedings{studer2023tumor,
  title = {Tumor Budding T-cell Graphs: Assessing the Need for Resection in pT1 Colorectal Cancer Patients},
  author = {Linda Studer and JM Bokhorst and I Nagtegaal and Inti Zlobec and Heather Dawson and Andreas Fischer},
  booktitle = {Medical Imaging with Deep Learning},
  year = {2023},
  url = {https://openreview.net/forum?id=ruaXPgZCk6i},
  abstract = {Colon resection is often the treatment of choice for colorectal cancer (CRC) patients. However, especially for minimally invasive cancer, such as pT1, simply removing the polyps may be enough to stop cancer progression. Different histopathological risk factors such as tumor grade and invasion depth currently found the basis for the need for colon resection in pT1 CRC patients. Here, we investigate two additional risk factors, tumor budding and lymphocyte infiltration at the invasive front, which are known to be clinically relevant. We capture the spatial layout of tumor buds and T-cells and use graph-based deep learning to investigate them as potential risk predictors. Our pT1 Hotspot Tumor Budding T-cell Graph (pT1-HBTG) dataset consists of 626 tumor budding hotspots from 575 patients. We propose and compare three different graph structures, as well as combinations of the node labels. The best-performing Graph Neural Network architecture is able to increase specificity by 20% compared to the currently recommended risk stratification based on histopathological risk factors, without losing any sensitivity. We believe that using a graph-based analysis can help to assist pathologists in making risk assessments for pT1 CRC patients, and thus decrease the number of patients undergoing potentially unnecessary surgery. Both the code and dataset are made publicly available.}
  }

• L. Vögtlin, A. Scius-Bertrand, P. Maergner, A. Fischer, and R. Ingold, "DIVA-DAF: A Deep Learning Framework for Historical Document Image Analysis," in Proceedings of the 7th International Workshop on Historical Document Imaging and Processing, New York, NY, USA, 2023, p. 61–66.
  [Bibtex]

  @inproceedings{vogtlin23diva,
  author = {V\"{o}gtlin, Lars and Scius-Bertrand, Anna and Maergner, Paul and Fischer, Andreas and Ingold, Rolf},
  title = {DIVA-DAF: A Deep Learning Framework for Historical Document Image Analysis},
  year = {2023},
  isbn = {9798400708411},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3604951.3605511},
  doi = {10.1145/3604951.3605511},
  abstract = {Deep learning methods have shown strong performance in solving tasks for historical document image analysis. However, despite current libraries and frameworks, programming an experiment or a set of experiments and executing them can be time-consuming. This is why we propose an open-source deep learning framework, DIVA-DAF, which is based on PyTorch Lightning and specifically designed for historical document analysis. Pre-implemented tasks such as segmentation and classification can be easily used or customized. It is also easy to create one’s own tasks with the benefit of powerful modules for loading data, even large data sets, and different forms of ground truth. The applications conducted have demonstrated time savings for the programming of a document analysis task, as well as for different scenarios such as pre-training or changing the architecture. Thanks to its data module, the framework also allows to reduce the time of model training significantly.},
  booktitle = {Proceedings of the 7th International Workshop on Historical Document Imaging and Processing},
  pages = {61–66},
  numpages = {6},
  keywords = {historical documents, document image analysis, deep neural networks, deep learning framework},
  location = {, San Jose, CA, USA, },
  series = {HIP '23}
  }

2022

• C. Abbet, L. Studer, A. Fischer, H. Dawson, I. Zlobec, B. Bozorgtabar, and J. -P. Thiran, "Self-rule to multi-adapt: Generalized multi-source feature learning using unsupervised domain adaptation for colorectal cancer tissue detection," Medical Image Analysis, vol. 79, p. 1–20, 2022.
  [Bibtex]

  @article{abbet22selfruletomultiadapt,
  Author = {C. Abbet and L. Studer and A. Fischer and H. Dawson and I. Zlobec and B. Bozorgtabar and J.-P. Thiran},
  Date-Added = {2022-09-27 14:00:13 +0200},
  Date-Modified = {2022-09-27 14:02:26 +0200},
  Journal = {Medical Image Analysis},
  Pages = {1--20},
  Title = {Self-rule to multi-adapt: Generalized multi-source feature learning using unsupervised domain adaptation for colorectal cancer tissue detection},
  Volume = {79},
  Year = {2022}}

