TY - JOUR A1 - Souza Jr., Luis Antonio de A1 - Passos, Leandro A. A1 - Santana, Marcos Cleison S. A1 - Mendel, Robert A1 - Rauber, David A1 - Ebigbo, Alanna A1 - Probst, Andreas A1 - Messmann, Helmut A1 - Papa, João Paulo A1 - Palm, Christoph T1 - Layer-selective deep representation to improve esophageal cancer classification JF - Medical & Biological Engineering & Computing N2 - 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.For this task, the deep learning techniques’ black-box nature must somehow be lightened up to clarify its promising results. Hence, we aim to investigate the impact of the ResNet-50 deep convolutional design for Barrett’s esophagus and adenocarcinoma classification. For such a task, and aiming at proposing a two-step learning technique, the output of each convolutional layer that composes the ResNet-50 architecture was trained and classified for further definition of layers that would provide more impact in the architecture. We showed that local information and high-dimensional features are essential to improve the classification for our task. Besides, we observed a significant improvement when the most discriminative layers expressed more impact in the training and classification of ResNet-50 for Barrett’s esophagus and adenocarcinoma classification, demonstrating that both human knowledge and computational processing may influence the correct learning of such a problem. KW - Multistep training KW - Barrett’s esophagus detection KW - Convolutional neural networks KW - Deep learning Y1 - 2024 U6 - https://doi.org/10.1007/s11517-024-03142-8 VL - 62 SP - 3355 EP - 3372 PB - Springer Nature CY - Heidelberg ER - TY - JOUR A1 - Ebigbo, Alanna A1 - Mendel, Robert A1 - Probst, Andreas A1 - Meinikheim, Michael A1 - Byrne, Michael F. A1 - Messmann, Helmut A1 - Palm, Christoph T1 - Multimodal imaging for detection and segmentation of Barrett’s esophagus-related neoplasia using artificial intelligence JF - Endoscopy N2 - The early diagnosis of cancer in Barrett’s esophagus is crucial for improving the prognosis. However, identifying Barrett’s esophagus-related neoplasia (BERN) is challenging, even for experts [1]. Four-quadrant biopsies may improve the detection of neoplasia, but they can be associated with sampling errors. The application of artificial intelligence (AI) to the assessment of Barrett’s esophagus could improve the diagnosis of BERN, and this has been demonstrated in both preclinical and clinical studies [2] [3]. In this video demonstration, we show the accurate detection and delineation of BERN in two patients ([Video 1]). In part 1, the AI system detects a mucosal cancer about 20 mm in size and accurately delineates the lesion in both white-light and narrow-band imaging. In part 2, a small island of BERN with high-grade dysplasia is detected and delineated in white-light, narrow-band, and texture and color enhancement imaging. The video shows the results using a transparent overlay of the mucosal cancer in real time as well as a full segmentation preview. Additionally, the optical flow allows for the assessment of endoscope movement, something which is inversely related to the reliability of the AI prediction. We demonstrate that multimodal imaging can be applied to the AI-assisted detection and segmentation of even small focal lesions in real time. KW - Video KW - Artificial Intelligence KW - Multimodal Imaging Y1 - 2022 U6 - https://doi.org/10.1055/a-1704-7885 VL - 54 IS - 10 PB - Georg Thieme Verlag CY - Stuttgart ET - E-Video ER - TY - JOUR A1 - Kolev, Kalin A1 - Kirchgeßner, Norbert A1 - Houben, Sebastian A1 - Csiszár, Agnes A1 - Rubner, Wolfgang A1 - Palm, Christoph A1 - Eiben, Björn A1 - Merkel, Rudolf A1 - Cremers, Daniel T1 - A variational approach to vesicle membrane reconstruction from fluorescence imaging JF - Pattern Recognition N2 - Biological applications like vesicle membrane analysis involve the precise segmentation of 3D structures in noisy volumetric data, obtained by techniques like magnetic resonance imaging (MRI) or laser scanning microscopy (LSM). Dealing with such data is a challenging task and requires robust and accurate segmentation methods. In this article, we propose a novel energy model for 3D segmentation fusing various cues like regional intensity subdivision, edge alignment and orientation information. The uniqueness of the approach consists in the definition of a new anisotropic regularizer, which accounts for the unbalanced slicing of the measured volume data, and the generalization of an efficient numerical scheme for solving the arising minimization problem, based on linearization and fixed-point iteration. We show how the proposed energy model can be optimized globally by making use of recent continuous convex relaxation techniques. The accuracy and robustness of the presented approach are demonstrated by evaluating it on multiple real data sets and comparing it to alternative segmentation methods based on level sets. Although the proposed model is designed with focus on the particular application at hand, it is general enough to be applied to a variety of different segmentation tasks. KW - 3D segmentation KW - Convex optimization KW - Vesicle membrane analysis KW - Fluorescence imaging KW - Dreidimensionale Bildverarbeitung KW - Bildsegmentierung KW - Konvexe Optimierung Y1 - 2011 U6 - https://doi.org/10.1016/j.patcog.2011.04.019 VL - 44 IS - 12 SP - 2944 EP - 2958 PB - Elsevier ER - TY - CHAP A1 - Metzler, V. A1 - Aach, T. A1 - Palm, Christoph A1 - Lehmann, Thomas M. T1 - Texture Classification of Graylevel Images by Multiscale Cross-Co-Occurrence Matrices T2 - Proceedings 15th International Conference on Pattern Recognition (ICPR-2000) N2 - Local gray level dependencies of natural images can be modelled by means of co-occurrence matrices containing joint probabilities of gray-level pairs. Texture, however, is a resolution-dependent phenomenon and hence, classification depends on the chosen scale. Since there is no optimal scale for all textures we employ a multiscale approach that acquires textural features at several scales. Thus linear and nonlinear scale-spaces are analyzed by multiscale co-occurrence matrices that describe the statistical behavior of a texture in scale-space. Classification is then performed on the basis of texture features taken from the individual scale with the highest discriminatory power. By considering cross-scale occurrences of gray level pairs, the impact of filters on the feature is described and used for classification of natural textures. This novel method was found to improve classification rates of the common co-occurrence matrix approach on standard textures significantly. Y1 - 2000 U6 - https://doi.org/10.1109/ICPR.2000.906133 SP - 549 EP - 552 ER - TY - GEN A1 - Meinikheim, Michael A1 - Mendel, Robert A1 - Scheppach, Markus W. A1 - Probst, Andreas A1 - Prinz, Friederike A1 - Schwamberger, Tanja A1 - Schlottmann, Jakob A1 - Gölder, Stefan Karl A1 - Walter, Benjamin A1 - Steinbrück, Ingo A1 - Palm, Christoph A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - INFLUENCE OF AN ARTIFICIAL INTELLIGENCE (AI) BASED DECISION SUPPORT SYSTEM (DSS) ON THE DIAGNOSTIC PERFORMANCE OF NON-EXPERTS IN BARRETT´S ESOPHAGUS RELATED NEOPLASIA (BERN) T2 - Endoscopy N2 - Aims Barrett´s esophagus related neoplasia (BERN) is difficult to detect and characterize during endoscopy, even for expert endoscopists. We aimed to assess the