@article{RueckertRauberMaerkletal., author = {Rueckert, Tobias and Rauber, David and Maerkl, Raphaela and Klausmann, Leonard and Yildiran, Suemeyye R. and Gutbrod, Max and Nunes, Danilo Weber and Moreno, Alvaro Fernandez and Luengo, Imanol and Stoyanov, Danail and Toussaint, Nicolas and Cho, Enki and Kim, Hyeon Bae and Choo, Oh Sung and Kim, Ka Young and Kim, Seong Tae and Arantes, Gon{\c{c}}alo and Song, Kehan and Zhu, Jianjun and Xiong, Junchen and Lin, Tingyi and Kikuchi, Shunsuke and Matsuzaki, Hiroki and Kouno, Atsushi and Manesco, Jo{\~a}o Renato Ribeiro and Papa, Jo{\~a}o Paulo and Choi, Tae-Min and Jeong, Tae Kyeong and Park, Juyoun and Alabi, Oluwatosin and Wei, Meng and Vercauteren, Tom and Wu, Runzhi and Xu, Mengya and Wang, An and Bai, Long and Ren, Hongliang and Yamlahi, Amine and Hennighausen, Jakob and Maier-Hein, Lena and Kondo, Satoshi and Kasai, Satoshi and Hirasawa, Kousuke and Yang, Shu and Wang, Yihui and Chen, Hao and Rodr{\´i}guez, Santiago and Aparicio, Nicol{\´a}s and Manrique, Leonardo and Palm, Christoph and Wilhelm, Dirk and Feussner, Hubertus and Rueckert, Daniel and Speidel, Stefanie and Nasirihaghighi, Sahar and Al Khalil, Yasmina and Li, Yiping and Arbel{\´a}ez, Pablo and Ayobi, Nicol{\´a}s and Hosie, Olivia and Lyons, Juan Camilo}, title = {Comparative validation of surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation in endoscopy: Results of the PhaKIR 2024 challenge}, series = {Medical Image Analysis}, volume = {109}, journal = {Medical Image Analysis}, publisher = {Elsevier}, issn = {1361-8415}, doi = {10.1016/j.media.2026.103945}, pages = {31}, abstract = {Reliable recognition and localization of surgical instruments in endoscopic video recordings are foundational for a wide range of applications in computer- and robot-assisted minimally invasive surgery (RAMIS), including surgical training, skill assessment, and autonomous assistance. However, robust performance under real-world conditions remains a significant challenge. Incorporating surgical context - such as the current procedural phase - has emerged as a promising strategy to improve robustness and interpretability. To address these challenges, we organized the Surgical Procedure Phase, Keypoint, and Instrument Recognition (PhaKIR) sub-challenge as part of the Endoscopic Vision (EndoVis) challenge at MICCAI 2024. We introduced a novel, multi-center dataset comprising thirteen full-length laparoscopic cholecystectomy videos collected from three distinct medical institutions, with unified annotations for three interrelated tasks: surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation. Unlike existing datasets, ours enables joint investigation of instrument localization and procedural context within the same data while supporting the integration of temporal information across entire procedures. We report results and findings in accordance with the BIAS guidelines for biomedical image analysis challenges. The PhaKIR sub-challenge advances the field by providing a unique benchmark for developing temporally aware, context-driven methods in RAMIS and offers a high-quality resource to support future research in surgical scene understanding.}, language = {en} } @misc{RueckertRauberKlausmannetal., author = {Rueckert, Tobias and Rauber, David and Klausmann, Leonard and Gutbrod, Max and Rueckert, Daniel and Feussner, Hubertus and Wilhelm, Dirk and Palm, Christoph}, title = {PhaKIR Dataset - Surgical Procedure Phase, Keypoint, and Instrument Recognition [Data set]}, doi = {10.5281/zenodo.15740620}, abstract = {Note: A script for extracting the individual frames from the video files while preserving the challenge-compliant directory structure and frame-to-mask naming conventions is available on GitHub and can be accessed here: https://github.com/remic-othr/PhaKIR_Dataset. The dataset is described in the following publications: Rueckert, Tobias et al.: Comparative validation of surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation in endoscopy: Results of the PhaKIR 2024 challenge. arXiv preprint, https://arxiv.org/abs/2507.16559. 2025. Rueckert, Tobias et al.: Video Dataset for Surgical Phase, Keypoint, and Instrument Recognition in Laparoscopic Surgery (PhaKIR). arXiv preprint, https://arxiv.org/abs/2511.06549. 