• J. Diesbach, A. Fischer, M. Bui, and A. Scius-Bertrand, "Generating synthetic styled Chu Nom characters," in Proc. 18th Int. Conf on Frontiers in Handwriting Recognition (ICFHR), 2022.
  [Bibtex]

  @inproceedings{diesbach22generating,
  Author = {Jonas Diesbach and Andreas Fischer and Marc Bui and Anna Scius-Bertrand},
  Booktitle = {Proc. 18th Int. Conf on Frontiers in Handwriting Recognition (ICFHR)},
  Date-Added = {2022-09-27 14:14:18 +0200},
  Date-Modified = {2022-09-27 14:16:12 +0200},
  Title = {Generating synthetic styled Chu Nom characters},
  Year = {2022}}

• A. Fornés, A. Bensalah, C. Carmona-Duarte, J. Chen, M. A. Ferrer, A. Fischer, J. Lladós, C. Martín, E. Opisso, R. Plamondon, A. Scius-Bertrand, and J. M. Tormos, "The RPM3D Project: 3D Kinematics for Remote Patient Monitoring," in Intertwining Graphonomics with Human Movements, Cham, 2022, p. 217–226.
  [Bibtex]

  @inproceedings{fornes22rpm3d,
  url = {https://doi.org/10.1007/978-3-031-19745-1_16},
  doi = {10.1007/978-3-031-19745-1_16},
  author = {Forn{\'e}s, Alicia and Bensalah, Asma and Carmona-Duarte, Cristina and Chen, Jialuo and Ferrer, Miguel A. and Fischer, Andreas and Llad{\'o}s, Josep and Mart{\'i}n, Cristina and Opisso, Eloy and Plamondon, R{\'e}jean and Scius-Bertrand, Anna and Tormos, Josep Maria},
  editor = {Carmona-Duarte, Cristina and Diaz, Moises and Ferrer, Miguel A. and Morales, Aythami},
  title = {The RPM3D Project: 3D Kinematics for Remote Patient Monitoring},
  booktitle = {Intertwining Graphonomics with Human Movements},
  year = {2022},
  publisher = {Springer International Publishing},
  address = {Cham},
  pages = {217--226},
  abstract = {This project explores the feasibility of remote patient monitoring based on the analysis of 3D movements captured with smartwatches. We base our analysis on the Kinematic Theory of Rapid Human Movement. We have validated our research in a real case scenario for stroke rehabilitation at the Guttmann Institute (https://www.guttmann.com/en/) (neurorehabilitation hospital), showing promising results. Our work could have a great impact in remote healthcare applications, improving the medical efficiency and reducing the healthcare costs. Future steps include more clinical validation, developing multi-modal analysis architectures (analysing data from sensors, images, audio, etc.), and exploring the application of our technology to monitor other neurodegenerative diseases.},
  isbn = {978-3-031-19745-1}
  }

• A. Scius-Bertrand, A. Fischer, and M. Bui, "Retrieving Keywords in Historical Vietnamese Stele Images Without Human Annotations," in Proc. 11th Int. Symposium on Information and Communication Technology (SoICT), 2022.
  [Bibtex]

  @inproceedings{scius22retrieving,
  Author = {A. Scius-Bertrand and A. Fischer and M. Bui},
  Booktitle = {Proc. 11th Int. Symposium on Information and Communication Technology (SoICT)},
  Date-Added = {2022-11-21 11:33:30 +0100},
  Date-Modified = {2022-11-21 11:34:59 +0100},
  Title = {Retrieving Keywords in Historical Vietnamese Stele Images Without Human Annotations},
  Year = {2022}}

• A. Scius-Bertrand, L. Studer, A. Fischer, and M. Bui, "Annotation-free keyword spotting in historical Vietnamese manuscripts using graph matching," in Proc. Int. Workshop on Structural and Syntactic Pattern Recognition (SSPR), 2022.
  [Bibtex]

  @inproceedings{scius22annotationfree,
  Author = {Anna Scius-Bertrand and Linda Studer and Andreas Fischer and Marc Bui},
  Booktitle = {Proc. Int. Workshop on Structural and Syntactic Pattern Recognition (SSPR)},
  Date-Added = {2022-09-27 14:11:54 +0200},
  Date-Modified = {2022-09-27 14:13:40 +0200},
  Title = {Annotation-free keyword spotting in historical Vietnamese manuscripts using graph matching},
  Year = {2022}}

