@inproceedings{RauberMendelScheppachetal., author = {Rauber, David and Mendel, Robert and Scheppach, Markus W. and Ebigbo, Alanna and Messmann, Helmut and Palm, Christoph}, title = {Analysis of Celiac Disease with Multimodal Deep Learning}, series = {Bildverarbeitung f{\"u}r die Medizin 2022: Proceedings, German Workshop on Medical Image Computing, Heidelberg, June 26-28, 2022}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2022: Proceedings, German Workshop on Medical Image Computing, Heidelberg, June 26-28, 2022}, publisher = {Springer Vieweg}, address = {Wiesbaden}, doi = {10.1007/978-3-658-36932-3_25}, pages = {115 -- 120}, abstract = {Celiac disease is an autoimmune disorder caused by gluten that results in an inflammatory response of the small intestine.We investigated whether celiac disease can be detected using endoscopic images through a deep learning approach. The results show that additional clinical parameters can improve the classification accuracy. In this work, we distinguished between healthy tissue and Marsh III, according to the Marsh score system. We first trained a baseline network to classify endoscopic images of the small bowel into these two classes and then augmented the approach with a multimodality component that took the antibody status into account.}, language = {en} } @misc{MendelSouzaJrRauberetal., author = {Mendel, Robert and Souza Jr., Luis Antonio de and Rauber, David and Papa, Jo{\~a}o Paulo and Palm, Christoph}, title = {Abstract: Semi-supervised Segmentation Based on Error-correcting Supervision}, series = {Bildverarbeitung f{\"u}r die Medizin 2021. Proceedings, German Workshop on Medical Image Computing, Regensburg, March 7-9, 2021}, journal = {Bildverarbeitung f{\"u}r die Medizin 2021. Proceedings, German Workshop on Medical Image Computing, Regensburg, March 7-9, 2021}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-33197-9}, doi = {10.1007/978-3-658-33198-6_43}, pages = {178}, abstract = {Pixel-level classification is an essential part of computer vision. For learning from labeled data, many powerful deep learning models have been developed recently. In this work, we augment such supervised segmentation models by allowing them to learn from unlabeled data. Our semi-supervised approach, termed Error-Correcting Supervision, leverages a collaborative strategy. Apart from the supervised training on the labeled data, the segmentation network is judged by an additional network.}, subject = {Deep Learning}, language = {en} } @article{EbigboPalmMessmann, author = {Ebigbo, Alanna and Palm, Christoph and Messmann, Helmut}, title = {Barrett esophagus: What to expect from Artificial Intelligence?}, series = {Best Practice \& Research Clinical Gastroenterology}, volume = {52-53}, journal = {Best Practice \& Research Clinical Gastroenterology}, number = {June-August}, publisher = {Elsevier}, issn = {1521-6918}, doi = {10.1016/j.bpg.2021.101726}, abstract = {The evaluation and assessment of Barrett's esophagus is challenging for both expert and nonexpert endoscopists. However, the early diagnosis of cancer in Barrett's esophagus is crucial for its prognosis, and could save costs. Pre-clinical and clinical studies on the application of Artificial Intelligence (AI) in Barrett's esophagus have shown promising results. In this review, we focus on the current challenges and future perspectives of implementing AI systems in the management of patients with Barrett's esophagus.}, subject = {Deep Learning}, language = {en} } @unpublished{MendelRueckertWilhelmetal., author = {Mendel, Robert and R{\"u}ckert, Tobias and Wilhelm, Dirk and R{\"u}ckert, Daniel and Palm, Christoph}, title = {Motion-Corrected Moving Average: Including Post-Hoc Temporal Information for Improved Video Segmentation}, doi = {10.48550/arXiv.2403.03120}, pages = {9}, abstract = {Real-time computational speed and a high degree of precision are requirements for computer-assisted interventions. Applying a segmentation network to a medical video processing task can introduce significant inter-frame prediction noise. Existing approaches can reduce inconsistencies by including temporal information but often impose requirements on the architecture or dataset. This paper proposes a method to include temporal information in any segmentation model and, thus, a technique to improve video segmentation performance without alterations during training or additional labeling. With Motion-Corrected Moving Average, we refine the exponential moving average between the current and previous predictions. Using optical flow to estimate the movement between consecutive frames, we can shift the prior term in the moving-average calculation to align with the geometry of the current frame. The optical flow calculation does not require the output of the model and can therefore be performed in parallel, leading to no significant runtime penalty for our approach. We evaluate our approach on two publicly available segmentation datasets and two proprietary endoscopic datasets and show improvements over a baseline approach.