add-on effect of an Artificial Intelligence (AI) algorithm (Barrett-Ampel) as a decision support system (DSS) for non-expert endoscopists in the evaluation of Barrett’s esophagus (BE) and BERN. Methods Twelve videos with multimodal imaging white light (WL), narrow-band imaging (NBI), texture and color enhanced imaging (TXI) of histologically confirmed BE and BERN were assessed by expert and non-expert endoscopists. For each video, endoscopists were asked to identify the area of BERN and decide on the biopsy spot. Videos were assessed by the AI algorithm and regions of BERN were highlighted in real-time by a transparent overlay. Finally, endoscopists were shown the AI videos and asked to either confirm or change their initial decision based on the AI support. Results Barrett-Ampel correctly identified all areas of BERN, irrespective of the imaging modality (WL, NBI, TXI), but misinterpreted two inflammatory lesions (Accuracy=75%). Expert endoscopists had a similar performance (Accuracy=70,8%), while non-experts had an accuracy of 58.3%. When AI was implemented as a DSS, non-expert endoscopists improved their diagnostic accuracy to 75%. Conclusions AI may have the potential to support non-expert endoscopists in the assessment of videos of BE and BERN. Limitations of this study include the low number of videos used. Randomized clinical trials in a real-life setting should be performed to confirm these results. KW - Artificial Intelligence KW - Barrett's Esophagus KW - Speiseröhrenkrankheit KW - Künstliche Intelligenz KW - Diagnose Y1 - 2022 U6 - https://doi.org/10.1055/s-00000012 VL - 54 IS - S 01 SP - S39 PB - Thieme ER - TY - JOUR A1 - Scheppach, Markus W. A1 - Mendel, Robert A1 - Probst, Andreas A1 - Meinikheim, Michael A1 - Palm, Christoph A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - ARTIFICIAL INTELLIGENCE (AI) – ASSISTED VESSEL AND TISSUE RECOGNITION IN THIRD-SPACE ENDOSCOPY JF - Endoscopy N2 - Aims Third-space endoscopy procedures such as endoscopic submucosal dissection (ESD) and peroral endoscopic myotomy (POEM) are complex interventions with elevated risk of operator-dependent adverse events, such as intra-procedural bleeding and perforation. We aimed to design an artificial intelligence clinical decision support solution (AI-CDSS, “Smart ESD”) for the detection and delineation of vessels, tissue structures, and instruments during third-space endoscopy procedures. Methods Twelve full-length third-space endoscopy videos were extracted from the Augsburg University Hospital database. 1686 frames were annotated for the following categories: Submucosal layer, blood vessels, electrosurgical knife and endoscopic instrument. A DeepLabv3+neural network with a 101-layer ResNet backbone was trained and validated internally. Finally, the ability of the AI system to detect visible vessels during ESD and POEM was determined on 24 separate video clips of 7 to 46 seconds duration and showing 33 predefined vessels. These video clips were also assessed by an expert in third-space endoscopy. Results Smart ESD showed a vessel detection rate (VDR) of 93.94%, while an average of 1.87 false positive signals were recorded per minute. VDR of the expert endoscopist was 90.1% with no false positive findings. On the internal validation data set using still images, the AI system demonstrated an Intersection over Union (IoU), mean Dice score and pixel accuracy of 63.47%, 76.18% and 86.61%, respectively. Conclusions This is the first AI-CDSS aiming to mitigate operator-dependent limitations during third-space endoscopy. Further clinical trials are underway to better understand the role of AI in such procedures. KW - Artificial Intelligence KW - Third-Space Endoscopy KW - Smart ESD Y1 - 2022 U6 - https://doi.org/10.1055/s-0042-1745037 VL - 54 IS - S01 SP - S175 PB - Thieme ER - TY - CHAP A1 - Weber Nunes, Danilo A1 - Rauber, David A1 - Palm, Christoph ED - Palm, Christoph ED - Breininger, Katharina ED - Deserno, Thomas M. ED - Handels, Heinz ED - Maier, Andreas ED - Maier-Hein, Klaus H. ED - Tolxdorff, Thomas T1 - Self-supervised 3D Vision Transformer Pre-training for Robust Brain Tumor Classification T2 - Bildverarbeitung für die Medizin 2025: Proceedings, German Conference on Medical Image Computing, Regensburg March 09-11, 2025 N2 - Brain tumors pose significant challenges in neurology, making precise classification crucial for prognosis and treatment planning. This work investigates the effectiveness of a self-supervised learning approach–masked autoencoding (MAE)–to pre-train a vision transformer (ViT) model for brain tumor classification. Our method uses non-domain specific data, leveraging the ADNI and OASIS-3 MRI datasets, which primarily focus on degenerative diseases, for pretraining. The model is subsequently fine-tuned and evaluated on the BraTS glioma and meningioma datasets, representing a novel use of these datasets for tumor classification. The pre-trained MAE ViT model achieves an average F1 score of 0.91 in a 5-fold cross-validation setting, outperforming the nnU-Net encoder trained from scratch, particularly under limited data conditions. These findings highlight the potential of self-supervised MAE in enhancing brain tumor classification accuracy, even with restricted labeled data. Y1 - 2025 U6 - https://doi.org/10.1007/978-3-658-47422-5_69 SP - 298 EP - 303 PB - Springer Vieweg CY - Wiesbaden ER - TY - CHAP A1 - Weiherer, Maximilian A1 - von Riedheim, Antonia A1 - Brébant, Vanessa A1 - Egger, Bernhard A1 - Palm, Christoph ED - Palm, Christoph ED - Breininger, Katharina ED - Deserno, Thomas M. ED - Handels, Heinz ED - Maier, Andreas ED - Maier-Hein, Klaus H. ED - Tolxdorff, Thomas T1 - iRBSM: A Deep Implicit 3D Breast Shape Model T2 - Bildverarbeitung für die Medizin 2025: Proceedings, German Conference on Medical Image Computing, Regensburg March 09-11, 2025 N2 - We present the first deep implicit 3D shape model of the female breast, building upon and improving the recently proposed Regensburg Breast Shape Model (RBSM). Compared to its PCA-based predecessor, our model employs implicit neural representations; hence, it can be trained on raw 3D breast scans and eliminates the need for computationally demanding non-rigid registration, a task that is particularly difficult for feature-less breast shapes. The resulting model, dubbed iRBSM, captures detailed surface geometry including fine structures such as nipples and belly buttons, is highly expressive, and outperforms the RBSM on different surface reconstruction tasks. Finally, leveraging the iRBSM, we present a prototype application to 3D reconstruct breast shapes from just a single image. Model and code publicly available at https://rbsm.re-mic.de/implicit. Y1 - 2025 U6 - https://doi.org/10.1007/978-3-658-47422-5_11 SP - 38 EP - 43 PB - Springer Vieweg CY - Wiesbaden ER - TY - CHAP A1 - Gutbrod, Max A1 - Geisler, Benedikt A1 - Rauber, David A1 - Palm, Christoph ED - Maier, Andreas ED - Deserno, Thomas M. ED - Handels, Heinz ED - Maier-Hein, Klaus H. ED - Palm, Christoph ED - Tolxdorff, Thomas T1 - Data Augmentation for Images of Chronic Foot Wounds T2 - Bildverarbeitung für die Medizin 2024: Proceedings, German Workshop on Medical Image Computing, March 10-12, 2024, Erlangen N2 - Training data for Neural Networks is often scarce in the medical domain, which often results in models that struggle to generalize and consequently showpoor performance on unseen