• M. Spoto, B. Wolf, A. Fischer, and A. Scius-Bertrand, "Improving Handwriting Recognition for Historical Documents Using Synthetic Text Lines," in Intertwining Graphonomics with Human Movements, Cham, 2022, p. 61–75.
  [Bibtex]

  @InProceedings{spoto22improving,
  url = {https://doi.org/10.1007/978-3-031-19745-1_5},
  doi = {10.1007/978-3-031-19745-1_5},
  author = {Spoto, Martin
  and Wolf, Beat
  and Fischer, Andreas
  and Scius-Bertrand, Anna},
  editor = {Carmona-Duarte, Cristina
  and Diaz, Moises
  and Ferrer, Miguel A.
  and Morales, Aythami},
  title = {Improving Handwriting Recognition for Historical Documents Using Synthetic Text Lines},
  booktitle = {Intertwining Graphonomics with Human Movements},
  year = {2022},
  publisher = {Springer International Publishing},
  address = {Cham},
  pages = {61--75},
  abstract = {Automatic handwriting recognition for historical documents is a key element for making our cultural heritage available to researchers and the general public. However, current approaches based on machine learning require a considerable amount of annotated learning samples to read ancient scripts and languages. Producing such ground truth is a laborious and time-consuming task that often requires human experts. In this paper, to cope with a limited amount of learning samples, we explore the impact of using synthetic text line images to support the training of handwriting recognition systems. For generating text lines, we consider lineGen, a recent GAN-based approach, and for handwriting recognition, we consider HTR-Flor, a state-of-the-art recognition system. Different meta-learning strategies are explored that schedule the addition of synthetic text line images to the existing real samples. In an experimental evaluation on the well-known Bentham dataset as well as the newly introduced Bullinger dataset, we demonstrate a significant improvement of the recognition performance when combining real and synthetic samples.},
  isbn = {978-3-031-19745-1}
  }

• C. Stammet, P. Dotti, U. Ultes-Nitsche, and A. Fischer, Analyzing Büchi Automata with Graph Neural Networks, 2022.
  [Bibtex]

  @misc{stammet2022analyzing,
  url = {https://doi.org/10.48550/arXiv.2206.09619},
  doi = {10.48550/arXiv.2206.09619},
  title={Analyzing B\"uchi Automata with Graph Neural Networks},
  author={Christophe Stammet and Prisca Dotti and Ulrich Ultes-Nitsche and Andreas Fischer},
  year={2022},
  eprint={2206.09619},
  archivePrefix={arXiv},
  primaryClass={cs.FL},
  abstract = {Büchi Automata on infinite words present many interesting problems and are used frequently in program verification and model checking. A lot of these problems on Büchi automata are computationally hard, raising the question if a learning-based data-driven analysis might be more efficient than using traditional algorithms. Since Büchi automata can be represented by graphs, graph neural networks are a natural choice for such a learning-based analysis. In this paper, we demonstrate how graph neural networks can be used to reliably predict basic properties of Büchi automata when trained on automatically generated random automata datasets.}
  }

• L. Studer, J. Bokhorst, F. Ciompi, A. Fischer, and H. Dawson, "Budding-T-cell score is a potential predictor for more aggressive treatment in pT1 colorectal cancers," in Proc. 18th European Congress on Digital Pathology (ECDP), 2022.
  [Bibtex]

  @inproceedings{studer22budding,
  Author = {Linda Studer and John-Melle Bokhorst and Francesco Ciompi and Andreas Fischer and Heather Dawson},
  Booktitle = {Proc. 18th European Congress on Digital Pathology (ECDP)},
  Date-Added = {2022-09-27 14:06:46 +0200},
  Date-Modified = {2022-09-27 14:08:07 +0200},
  Title = {Budding-T-cell score is a potential predictor for more aggressive treatment in pT1 colorectal cancers},
  Year = {2022}}

2021

• C. Abbet, L. Studer, A. Fischer, B. Bozorgtabar, J. -P. Thiran, F. Müller, H. Dawson, and I. Zlobec, "Reducing the annotation workload: using self-supervised methods to learn from publicly available colorectal cancer datasets," in Proc. 87th Annual Congress of the Swiss Society of Pathology, 2021, p. 634–635.
  [Bibtex]