2025. The proposed dataset was used as the training dataset in the PhaKIR challenge (https://phakir.re-mic.de/) as part of EndoVis-2024 at MICCAI 2024 and consists of eight real-world videos of human cholecystectomies ranging from 23 to 60 minutes in duration. The procedures were performed by experienced physicians, and the videos were recorded in three hospitals. In addition to existing datasets, our annotations provide pixel-wise instance segmentation masks of surgical instruments for a total of 19 categories, coordinates of relevant instrument keypoints (instrument tip(s), shaft-tip transition, shaft), both at an interval of one frame per second, and specifications regarding the intervention phases for a total of eight different phase categories for each individual frame in one dataset and thus comprehensively cover instrument localization and the context of the operation. Furthermore, the provision of the complete video sequences offers the opportunity to include the temporal information regarding the respective tasks and thus further optimize the resulting methods and outcomes.}, language = {en} } @misc{GutbrodRauberWeberNunesetal., author = {Gutbrod, Max and Rauber, David and Weber Nunes, Danilo and Palm, Christoph}, title = {A cleaned subset of the first five CATARACTS test videos [Data set]}, doi = {10.5281/zenodo.14924735}, abstract = {This dataset is a subset of the original CATARACTS test dataset and is used by the OpenMIBOOD framework to evaluate a specific out-of-distribution setting. When using this dataset, it is mandatory to cite the corresponding publication (OpenMIBOOD (10.1109/CVPR52734.2025.02410)) and follow the acknowledgement and citation requirements of the original dataset (CATARACTS). The original CATARACTS dataset (associated publication,Homepage) consists of 50 videos of cataract surgeries, split into 25 train and 25 test videos. This subset contains the frames of the first 5 test videos. Further, black frames at the beginning of each video were removed.}, language = {en} } @misc{GutbrodRauberWeberNunesetal., author = {Gutbrod, Max and Rauber, David and Weber Nunes, Danilo and Palm, Christoph}, title = {Cropped single instrument frames subset from Cholec80 [Data set]}, doi = {10.5281/zenodo.14921670}, abstract = {This dataset is a subset of the original Cholec80 dataset and is used by the OpenMIBOOD framework to evaluate a specific out-of-distribution setting. When using this dataset, it is mandatory to cite the corresponding publication (OpenMIBOOD) and to follow the acknowledgement and citation requirements of the original dataset (Cholec80). The original Cholec80 dataset (associated paper,Homepage) consists of 80 cholecystectomy surgery videos recorded at 25 fps, performed by 13 surgeons. It includes phase annotations (25 fps) and tool presence labels (1 fps), with phase definitions provided by a senior surgeon. A tool is considered present if at least half of its tip is visible. The dataset categorizes tools into seven types: Grasper, Bipolar, Hook, Scissors, Clipper, Irrigator, and Specimen bag. Multiple tools may be present in each frame. Additionally, 76 of the 80 videos exhibit a strong black vignette. For this dataset subset, frames were extracted based on tool presence labels, selecting only those containing Grasper, Bipolar, Hook, or Clipper while ensuring that only a single tool appears per frame. To enhance visual consistency, the black vignette was removed by extracting an inner rectangular region, where applicable.}, language = {en} } @misc{GutbrodRauberWeberNunesetal., author = {Gutbrod, Max and Rauber, David and Weber Nunes, Danilo and Palm, Christoph}, title = {OpenMIBOOD's classification models for the MIDOG, PhaKIR, and OASIS-3 benchmarks [Data set]}, doi = {10.5281/zenodo.14982267}, abstract = {These models are provided for evaluating post-hoc out-of-distribution methods on the three OpenMIBOOD benchmarks: MIDOG, PhaKIR, and OASIS-3. When using these models, make sure to give appropriate credit and cite the OpenMIBOOD publication.}, language = {en} } @article{deSouzaJuniorPachecoOliveiradosSantosetal., author = {de Souza J{\´u}nior, Luis Antonio and Pacheco, Andr{\´e} Georghton Cardoso and Oliveira dos Santos, Thiago and Fogos da Rocha, Wyctor and Bouzon, Pedro Henrique and Palm, Christoph and Papa, Jo{\~a}o Paulo}, title = {LiwTERM-r: a Revised Lightweight Transformer-based Model for Multimodal Skin Lesion Detection Robust to Incomplete Input}, series = {Journal of the Brazilian Computer Society}, volume = {32}, journal = {Journal of the Brazilian Computer Society}, number = {1}, publisher = {Brazilian Computer Society}, doi = {10.5753/jbcs.2026.5871}, pages = {11}, abstract = {As the most common type of cancer in the world, skin cancer accounts for approximately 30\% of all diagnosed tumor-based