datasets. Generally, adding augmentation methods to the training pipeline considerably enhances a model’s performance. Using the dataset of the Foot Ulcer Segmentation Challenge, we analyze two additional augmentation methods in the domain of chronic foot wounds - local warping of wound edges along with projection and blurring of shapes inside wounds. Our experiments show that improvements in the Dice similarity coefficient and Normalized Surface Distance metrics depend on a sensible selection of those augmentation methods. Y1 - 2024 U6 - https://doi.org/10.1007/978-3-658-44037-4_71 SP - 261 EP - 266 PB - Springer CY - Wiesbaden ER - TY - GEN A1 - Schroeder, Josef A. A1 - Semmelmann, Matthias A1 - Siegmund, Heiko A1 - Grafe, Claudia A1 - Evert, Matthias A1 - Palm, Christoph T1 - Improved interactive computer-assisted approach for evaluation of ultrastructural cilia abnormalities T2 - Ultrastructural Pathology KW - Zilie KW - Ultrastruktur KW - Anomalie KW - Bildverarbeitung KW - Computerunterstütztes Verfahren Y1 - 2017 U6 - https://doi.org/10.1080/01913123.2016.1270978 VL - 41 IS - 1 SP - 112 EP - 113 ER - TY - CHAP A1 - Rückert, Tobias A1 - Rieder, Maximilian A1 - Feussner, Hubertus A1 - Wilhelm, Dirk A1 - Rückert, Daniel A1 - Palm, Christoph ED - Maier, Andreas ED - Deserno, Thomas M. ED - Handels, Heinz ED - Maier-Hein, Klaus H. ED - Palm, Christoph ED - Tolxdorff, Thomas T1 - Smoke Classification in Laparoscopic Cholecystectomy Videos Incorporating Spatio-temporal Information T2 - Bildverarbeitung für die Medizin 2024: Proceedings, German Workshop on Medical Image Computing, March 10-12, 2024, Erlangen N2 - Heavy smoke development represents an important challenge for operating physicians during laparoscopic procedures and can potentially affect the success of an intervention due to reduced visibility and orientation. Reliable and accurate recognition of smoke is therefore a prerequisite for the use of downstream systems such as automated smoke evacuation systems. Current approaches distinguish between non-smoked and smoked frames but often ignore the temporal context inherent in endoscopic video data. In this work, we therefore present a method that utilizes the pixel-wise displacement from randomly sampled images to the preceding frames determined using the optical flow algorithm by providing the transformed magnitude of the displacement as an additional input to the network. Further, we incorporate the temporal context at evaluation time by applying an exponential moving average on the estimated class probabilities of the model output to obtain more stable and robust results over time. We evaluate our method on two convolutional-based and one state-of-the-art transformer architecture and show improvements in the classification results over a baseline approach, regardless of the network used. Y1 - 2024 U6 - https://doi.org/10.1007/978-3-658-44037-4_78 SP - 298 EP - 303 PB - Springeer CY - Wiesbaden ER - TY - INPR A1 - Mendel, Robert A1 - Rückert, Tobias A1 - Wilhelm, Dirk A1 - Rückert, Daniel A1 - Palm, Christoph T1 - Motion-Corrected Moving Average: Including Post-Hoc Temporal Information for Improved Video Segmentation N2 - 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. KW - Deep Learning KW - Video KW - Segmentation Y1 - 2024 U6 - https://doi.org/10.48550/arXiv.2403.03120 ER - TY - GEN A1 - Scheppach, Markus W. A1 - Mendel, Robert A1 - Probst, Andreas A1 - Meinikheim, Michael A1 - Palm, Christoph A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - Artificial Intelligence (AI) – assisted vessel and tissue recognition during third space endoscopy (Smart ESD) T2 - Zeitschrift für Gastroenterologie N2 - Clinical setting  Third space procedures such as endoscopic submucosal dissection (ESD) and peroral endoscopic myotomy (POEM) are complex minimally invasive techniques with an elevated risk for operator-dependent adverse events such as bleeding and perforation. This risk arises from accidental dissection into the muscle layer or through submucosal blood vessels as the submucosal cutting plane within the expanding resection site is not always apparent. Deep learning algorithms have shown considerable potential for the detection and characterization of gastrointestinal lesions. So-called AI – clinical decision support solutions (AI-CDSS) are commercially available for polyp detection during colonoscopy. Until now, these computer programs have concentrated on diagnostics whereas an AI-CDSS for interventional endoscopy has not yet been introduced. We aimed to develop an AI-CDSS („Smart ESD“) for real-time intra-procedural detection and delineation of blood vessels, tissue structures and endoscopic instruments during third-space endoscopic procedures. Characteristics of Smart ESD  An AI-CDSS was invented that delineates blood vessels, tissue structures and endoscopic instruments during third-space endoscopy in real-time. The output can be displayed by an overlay over the endoscopic image with different modes of visualization, such as a color-coded semitransparent area overlay, or border tracing (demonstration video). Hereby the optimal layer for dissection can be visualized, which is close above or directly at the muscle layer, depending on the applied technique (ESD or POEM). Furthermore, relevant blood vessels (thickness> 1mm) are delineated. Spatial proximity between the electrosurgical knife and a blood vessel triggers a warning signal. By this guidance system, inadvertent dissection through blood vessels could be averted. Technical specifications  A DeepLabv3+ neural network architecture with KSAC and a 101-layer ResNeSt backbone was used for the development of Smart ESD. It was trained and validated with 2565 annotated still images from 27 full length third-space endoscopic videos. The annotation classes were blood vessel, submucosal layer, muscle layer, electrosurgical knife and endoscopic instrument shaft. A test on a separate data set yielded an intersection over union (IoU) of 68%, a Dice Score of 80% and a pixel accuracy of 87%, demonstrating a high overlap between expert and AI segmentation. Further experiments on standardized video clips showed a mean vessel detection rate (VDR) of 85% with values of 92%, 70% and 95% for POEM, rectal ESD and esophageal ESD respectively. False positive measurements occurred 0.75 times per minute. 