  @inproceedings{abbet21reducing,
  Author = {C. Abbet and L. Studer and A. Fischer and B. Bozorgtabar and J.-P. Thiran and F. M{\"u}ller and H. Dawson and I. Zlobec},
  Booktitle = {Proc. 87th Annual Congress of the Swiss Society of Pathology},
  Date-Added = {2022-09-27 13:57:32 +0200},
  Date-Modified = {2022-09-27 13:59:09 +0200},
  Pages = {634--635},
  Title = {Reducing the annotation workload: using self-supervised methods to learn from publicly available colorectal cancer datasets},
  Year = {2021}}

• C. Abbet, L. Studer, A. Fischer, H. Dawson, I. Zlobec, B. Bozorgtabar, and J. -P. Thiran, "Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping," in Proc. 4th Int. Conf. on Medical Imaging with Deep Learning (MIDL), 2021, p. 1–16.
  [Bibtex]

  @inproceedings{abbet21selfrule,
  Author = {C. Abbet and L. Studer and A. Fischer and H. Dawson and I. Zlobec and B. Bozorgtabar and J.-P. Thiran},
  Booktitle = {Proc. 4th Int. Conf. on Medical Imaging with Deep Learning (MIDL)},
  Date-Added = {2022-09-27 13:55:40 +0200},
  Date-Modified = {2022-09-27 13:56:42 +0200},
  Pages = {1--16},
  Title = {Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping},
  Year = {2021}}

• P. Riba, A. Fischer, J. Llados, and A. Fornes, "Learning Graph Edit Distance by Graph Neural Networks," Pattern Recognition, vol. 120, p. 1–11, 2021.
  [Bibtex]

  @article{riba21learning,
  Author = {P. Riba and A. Fischer and J. Llados and A. Fornes},
  Date-Added = {2022-09-27 13:38:19 +0200},
  Date-Modified = {2022-09-27 13:40:11 +0200},
  Journal = {Pattern Recognition},
  Pages = {1--11},
  Title = {Learning Graph Edit Distance by Graph Neural Networks},
  Volume = {120},
  Year = {2021}}

• A. Scius-Bertrand, M. Jungo, B. Wolf, A. Fischer, and M. Bui, "Annotation-Free Character Detection in Historical Vietnamese Stele Images," in Proc. 16th Int. Conf. on Document Analysis and Recognition (ICDAR), 2021, p. 432–447.
  [Bibtex]

  @inproceedings{scius21annotationfree,
  Author = {A. Scius-Bertrand and M. Jungo and B. Wolf and A. Fischer and M. Bui},
  Booktitle = {Proc. 16th Int. Conf. on Document Analysis and Recognition (ICDAR)},
  Date-Added = {2022-09-27 13:51:05 +0200},
  Date-Modified = {2022-09-27 13:52:23 +0200},
  Pages = {432--447},
  Title = {Annotation-Free Character Detection in Historical Vietnamese Stele Images},
  Year = {2021}}

• A. Scius-Bertrand, M. Jungo, B. Wolf, A. Fischer, and M. Bui, "Transcription Alignment of Historical Vietnamese Manuscripts without Human-Annotated Learning Samples," Applied Sciences, vol. 11, p. 1–18, 2021.
  [Bibtex]

  @article{scius21transcription,
  Author = {A. Scius-Bertrand and M. Jungo and B. Wolf and A. Fischer and M. Bui},
  Date-Added = {2022-09-27 13:42:23 +0200},
  Date-Modified = {2022-09-27 13:43:25 +0200},
  Journal = {Applied Sciences},
  Pages = {1--18},
  Title = {Transcription Alignment of Historical Vietnamese Manuscripts without Human-Annotated Learning Samples},
  Volume = {11},
  Year = {2021}}

• L. Studer, J. Wallau, H. Dawson, I. Zlobec, and A. Fischer, "Classification of Intestinal Gland Cell-Graphs Using Graph Neural Networks," in Proc. 25th Int. Conf. on Pattern Recognition (ICPR), 2021, p. 3636–3643.
  [Bibtex]