}, subject = {Deep Learning}, language = {en} } @misc{ScheppachMendelProbstetal., author = {Scheppach, Markus W. and Mendel, Robert and Probst, Andreas and Meinikheim, Michael and Palm, Christoph and Messmann, Helmut and Ebigbo, Alanna}, title = {Intraprozedurale Strukturerkennung bei Third-Space Endoskopie mithilfe eines Deep-Learning Algorithmus}, series = {Zeitschrift f{\"u}r Gastroenterologie}, volume = {60}, journal = {Zeitschrift f{\"u}r Gastroenterologie}, number = {04}, publisher = {Thieme}, address = {Stuttgart}, doi = {10.1055/s-0042-1745652}, pages = {e250-e251}, abstract = {Einleitung Third-Space Interventionen wie die endoskopische Submukosadissektion (ESD) und die perorale endoskopische Myotomie (POEM) sind technisch anspruchsvoll und mit einem erh{\"o}hten Risiko f{\"u}r intraprozedurale Komplikationen wie Blutung oder Perforation assoziiert. Moderne Computerprogramme zur Unterst{\"u}tzung bei diagnostischen Entscheidungen werden unter Einsatz von k{\"u}nstlicher Intelligenz (KI) in der Endoskopie bereits erfolgreich eingesetzt. Ziel der vorliegenden Arbeit war es, relevante anatomische Strukturen mithilfe eines Deep-Learning Algorithmus zu detektieren und segmentieren, um die Sicherheit und Anwendbarkeit von ESD und POEM zu erh{\"o}hen. Methoden Zw{\"o}lf Videoaufnahmen in voller L{\"a}nge von Third-Space Endoskopien wurden aus der Datenbank des Universit{\"a}tsklinikums Augsburg extrahiert. 1686 Einzelbilder wurden f{\"u}r die Kategorien Submukosa, Blutgef{\"a}ß, Dissektionsmesser und endoskopisches Instrument annotiert und segmentiert. Mit diesem Datensatz wurde ein DeepLabv3+neuronales Netzwerk auf der Basis eines ResNet mit 101 Schichten trainiert und intern anhand der Parameter Intersection over Union (IoU), Dice Score und Pixel Accuracy validiert. Die F{\"a}higkeit des Algorithmus zur Gef{\"a}ßdetektion wurde anhand von 24 Videoclips mit einer Spieldauer von 7 bis 46 Sekunden mit 33 vordefinierten Gef{\"a}ßen evaluiert. Anhand dieses Tests wurde auch die Gef{\"a}ßdetektionsrate eines Experten in der Third-Space Endoskopie ermittelt. Ergebnisse Der Algorithmus zeigte eine Gef{\"a}ßdetektionsrate von 93,94\% mit einer mittleren Rate an falsch positiven Signalen von 1,87 pro Minute. Die Gef{\"a}ßdetektionsrate des Experten lag bei 90,1\% ohne falsch positive Ergebnisse. In der internen Validierung an Einzelbildern wurde eine IoU von 63,47\%, ein mittlerer Dice Score von 76,18\% und eine Pixel Accuracy von 86,61\% ermittelt. Zusammenfassung Dies ist der erste KI-Algorithmus, der f{\"u}r den Einsatz in der therapeutischen Endoskopie entwickelt wurde. Pr{\"a}limin{\"a}re Ergebnisse deuten auf eine mit Experten vergleichbare Detektion von Gef{\"a}ßen w{\"a}hrend der Untersuchung hin. Weitere Untersuchungen sind n{\"o}tig, um die Leistung des Algorithmus im Vergleich zum Experten genauer zu eruieren sowie einen m{\"o}glichen klinischen Nutzen zu ermitteln.}, language = {de} } @article{SouzaJrPachecoPassosetal., author = {Souza Jr., Luis Antonio de and Pacheco, Andr{\´e} G.C. and Passos, Leandro A. and Santana, Marcos Cleison S. and Mendel, Robert and Ebigbo, Alanna and Probst, Andreas and Messmann, Helmut and Palm, Christoph and Papa, Jo{\~a}o Paulo}, title = {DeepCraftFuse: visual and deeply-learnable features work better together for esophageal cancer detection in patients with Barrett's esophagus}, series = {Neural Computing and Applications}, volume = {36}, journal = {Neural Computing and Applications}, publisher = {Springer}, address = {London}, doi = {10.1007/s00521-024-09615-z}, pages = {10445 -- 10459}, abstract = {Limitations in computer-assisted diagnosis include lack of labeled data and inability to model the relation between what experts see and what computers learn. Even though artificial intelligence and machine learning have demonstrated remarkable performances in medical image computing, their accountability and transparency level must be improved to transfer this success into clinical practice. The reliability of machine learning decisions must be explained and interpreted, especially for supporting the medical diagnosis. While deep learning techniques are broad so that unseen information might help learn patterns of interest, human insights to describe objects of interest help in decision-making. This paper proposes a novel approach, DeepCraftFuse, to address the challenge of combining information provided by deep networks with visual-based features to significantly enhance the correct identification of cancerous tissues in patients affected with Barrett's esophagus (BE). We demonstrate that DeepCraftFuse outperforms state-of-the-art techniques on private and public datasets, reaching results of around 95\% when distinguishing patients affected by BE that is either positive or negative to esophageal cancer.}, subject = {Deep Learning}, 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} }