7 out of 9 vessels which caused intraprocedural bleeding were caught by the algorithm, as well as both vessels which required hemostasis via hemostatic forceps. Future perspectives  Smart ESD performed well for vessel and tissue detection and delineation on still images, as well as on video clips. During a live demonstration in the endoscopy suite, clinical applicability of the innovation was examined. The lag time for processing of the live endoscopic image was too short to be visually detectable for the interventionist. Even though the algorithm could not be applied during actual dissection by the interventionist, Smart ESD appeared readily deployable during visual assessment by ESD experts. Therefore, we plan to conduct a clinical trial in order to obtain CE-certification of the algorithm. This new technology may improve procedural safety and speed, as well as training of modern minimally invasive endoscopic resection techniques. KW - Artificial Intelligence KW - Medical Image Computing KW - Endoscopy KW - Bildgebendes Verfahren KW - Medizin KW - Künstliche Intelligenz KW - Endoskopie Y1 - 2022 U6 - https://doi.org/10.1055/s-0042-1755110 VL - 60 IS - 08 PB - Georg Thieme Verlag CY - Stuttgart ER - TY - JOUR A1 - Souza Jr., Luis Antonio de A1 - Pacheco, André G.C. A1 - Passos, Leandro A. A1 - Santana, Marcos Cleison S. A1 - Mendel, Robert A1 - Ebigbo, Alanna A1 - Probst, Andreas A1 - Messmann, Helmut A1 - Palm, Christoph A1 - Papa, João Paulo T1 - DeepCraftFuse: visual and deeply-learnable features work better together for esophageal cancer detection in patients with Barrett’s esophagus JF - Neural Computing and Applications N2 - 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. KW - Deep Learning KW - Speiseröhrenkrebs KW - Adenocarcinom KW - Endobrachyösophagus KW - Diagnose KW - Maschinelles Lernen KW - Machine learning KW - Adenocarcinoma KW - Object detector KW - Barrett’s esophagus KW - Deep Learning Y1 - 2024 U6 - https://doi.org/10.1007/s00521-024-09615-z VL - 36 SP - 10445 EP - 10459 PB - Springer CY - London ER - TY - JOUR A1 - Souza, Luis A. A1 - Pacheco, André G.C. A1 - de Souza, Alberto F. A1 - Oliveira-Santos, Thiago A1 - Badue, Claudine A1 - Palm, Christoph A1 - Papa, João Paulo T1 - TransConv: a lightweight architecture based on transformers and convolutional neural networks for adenocarcinoma and Barrett’s esophagus identification JF - Neural Computing and Applications N2 - Barrett’s esophagus, also known as BE, is commonly associated with repeated exposure to stomach acid. If not treated properly, it may evolve into esophageal adenocarcinoma, aka esophageal cancer. This paper proposes TransConv, a hybrid architecture that benefits from features learned by pre-trained vision transformers (ViTs) and convolutional neural networks (CNNs), followed by a shallow neural network composed of three normalizations, ReLU activations, and fully connected layers, and a SoftMax head to distinguish between BE and esophageal cancer. TransConv is designed to be training-lightweight, and for the ViT and CNN backbone models, weights are kept frozen during training, i.e., the primary goal of TransConv is to learn the weights of the fully connected layer from both backbones only, avoiding the burden of updating their weights but still learning their final descriptions for the lightweight convolutional model. We report promising results with low computational training costs in two datasets, one public and another private. From our achievements, TransConv was able to deliver balanced accuracy results around 85% and 86% for each evaluated dataset, respectively, in a design that required only 50 epochs of model training, a very reduced number compared to state-of-the-art conducted studies in the same domain. Y1 - 2025 U6 - https://doi.org/10.1007/s00521-025-11299-y IS - 37 SP - 15535 EP - 15546 PB - Springer ER - TY - INPR A1 - Rückert, Tobias A1 - Rückert, Daniel A1 - Palm, Christoph T1 - Methods and datasets for segmentation of minimally invasive surgical instruments in endoscopic images and videos: A review of the state of the art N2 - In the field of computer- and robot-assisted minimally invasive surgery, enormous progress has been made in recent years based on the recognition of surgical instruments in endoscopic images. Especially the determination of the position and type of the instruments is of great interest here. Current work involves both spatial and temporal information with the idea, that the prediction of movement of surgical tools over time may improve the quality of final segmentations. The provision of publicly available datasets has recently encouraged the development of new methods, mainly based on deep learning. In this review, we identify datasets used for method development and evaluation, as well as quantify their frequency of use in the literature. We further present an overview of the current state of research regarding the segmentation and tracking of minimally invasive surgical instruments in endoscopic images. The paper focuses on methods that work purely visually without attached markers of any kind on the instruments, taking into account both single-frame segmentation approaches as well as those involving temporal information. A discussion of the reviewed literature is provided, highlighting existing shortcomings and emphasizing available potential for future developments. The publications considered were identified through the platforms Google Scholar, Web of Science, and PubMed. The search terms used were "instrument segmentation", "instrument tracking", "surgical tool segmentation", and "surgical tool tracking" and result in 408 articles published between 2015 and 2022 from which 109 were included using systematic selection criteria. Y1 - 2023 U6 - https://doi.org/10.48550/arXiv.2304.13014 ER - TY - GEN A1 - Currle, Edda A1 - Haug, Sonja A1 - Weber, Karsten T1 - Artificial Intelligence and anamnesis: Results of a population survey N2 - Digital procedures are increasingly implemented to enhance efficiency in healthcare, with Artificial Intelligence (AI) — particularly chatbot s— showing significant potential for future applications. However, little is known about patients’ acceptance of such technologies. The study “AI and Anamnesis” addresses this gap by investigating the German population’s acceptance of and willingness to use AI-driven technologies for digital anamnesis. This poster presents results from the first wave of the survey, offering initial insights into public attitudes and potential barriers to adoption. KW - Künstliche Intelligenz KW - Anamnese KW - Technologieakzeptanz KW - Bevölkerungsbefragung Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-83325 ER - TY - JOUR A1 - Lehle, Karla A1 - Philipp, Alois A1 - Krenkel, Lars A1 - Gruber, Michael A1 - Hiller, Karl-Anton A1 - Müller, Thomas A1 - Lubnow, Matthias T1 - Thrombocytopenia During Venovenous Extracorporeal Membrane Oxygenation in Adult Patients With Bacterial, Viral, and COVID-19 Pneumonia JF - ASAIO Journal N2 - Contact of blood with artificial surfaces triggers platelet activation. The aim was to compare platelet kinetics after venovenous extracorporeal membrane oxygenation (V-V ECMO) start and after system exchange in different etiologies of acute lung failure. Platelet counts and coagulation parameters were analyzed from adult patients with long and exchange-free (≥8 days) ECMO runs (n = 330) caused by bacterial (n = 142), viral (n = 76), or coronavirus disease 2019 (COVID-19) (n = 112) pneumonia. A subpopulation requiring a system exchange and with long, exchange-free runs of the second oxygenator (≥7 days) (n = 110) was analyzed analogously. Patients with COVID-19 showed the highest platelet levels before ECMO implantation. Independent of the underlying disease and ECMO type, platelet counts decreased significantly within 24 hours and reached a steady state after 5 days. In the subpopulation, at the day of a system exchange, platelet counts were lower compared with ECMO start, but without differences between underlying diseases. Subsequently, platelets remained unchanged in the bacterial pneumonia group, but increased in the COVID-19 and viral pneumonia groups within 2–4 days, whereas D-dimers decreased and fibrinogen levels increased. Thus, overall platelet counts on V-V ECMO show disease-specific initial dynamics followed by an ongoing consumption by the ECMO device, which is not boosted by new artificial surfaces after a system exchange. Y1 - 2025 U6 - https://doi.org/10.1097/MAT.0000000000002383 