lesions. Early diagnosis can reduce mortality and prevent disfiguring in different skin regions. With the application of machine learning techniques in recent years, especially deep learning, promising results in this task could be achieved, presenting studies demonstrating that the combination of patients' clinical anamneses and images of the injured lesion is essential for improving the correct classification of skin lesions. Despite that, meaningful use of anamneses with multiple collected images of the same skin lesion is mandatory, requiring further investigation. Thus, this project aims to contribute to developing multimodal machine learning-based models to solve the skin lesion classification problem by employing a lightweight transformer model that is robust to missing clinical information input. As a main hypothesis, models can be fed by multiple images from different sources as input along with clinical anamneses from the patient's historical evaluations, leading to a more factual and trustworthy diagnosis. Our model deals with the not-trivial task of combining images and clinical information concerning the skin lesions in a lightweight transformer architecture that does not demand high computation resources or even all the information from the anamneses but still presents competitive classification results.}, language = {en} } @misc{KlausmannRueckertRauberetal., author = {Klausmann, Leonard and Rueckert, Tobias and Rauber, David and Maerkl, Raphaela and Yildiran, Suemeyye R. and Gutbrod, Max and Palm, Christoph}, title = {Abstract: DIY Challenge Blueprint}, series = {Bildverarbeitung f{\"u}r die Medizin 2025: Proceedings, German Conference on Medical Image Computing, L{\"u}beck March 15-17, 2026}, journal = {Bildverarbeitung f{\"u}r die Medizin 2025: Proceedings, German Conference on Medical Image Computing, L{\"u}beck March 15-17, 2026}, editor = {Handels, Heinz and Breininger, Katharina and Deserno, Thomas M. and Maier, Andreas and Maier-Hein, Klaus H. and Palm, Christoph and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, doi = {10.1007/978-3-658-51100-5_27}, pages = {131 -- 131}, abstract = {The high cost of challenge platforms prevents many people from organizing their own competitions. The do-it-yourself (DIY) challenge blueprint [1] allows you to host your own biomedical AI benchmark challenge. Our DIY approach circumvents the current constraints of commercial challenge platforms. A sovereign, extensible and cost-efficient deployment is provided via containerised, identity-managed and reproducible pipelines. Focus lies on GDPR-compliant hosting via infrastructure-as-code, automated evaluation, modular orchestration, and role-based identity and access management. The framework integrates Docker-based execution and standardised interfaces for task definitions, dataset curation and evaluation. All in all it is designed to be flexible and modular, as demonstrated in the MICCAI 2024 PhaKIR challenge [2, 3]. In this case study, different medical tasks on a multicentre laparoscopic dataset with framewise labels for phases and spatial annotations for instruments across fulllength videos were supported. This case study empirically validates the DIY challenge blueprint as a reproducible and customizable challenge-hosting infrastructure. The full code can be found at https://github.com/remic-othr/PhaKIR_DIY.}, subject = {Bildverarbeitung}, language = {en} } @inproceedings{GutbrodRauberPalm, author = {Gutbrod, Max and Rauber, David and Palm, Christoph}, title = {Improving Generalization in Mitotic Cell Detection via Domain Transformations}, series = {Bildverarbeitung f{\"u}r die Medizin 2025: Proceedings, German Conference on Medical Image Computing, L{\"u}beck March 15-17, 2026}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2025: Proceedings, German Conference on Medical Image Computing, L{\"u}beck March 15-17, 2026}, editor = {Handels, Heinz and Breininger, Katharina and Deserno, Thomas M. and Maier, Andreas and Maier-Hein, Klaus H. and Palm, Christoph and Tolxdorff, Thomas}, publisher = {Springer Vieweg}, address = {Wiesbaden}, doi = {10.1007/978-3-658-51100-5_71}, pages = {362 -- 367}, abstract = {We address domain generalization (DG) in mitotic-cell (MC) detection by combining a β-variational autoencoder (VAE) for domain transformations with feature-space alignment together with an object detector. The β-VAE synthesizes domain-transformed images, and the detector is trained to map originals and their transformed counterparts to equal representations. On the MIDOG++ dataset, this approach improves out-of-domain detection F1 scores by 7 and 3 percentage points compared to the color-variation augmentation and stain-normalization baselines. Results further suggest that morphology shifts hinder generalization more than stain shifts.