  @inproceedings{studer21classification,
  Author = {L. Studer and J. Wallau and H. Dawson and I. Zlobec and A. Fischer},
  Booktitle = {Proc. 25th Int. Conf. on Pattern Recognition (ICPR)},
  Date-Added = {2022-09-27 13:54:23 +0200},
  Date-Modified = {2022-09-27 13:55:20 +0200},
  Pages = {3636--3643},
  Title = {Classification of Intestinal Gland Cell-Graphs Using Graph Neural Networks},
  Year = {2021}}

• L. Studer, A. Blank, J. -M. Bokhorst, I. Nagtegaal, I. Zlobec, A. Lugli, A. Fischer, and H. Dawson, "Taking tumour budding to the next frontier–-a post International Tumour Budding Consensus Conference (ITBCC) 2016 review," Histopathology, vol. 78, iss. 4, p. 476–484, 2021.
  [Bibtex]

  @article{studer21taking,
  Author = {L. Studer and A. Blank and J.-M. Bokhorst and I. Nagtegaal and I. Zlobec and A. Lugli and A. Fischer and H. Dawson},
  Date-Added = {2022-09-27 13:43:50 +0200},
  Date-Modified = {2022-09-27 13:50:51 +0200},
  Journal = {Histopathology},
  Number = {4},
  Pages = {476--484},
  Title = {Taking tumour budding to the next frontier---a post International Tumour Budding Consensus Conference (ITBCC) 2016 review},
  Volume = {78},
  Year = {2021}}

• F. Wolf, A. Fischer, and G. A. Fink, "Graph Convolutional Neural Networks for Learning Attribute Representations for Word Spotting," in Proc. 16th Int. Conf. on Document Analysis and Recognition (ICDAR), 2021, p. 50–64.
  [Bibtex]

  @inproceedings{wolf21graph,
  Author = {F. Wolf and A. Fischer and G.A. Fink},
  Booktitle = {Proc. 16th Int. Conf. on Document Analysis and Recognition (ICDAR)},
  Date-Added = {2022-09-27 13:53:12 +0200},
  Date-Modified = {2022-09-27 13:54:08 +0200},
  Pages = {50--64},
  Title = {Graph Convolutional Neural Networks for Learning Attribute Representations for Word Spotting},
  Year = {2021}}

2015

• A. Bou Hernandez, A. Fischer, and R. Plamondon, "Omega-Lognormal Analysis of Oscillatory Movements as a Function of Brain Stroke Risk Factors," in Proc. 17th Conf. of the International Graphonomics Society, 2015, p. 59–62.
  [Bibtex]

  @inproceedings{bou15omega,
  Author = {A. {Bou Hernandez} and A. Fischer and R. Plamondon},
  Booktitle = {Proc. 17th Conf. of the International Graphonomics Society},
  Date-Added = {2017-01-17 10:41:35 +0000},
  Date-Modified = {2017-01-17 10:41:35 +0000},
  Pages = {59--62},
  Title = {Omega-Lognormal Analysis of Oscillatory Movements as a Function of Brain Stroke Risk Factors},
  Year = {2015}}

• M. Diaz-Cabrera, A. Fischer, R. Plamondon, and M. A. Ferrer, "Towards an On-line Automatic Signature Verifier Using Only One Reference Per Signer," in Proc. 13th Int. Conf. on Document Analysis and Recognition, 2015, p. 631–635.
  [Bibtex]

  @inproceedings{diaz15towards,
  Author = {Diaz-Cabrera, M. and Fischer, A. and Plamondon, R. and Ferrer, M.A.},
  Booktitle = {Proc. 13th Int. Conf. on Document Analysis and Recognition},
  Date-Added = {2017-01-16 23:38:18 +0000},
  Date-Modified = {2017-01-16 23:38:18 +0000},
  Pages = {631--635},
  Title = {Towards an On-line Automatic Signature Verifier Using Only One Reference Per Signer},
  Year = {2015}}

• A. Fischer and R. Plamondon, "A Dissimilarity Measure for On-Line Signature Verification Based on the Sigma-Lognormal Model," in Proc. 17th Conf. of the International Graphonomics Society, 2015, p. 83–86.
  [Bibtex]