SN - 1058-2916 SN - 1538-943X PB - Wolters Kluwer ER - TY - GEN A1 - Weber, Karsten T1 - The weapons of war and conflict are technology BT - Security from the viewpoint of Technology Assessment Y1 - 2023 ER - TY - GEN A1 - Weber, Karsten T1 - Social science research, research on acceptance, and applied ethics on technology and artificial intelligence in the health sector Y1 - 2024 ER - TY - JOUR A1 - Tröster, Mark A1 - Eckstein, Simon A1 - Kennel, Paula A1 - Kopp, Verna A1 - Benkiser, Alina A1 - Bihlmeier, Felicitas A1 - Daub, Urban A1 - Maufroy, Christophe A1 - Dendorfer, Sebastian A1 - Fritzsche, Lars A1 - Schneider, Urs A1 - Bauernhasl, Thomas T1 - Person-specific evaluation method for occupational exoskeletons - Biomechanical body heat map JF - Applied Ergonomics N2 - Human-centred and ergonomic work design is one of the most important drivers for increasing the competitiveness of the European Union. As a flexible, person-specific occupational measure, exoskeletons promise great potential for effectively reducing individual ergonomic stress. Digital human models can provide important insights and offer great potential for systematising the effect and targeted use of exoskeletons, supporting their effective implementation in practice. In this article, digital human models are applied on two levels. Firstly, a realistic industrial logistics scenario in which boxes had to be relocated is designed with the help of a digital human model for workplace and process planning and secondly, a new biomechanical evaluation methodology to analyse intended and unintended effects on internal stress on the human body is demonstrated by applying musculoskeletal exoskeleton human models of four test subjects. Finally, the modelled biomechanical support tendencies of one exoskeleton are preliminary validated using EMG measurement data of the back muscles collected from the four male workers. The preliminary analysis of two back-support exoskeletons to demonstrate the new methodological approach confirms the expected, intended effects in the lower back and reveals unintended effects, such as e.g. changes in knee kinetics when applying a soft or hard-frame exoskeleton. Furthermore, the exemplary results to demonstrate the methodological approach expose notable differences between the test subjects, which underlines the relevance of person-specific evaluation and consideration of exoskeleton support. The preliminary validation shows a correlation between the modelled and the EMG-measured biomechanical exoskeleton support of the considered back muscles. KW - Industry 5.0 KW - Ergonomics KW - Digital human modelling KW - Biomechanics KW - Musculoskeletal modelling KW - Occupational exoskeletons Y1 - 2025 U6 - https://doi.org/10.1016/j.apergo.2025.104671 VL - 132 PB - Elsevier ER - TY - JOUR A1 - Seifert, Ruth T1 - The “Human Rights Profession” and War: Humanitarian Help and Social Work in Armed Conflicts JF - Journal of Human Rights and Social Work N2 - Whereas, in the past, social work publications dealing with armed conflict were of interest to a relatively small community; beginning with the Ukraine conflict, this issue has moved into the center of social work discourses. This has raised issues concerning the positioning of social work in political and armed conflict again. Looking at the literature on social work and armed conflicts, it is generally assumed that the definition of the human rights profession also holds for social work in situations of violent collective conflicts. This, however, raises questions. This paper will argue that antinomies emerge for social work resulting from the definition as a human rights profession and the realities of humanitarian work in armed conflicts which have yet to be dealt with. Y1 - 2024 U6 - https://doi.org/10.1007/s41134-024-00349-5 VL - 9 IS - 3 SP - 360 EP - 371 PB - Springer Nature ER - TY - JOUR A1 - Schmiedt, Anja B. A1 - Empacher, Christina A1 - Kamps, Udo T1 - One- and two-sided prediction intervals for future Pareto record values with applications JF - Journal of Statistical Theory and Applications N2 - Based on upper record values in a series of observations over time from a Pareto distribution, an exact and several approximate one-sided and two-sided prediction intervals for the next record value (or for another future record value) to appear are studied and compared by means of an extensive simulation study. The performances of the prediction intervals are evaluated and recommendations are proposed for what method should be used in a given situation with respect to the form of the prediction interval on the one hand and to the tail behaviour on the other. The proposed methods are applied to insurance, environmental and sports data, where the respectively fitted Pareto distributions show a different tail behaviour. As a result, it is seen that although the number of observed record values in the presented real data applications is rather small, as it is usually the case, the selected prediction intervals are of practical use. KW - Interval prediction KW - Pareto distribution KW - Real data analyses KW - Record values KW - Simulation study Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-81605 SN - 2214-1766 N1 - Corresponding author der OTH Regensburg: Anja B. Schmiedt PB - Springer ER - TY - JOUR A1 - Dotter, Caroline A1 - Haug, Sonja A1 - Schnell, Rainer A1 - Raptis, Georgios A1 - Weber, Karsten T1 - Sharing health data for research purposes: results of a population survey in Germany JF - BMC health services research N2 - BACKGROUND: Increased use of health data has the potential to improve both health care and health policies. Several recent policy initiatives at the European and German legislative levels aim to increase the primary and secondary use of health data. However, little is known about general population views on health data access for research. Most studies are based on subsets defined by specific illnesses. METHODS: We commissioned a national computer-assisted dual-frame telephone survey (landline and mobile). Logit estimation models were used to identify predictors of willingness to provide access to health data to different organizations (universities in Germany, universities worldwide, German government organizations, pharmaceutical companies). RESULTS: A high willingness to share health data for research purposes is observed, depending on the specific data recipient. The willingness is highest for research at universities in Germany and German governmental organizations, and lowest regarding research by pharmaceutical companies. The main drivers for sharing health data are the level of trust in public institutions, the respondents' assessment of the seriousness and likelihood of data misuse, and the level of digital literacy. Age, gender, and level of education have small effects and do not determine the willingness to share health data for all organizations. CONCLUSION: We present evidence from a random sample of the German population. The results indicate widespread support among the population for providing access to health data for research purposes. Similar to findings in other countries, the willingness depends strongly on the recipient of the data. This paper evaluates the impact of various determinants - identified in previous qualitative and quantitative research - on the willingness of the German population to share health data. While previous studies have found that patients are generally more willing to share health data, we found that the presence of a medical precondition does not translate into respondents' unequivocal support for health data sharing. We identify privacy concerns, general trust, and digital literacy as key factors influencing the willingness to share health data. Therefore, policymakers and stakeholders need to ensure and communicate the necessary privacy protection measures to increase the willingness of the German population to share health data. Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-81622 N1 - Corresponding author der OTH Regensburg: Caroline Dotter VL - 25 PB - BMC ER - TY - GEN A1 - Rückert, Tobias A1 - Rückert, Daniel A1 - Palm, Christoph T1 - Corrigendum to “Methods and datasets for segmentation of minimally invasive surgical instruments in endoscopic images and videos: A review of the state of the art” [Comput. Biol. Med. 169 (2024) 107929] T2 - Computers in Biology and Medicine N2 - The authors regret that the SAR-RARP50 dataset is missing from the description of publicly available datasets presented in Chapter 4. Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-70337 N1 - Aufsatz unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/6983 PB - Elsevier ER - TY - JOUR A1 - Hammer, Simone A1 - Nunes, Danilo Weber A1 - Hammer, Michael A1 - Zeman, Florian A1 - Akers, Michael A1 - Götz, Andrea A1 - Balla, Annika A1 - Doppler, Michael Christian A1 - Fellner, Claudia A1 - Da Platz Batista Silva, Natascha A1 - Thurn, Sylvia A1 - Verloh, Niklas A1 - Stroszczynski, Christian A1 - Wohlgemuth, Walter Alexander A1 - Palm, Christoph A1 - Uller, Wibke T1 - Deep learning-based differentiation of peripheral high-flow and low-flow vascular malformations in T2-weighted short tau inversion recovery MRI JF - Clinical hemorheology and microcirculation N2 - BACKGROUND Differentiation of high-flow from low-flow vascular malformations (VMs) is crucial for therapeutic management of this orphan disease. OBJECTIVE A convolutional neural network (CNN) was evaluated for differentiation of peripheral vascular malformations (VMs) on T2-weighted short tau inversion recovery (STIR) MRI. METHODS 527 MRIs (386 low-flow and 141 high-flow VMs) were randomly divided into training, validation and test set for this single-center study. 1) Results of the CNN's diagnostic performance were compared with that of two expert and four junior radiologists. 2) The influence of CNN's prediction on the radiologists' performance and diagnostic certainty was evaluated. 3) Junior radiologists' performance after self-training was compared with that of the CNN. RESULTS Compared with the expert radiologists the CNN achieved similar accuracy (92% vs. 97%, p = 0.11), sensitivity (80% vs. 93%, p = 0.16) and specificity (97% vs. 100%, p = 0.50). In comparison to the junior radiologists, the CNN had a higher specificity and accuracy (97% vs. 80%, p <  0.001; 92% vs. 77%, p <  0.001). CNN assistance had no significant influence on their diagnostic performance and certainty. After self-training, the junior radiologists' specificity and accuracy improved and were comparable to that of the CNN. CONCLUSIONS Diagnostic performance of the CNN for differentiating high-flow from low-flow VM was comparable to that of expert radiologists. CNN did not significantly improve the simulated daily practice of junior radiologists, self-training was more effective. KW - magnetic resonance imaging KW - deep learning KW - Vascular malformation Y1 - 2024 U6 - https://doi.org/10.3233/CH-232071 SP - 1 EP - 15 PB - IOP Press ET - Pre-press ER - TY - GEN A1 - Braune, N. A1 - Greiner, M. A1 - Kappen, F. A1 - Kerscher, S. A1 - Mezler, C. A1 - Walter, L. A1 - Mertens, J. A1 - Pfingsten, Andrea T1 - Investigation of Cognitive-Motor Interference in Dual Tasking T2 - physioscience N2 - The transferability of the results to the overall population is limited due to the small sample size, unequal gender distribution, and a low average age. The researchers suspect a subconscious prioritization of the Cognitive Task during Dual Tasking through a division of limited attention resources of the central nervous system. In general, the findings of this study are closely aligned with the key points of the Central Capacity Sharing Model [1] and the Bottleneck Theory [2]. For future research, it is essential to consider a larger sample size, a more balanced gender distribution, and the inclusion of diverse age groups in order to achieve more reliable results. Y1 - 2025 U6 - https://doi.org/10.1055/s-0045-1808174 N1 - Abstract/Poster-Beitrag zu 8. Forschungssymposium Physiotherapie der Deutschen Gesellschaft für Physiotherapiewissenschaft e. V. Cottbus Senftenberg, 22.–23.11.2025 VL - 21 IS - S 01 SP - S36 PB - Thieme ER - TY - THES A1 - Süß, Franz T1 - The influence of mental stress on the musculoskeletal human back during static posture and trunk motion N2 - The investigation of the influence of mental stress on muscle recruitment of the back and its effect on the intervertebral discs was the main focus of this work. Furthermore, the goal was to develop algorithms to use mental stress as an input parameter in musculoskeletal simulation models. In the first step, a study was designed to investigate the influence of emotional and cognitive stress without kinetic influencing factors during sitting. At the muscular level, emotional stress was found to affect the upper back, while cognitive stress elicited higher muscle activity in the upper and lower back. Using a newly developed algorithm to apply back muscle recruitment changes to static inverse kinematic simulation models, load increases at the discs of up to 189 N on average and up to 907 N at peak were found.Based on the results of the first study, a second dynamic study was designed and conducted. In this case, the focus was on the cognitive stressor and the lower back. Using a dynamometer, subject-specific loads were applied during extension and flexion of the upper back. In contrast to the first study, in the upper back, only the right m. trapezius pars descendens showed a load-induced difference in muscle activity, but the lower back did. To investigate the effects of muscle tone increase in detail, the algorithm developed in the first study was extended to the dynamic case. The use of simulation models allowed the inference of the effects of the purely stress-induced tone increase. For this purpose, the kinetic and muscular effects were isolated and simulated. The study revealed a stress increase of 47% of the body weight in the L4L5 disc. The final numerical study focused on the general application of muscle activities to inverse kinematic simulation models. This was based on the novel simulation algorithm used in study two and the measured muscle activities. The simulation of the measured muscle activities formed the link between reality and simulation. To simulate the activities, neural networks and gradient boosting regression algorithms were investigated. The latter were found to be better suited to represent the data. However, the data is too small for a detailed statement, especially for loads below 100%. The results of this work can help to better assess the musculoskeletal effects of psychological stress on the musculoskeletal system and, if necessary, to develop ergonomic prevention strategies. By recognizing stress-related kinematic