}, subject = {K{\"u}nstliche Intelligenz}, language = {en} } @article{WallnerGutbrodRauberetal., author = {Wallner, M. and Gutbrod, Max and Rauber, David and Ebigbo, Alanna and Probst, Andreas and Palm, Christoph and Messmann, Helmut and Roser, David}, title = {KI-gest{\"u}tzte Detektion und Segmentierung von Magenkarzinomen in westlichen endoskopischen Bilddaten anhand eines fine-tuned Deep-Learning Ansatzes}, series = {Zeitschrift f{\"u}r Gastroenterologie}, volume = {64}, journal = {Zeitschrift f{\"u}r Gastroenterologie}, number = {03}, publisher = {Thieme}, doi = {10.1055/s-0046-1817751}, pages = {e64 -- e65}, abstract = {Diese vorl{\"a}ufige monozentrische Studie zeigt, dass ein aus einem Barrett-{\"O}sophagus-KI-System feinjustiertes Deep-Learning-Modell Magenkarzinome in westlichen multimodalen endoskopischen Bilddaten zuverl{\"a}ssig detektieren und pr{\"a}zise segmentieren kann. Die hohe Segmentierungsgenauigkeit und Detektionssensitivit{\"a}t {\"u}ber verschiedene Bildmodalit{\"a}ten hinweg unterstreichen die Machbarkeit eines pathologiegest{\"u}tzten KI-Ansatzes auch in einer westlichen Niedriginzidenzpopulation. Aufgrund der ausschließlichen Verwendung von Bildern mit sichtbaren Tumoren lassen sich keine Aussagen zur Spezifit{\"a}t treffen; eine {\"U}bertragbarkeit auf Screening- oder Mischkollektive ist daher limitiert. Weitere Studien mit a) gr{\"o}ßerem Datensatz inklusive Videodaten, b) externer Validierung an einer multizentrischen westlichen Kohorte, sowie c) Anwendung und Pr{\"u}fung an nicht-neoplastischen Vergleichsbildern oder anderen Pathologien sind erforderlich. Nach unserem Kenntnisstand z{\"a}hlt dieses System zu den ersten in einer westlichen Population entwickelten endoskopischen KI-Ans{\"a}tzen zur Magenkarzinomdetektion, und zu wenigen, die vollst{\"a}ndige ESD-pr{\"a}paratbasierte Referenzdaten f{\"u}r Training und Validierung nutzen.}, language = {de} } @article{WeiherervonRiedheimBrebantetal., author = {Weiherer, Maximilian and von Riedheim, Antonia and Br{\´e}bant, Vanessa and Egger, Bernhard and Palm, Christoph}, title = {Learning Neural Parametric 3D Breast Shape Models for Metrical Surface Reconstruction From Monocular RGB Videos}, series = {Machine Learning for Biomedical Imaging (MELBA)}, journal = {Machine Learning for Biomedical Imaging (MELBA)}, number = {MELBA-BVM 2025 Special Issue}, publisher = {Melba}, doi = {10.59275/j.melba.2026-8b23}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-89791}, pages = {95 -- 114}, abstract = {We present a neural parametric 3D breast shape model and, based on this model, introduce a low-cost and accessible 3D surface reconstruction pipeline capable of recovering accurate breast geometry from a monocular RGB video. In contrast to widely used, commercially available yet expensive 3D breast scanning solutions and existing low-cost alternatives, our method requires neither specialized hardware nor proprietary software and can be used with any device that is able to record RGB videos. The key building blocks of our pipeline are a state-of-the-art, off-the-shelf Structure-from-Motion pipeline, paired with a parametric breast model for robust surface reconstruction. Our model, similarly to the recently proposed implicit Regensburg Breast Shape Model (iRBSM), leverages implicit neural representations to model breast shapes. However, unlike the iRBSM, which employs a single global neural Signed Distance Function (SDF), our approach—inspired by recent state-of-the-art face models—decomposes the implicit breast domain into multiple smaller regions, each represented by a local neural SDF anchored at anatomical landmark positions. When incorporated into our surface reconstruction pipeline, the proposed model, dubbed liRBSM (short for localized iRBSM), significantly outperforms the iRBSM in terms of reconstruction quality, yielding more detailed surface reconstruction than its global counterpart. Overall, we find that the introduced pipeline is able to recover high-quality and metrically correct 3D breast geometry within an error margin of less than 2 mm. Our method is fast (requires less than six minutes), fully transparent and open-source, and together with the model publicly available at https://rbsm.re-mic.de/local-implicit.}, language = {en} }