  @inproceedings{fischer15adissimilarity,
  Author = {A. Fischer and R. Plamondon},
  Booktitle = {Proc. 17th Conf. of the International Graphonomics Society},
  Date-Added = {2017-01-17 10:41:47 +0000},
  Date-Modified = {2017-01-17 10:41:47 +0000},
  Pages = {83--86},
  Title = {A Dissimilarity Measure for On-Line Signature Verification Based on the Sigma-Lognormal Model},
  Year = {2015}}

• A. Fischer, S. Uchida, V. Frinken, K. Riesen, and H. Bunke, "Improving Hausdorff edit distance using structural node context," in Proc. 10th Int. Workshop on Graph-based Representations in Pattern Recognition, 2015, p. 148–157.
  [Bibtex]

  @inproceedings{fischer15improving,
  Author = {A. Fischer and S. Uchida and V. Frinken and K. Riesen and H. Bunke},
  Booktitle = {Proc. 10th Int. Workshop on Graph-based Representations in Pattern Recognition},
  Date-Added = {2017-01-17 10:41:03 +0000},
  Date-Modified = {2017-01-17 10:41:03 +0000},
  Pages = {148--157},
  Title = {Improving {H}ausdorff edit distance using structural node context},
  Year = {2015}}

• A. Fischer, M. Diaz-Cabrera, R. Plamondon, and M. A. Ferrer, "Robust Score Normalization for DTW-Based On-Line Signature Verification," in Proc. 13th Int. Conf. on Document Analysis and Recognition, 2015, p. 241–245.
  [Bibtex]

  @inproceedings{fischer15robust,
  Author = {Fischer, A. and Diaz-Cabrera, M. and Plamondon, R. and Ferrer, M.A.},
  Booktitle = {Proc. 13th Int. Conf. on Document Analysis and Recognition},
  Date-Added = {2017-01-16 23:38:37 +0000},
  Date-Modified = {2017-01-16 23:38:37 +0000},
  Pages = {241--245},
  Title = {Robust Score Normalization for {DTW}-Based On-Line Signature Verification},
  Year = {2015}}

• K. Riesen, M. Ferrer, A. Fischer, and H. Bunke, "Approximation of graph edit distance in quadratic time," in Proc. 10th Int. Workshop on Graph-based Representations in Pattern Recognition, 2015, p. 3–12.
  [Bibtex]

  @inproceedings{riesen15approximation,
  Author = {K. Riesen and M. Ferrer and A. Fischer and H. Bunke},
  Booktitle = {Proc. 10th Int. Workshop on Graph-based Representations in Pattern Recognition},
  Date-Added = {2017-01-17 10:41:22 +0000},
  Date-Modified = {2017-01-17 10:41:22 +0000},
  Pages = {3--12},
  Title = {Approximation of graph edit distance in quadratic time},
  Year = {2015}}

• M. Seuret, A. Fischer, A. Garz, M. Liwicki, and R. Ingold, "Clustering Historical Documents Based on the Reconstruction Error of Autoencoders," in Proc. 3rd Int. Workshop on Historical Document Imaging and Processing, 2015, p. 85–91.
  [Bibtex]

  @inproceedings{seuret15clustering,
  Author = {Seuret, M. and Fischer, A. and Garz, A. and Liwicki, M. and Ingold, R.},
  Booktitle = {Proc. 3rd Int. Workshop on Historical Document Imaging and Processing},
  Date-Added = {2017-01-17 09:37:18 +0000},
  Date-Modified = {2017-01-17 10:39:35 +0000},
  Pages = {85--91},
  Title = {Clustering Historical Documents Based on the Reconstruction Error of Autoencoders},
  Year = {2015}}

• H. Wei, M. Seuret, K. Chen, A. Fischer, M. Liwicki, and R. Ingold, "Selecting Autoencoder Features for Layout Analysis of Historical Documents," in Proc. 3rd Int. Workshop on Historical Document Imaging and Processing, 2015, p. 55–62.
  [Bibtex]

  @inproceedings{wei15selecting,
  Author = {Wei, H. and Seuret, M. and Chen, K. and Fischer, A. and Liwicki, M. and Ingold, R.},
  Booktitle = {Proc. 3rd Int. Workshop on Historical Document Imaging and Processing},
  Date-Added = {2017-01-17 10:38:06 +0000},
  Date-Modified = {2017-01-17 10:39:57 +0000},
  Pages = {55--62},
  Title = {Selecting Autoencoder Features for Layout Analysis of Historical Documents},
  Year = {2015}}