difference, as well as subsequent prompting of trunk movement, could help prevent long-term effects. When examining any situation, the combination of machine learning and musculoskeletal simulation tools can help examine and minimize the effects of psychological stress. N2 - Die Untersuchung des Einflusses von mentalem Stress auf die Muskelrekrutierung des Rückens und dessen Auswirkung auf die Bandscheiben war der Schwerpunkt dieser Arbeit. Des Weiteren war es das Ziel, Algorithmen zu entwickeln, um die mentale Belastung als Eingabeparameter in muskuloskelettalen Simulationsmodellen zu nutzen. Im ersten Schritt wurde eine Studie konzipiert, um den Einfluss von emotionalem und kognitivem Stress ohne kinetische Einflussfaktoren beim Sitzen zu untersuchen. Auf muskulärer Ebene wurde festgestellt, dass emotionaler Stress den oberen Rücken beeinflusste, während kognitiver Stress höhere Muskelaktivität im oberen und unteren Rücken auslöste. Unter Verwendung eines neu entwickelten Algorithmus zur Anwendung von Rekrutierungsänderungen der Rückenmuskulatur auf statische inverse kinematische Simulationsmodelle wurden Belastungserhöhungen an den Bandscheiben von bis zu 189 N im Mittel und bis zu 907 N in der Spitze gefunden. Basierend auf den Ergebnissen der ersten Studie wurde eine zweite dynamische Studie konzipiert und durchgeführt. In diesem Fall lag der Fokus auf dem kognitiven Stressor und dem unteren Rücken. Mit Hilfe eines Dynamometers wurden subjektspezifische Belastungen während der Extension und Flexion des Oberkörpers aufgebracht. Im Gegensatz zur ersten Studie konnte im oberen Rücken, nur im rechten m. trapezius pars descendens ein Lastfall bedingter Unterschied in der Muskelaktivität festgestellt werden, wohl aber im unteren Rücken. Um die Auswirkungen der Muskeltonuserhöhung im Detail zu untersuchen, wurde der in der ersten Studie entwickelte Algorithmus auf den dynamischen Fall erweitert. Die Verwendung von Simulationsmodellen erlaubte den Rückschluss auf die Auswirkungen der rein stressinduzierten Tonuserhöhung. Hierfür wurden die kinetischen und muskulären Effekte isoliert und simuliert. Die Studie ergab eine Belastungserhöhung von 47 % des Körpergewichts in der L4L5 Bandscheibe. Die abschließende numerische Studie konzentrierte sich auf die allgemeine Anwendung von Muskelaktivitäten auf inverse kinematische Simulationsmodelle. Grundlage dafür waren der neuartige Simulationsalgorithmus, der in Studie zwei verwendet wurde, und die gemessenen Muskelaktivitäten. Die Simulation der gemessenen Muskelaktivitäten bildete das Bindeglied zwischen Realität und Simulation. Um die Aktivitäten zu simulieren, wurden neuronale Netze und Gradient-Boosting-Regressionsalgorithmen untersucht. Es zeigte sich, dass letztere besser geeignet sind, die Daten abzubilden. Für eine detaillierte Aussage, insbesondere für Belastungen unter 100 %, ist die Datenlage jedoch zu klein. Die Ergebnisse dieser Arbeit können helfen, die muskuloskelettalen Auswirkungen psychischer Belastungen auf den Bewegungsapparat besser einzuschätzen und ggf. ergonomische Präventionsstrategien zu entwickeln. Durch das Erkennen stressbedingter kinematischer Unterschied sowie die darauffolgende Aufforderung zur Rumpfbewegung könnte helfen, langfristige Auswirkungen zu vermeiden. Bei der Untersuchung beliebiger Situationen, kann die Kombination aus maschinellem Lernen und muskuloskelettalen Simulationswerkzeugen helfen, die Auswirkungen psychischer Belastungen zu untersuchen und zu minimieren. KW - Musculoskeletal simulation KW - mental stress KW - spinal disc loads Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:355-epub-460554 ER - TY - JOUR A1 - Ott, Christian A1 - Rosengarth, Katharina A1 - Doenitz, Christian A1 - Hoehne, Julius A1 - Wendl, Christina A1 - Dodoo-Schittko, Frank A1 - Lang, Elmar Wolfgang A1 - Schmidt, Nils Ole A1 - Goldhacker, Markus T1 - Preoperative Assessment of Language Dominance through Combined Resting-State and Task-Based Functional Magnetic Resonance Imaging JF - Journal of personalized medicine N2 - Brain lesions in language-related cortical areas remain a challenge in the clinical routine. In recent years, the resting-state fMRI (RS-fMRI) was shown to be a feasible method for preoperative language assessment. The aim of this study was to examine whether language-related resting-state components, which have been obtained using a data-driven independent-component-based identification algorithm, can be supportive in determining language dominance in the left or right hemisphere. Twenty patients suffering from brain lesions close to supposed language-relevant cortical areas were included. RS-fMRI and task-based (TB-fMRI) were performed for the purpose of preoperative language assessment. TB-fMRI included a verb generation task with an appropriate control condition (a syllable switching task) to decompose language-critical and language-supportive processes. Subsequently, the best fitting ICA component for the resting-state language network (RSLN) referential to general linear models (GLMs) of the TB-fMRI (including models with and without linguistic control conditions) was identified using an algorithm based on the Dice index. Thereby, the RSLNs associated with GLMs using a linguistic control condition led to significantly higher laterality indices than GLM baseline contrasts. LIs derived from GLM contrasts with and without control conditions alone did not differ significantly. In general, the results suggest that determining language dominance in the human brain is feasible both with TB-fMRI and RS-fMRI, and in particular, the combination of both approaches yields a higher specificity in preoperative language assessment. Moreover, we can conclude that the choice of the language mapping paradigm is crucial for the mentioned benefits. KW - resting-state fMRI KW - task-based fMRI KW - brain mapping KW - language assessment KW - data-driven analysis Y1 - 2021 U6 - https://doi.org/10.3390/jpm11121342 VL - 11 IS - 12 PB - MDPI ER - TY - JOUR A1 - Plank, Tina A1 - Rosengarth, Katharina A1 - Schmalhofer, Carolin A1 - Goldhacker, Markus A1 - Brandl-Rühle, Sabine A1 - Greenlee, Mark W. T1 - Perceptual learning in patients with macular degeneration JF - Frontiers in psychology N2 - Patients with age-related macular degeneration (AMD) or hereditary macular dystrophies (JMD) rely on an efficient use of their peripheral visual field. We trained eight AMD and five JMD patients to perform a texture-discrimination task (TDT) at their preferred retinal locus (PRL) used for fixation. Six training sessions of approximately one hour duration were conducted over a period of approximately 3 weeks. Before, during and after training twelve patients and twelve age-matched controls (the data from two controls had to be discarded later) took part in three functional magnetic resonance imaging (fMRI) sessions to assess training-related changes in the BOLD response in early visual cortex. Patients benefited from the training measurements as indexed by significant decrease (p = 0.001) in the stimulus onset asynchrony (SOA) between the presentation of the texture target on background and the visual mask, and in a significant location specific effect of the PRL with respect to hit rate (p = 0.014). The following trends were observed: (i) improvement in Vernier acuity for an eccentric line-bisection task; (ii) positive correlation between the development of BOLD signals in early visual cortex and initial fixation stability (r = 0.531); (iii) positive correlation between the increase in task performance and initial fixation stability (r = 0.730). The first two trends were non-significant, whereas the third trend was significant at p = 0.014, Bonferroni corrected. Consequently, our exploratory study suggests that training on the TDT can enhance eccentric vision in patients with central vision loss. This enhancement is accompanied by a modest alteration in the BOLD response in early visual cortex. KW - perceptual learning KW - fMRI BOLD KW - cortical plasticity KW - visual cortex KW - macular degeneration Y1 - 2014 U6 - https://doi.org/10.3389/fpsyg.2014.01189 SN - 1664-1078 VL - 5 SP - 1 EP - 14 PB - Frontiers Research Foundation CY - Lausanne ER - TY - GEN A1 - Greenlee, Mark W. A1 - Anstis, Stuart A1 - Rosengarth, Katharina A1 - Goldhacker, Markus A1 - Brandl-Rühle, Sabine A1 - Plank, Tina T1 - Neural correlates of perceptual filling-in: fMRI evidence in the foveal projection zone of patients with central scotoma T2 - Journal of Vision / Vision Sciences Society Annual Meeting Abstract N2 - Patients with juvenile retinal dystrophy often report that they are unaware of their central scotoma, suggesting the presence of perceptual filling-in. We used functional Magnetic Resonance Imaging (fMRI) to determine possible neural correlates of perceptual filling-in in patients with retinal distrophy and clinically established central scotoma in both eyes. The data of 5 patients (Stargardt disease, cone-rod dystrophy; mean age 45 yrs; scotoma diameter 10-20°) and of 5 normally sighted controls were analyzed. Fixation behaviour and perimetry were measured with a Nidek microperimeter. Magnetic resonance imaging was performed using a Siemens 3T Allegra scanner. We stimulated the central visual field (30 deg) with a vertically oriented, low spatial frequency (1 c/deg) high-contrast sinewave grating that was either a) continuous, or b) was interrupted by a central grey disk. The disk was either slightly larger than the scotoma (detectable on 75% of trials) or slightly smaller (detectable on 25% of trials). To control for attention, an eccentric fixation task was performed during scanning. Data were analyzed using SPM8 (GLM with ROI analysis to obtain percent signal change for foveal projection zone). Results: for all patients, the BOLD signal in the foveal projection area was significantly higher for the small disk (i.e., condition leading to complete filling-in) than for the large disk (i.e., no filling-in). This effect was absent in the control subjects. Our findings support the existence of an active neural process that leads to filling-in in patients with central visual field scotomata. Y1 - 2012 U6 - https://doi.org/10.1167/12.9.1303 SN - 1534-7362 VL - 12 IS - 9 PB - ARVO ER - TY - JOUR A1 - Goldhacker, Markus A1 - Rosengarth, Katharina A1 - Plank, Tina A1 - Greenlee, Mark W. T1 - The effect of feedback on performance and brain activation during perceptual learning JF - Vision research N2 - We investigated the role of informative feedback on the neural correlates of perceptual learning in a coherent-motion detection paradigm. Stimulus displays consisted of four patches of moving dots briefly (500 ms) presented simultaneously, one patch in each visual quadrant. The coherence level was varied in the target patch from near threshold to high, while the other three patches contained only noise. The participants judged whether coherent motion was present or absent in the target patch. To guarantee central fixation, a secondary RSVP digit-detection task was performed at fixation. Over six training sessions subjects learned to detect coherent motion in a predefined quadrant (i.e., the learned location). Half of our subjects were randomly assigned to the feedback group, where they received informative feedback after each response during training, whereas the other group received non-informative feedback during training that a response button was pressed. We investigated whether the presence of informative feedback during training had an influence on the learning success and on the resulting BOLD response. Behavioral data of 24 subjects showed improved performance with increasing practice. Informative feedback promoted learning for motion displays with high coherence levels, whereas it had little effect on learning for displays with near-threshold coherence levels. Learning enhanced fMRI responses in early visual cortex and motion-sensitive area MT+ and these changes were most pronounced for high coherence levels. Activation in the insular and cingulate cortex was mainly influenced by coherence level and trained location. We conclude that feedback modulates behavioral performance and, to a lesser extent, brain activation in areas responsible for monitoring perceptual learning. KW - Perceptual learning KW - Functional MRI KW - Feedback KW - Coherent motion detection Y1 - 2014 U6 - https://doi.org/10.1016/j.visres.2013.11.010 SN - 1878-5646 SN - 0042-6989 VL - 99 SP - 99 EP - 110 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Greenlee, Mark W. A1 - Rosengarth, Katharina A1 - Schmalhofer, Carolin A1 - Goldhacker, Markus A1 - Brandl-Rühle, Sabine A1 - Plank, Tina T1 - Perceptual learning in patients with central scotomata due to hereditary and age-related macular dystrophy JF - Journal of Vision N2 - Hereditary and age-related forms of macular dystrophy (MD) are characterized by loss of cone function in the fovea, leading to central scotomata and eccentric fixation at the so-called preferred retinal locus (PRL). We investigated whether perceptual learning enhances visual abilities at the PRL. We also determined the neural correlates (3-Tesla fMRI) of learning success. Twelve MD patients (eight with age-related macular dystrophy, four with hereditary macular dystrophies) were trained on a texture discrimination task (TDT) over six days. Patients underwent three fMRI sessions (before, during and after training) while performing the TDT (target at PRL or opposite PRL). Reading speed, visual acuity (Vernier task) and contrast sensitivity were also assessed before and after training. With one exception, all patients showed improved performance (i.e. significant decrease in stimulus onset asynchronies and reaction times, significant increase in hit rates) on the TDT. Eight patients also showed moderate increases in reading speed, six patients showed improved thresholds in contrast sensitivity and nine patients showed improved thresholds in a vernier visual acuity task after TDT training. We found an increase in BOLD response in the projections zone of the PRL in the primary visual cortex in nine of twelve patients after training. The change in fMRI signal correlated (r = .8; p = .02) with the patients’ performance enhancements when the target was in the PRL. The results suggest that perceptual learning can enhance eccentric vision and cortical processing in MD patients. Y1 - 2014 U6 - https://doi.org/10.1167/14.10.666 SN - 1468-4233 SN - 0301-0066 VL - 14 IS - 10 PB - ARVO ER - TY - JOUR A1 - Goldhacker, Markus A1 - Rosengarth, Katharina A1 - Anstis, Stuart A1 - Wirth, Anna A1 - Plank, Tina A1 - Greenlee, Mark W. T1 - FMRI evidence for perceptual filling-in in patients with macular dystrophy JF - Perception Y1 - 2013 UR - https://www.researchgate.net/publication/298247764_FMRI_evidence_for_perceptual_filling-in_in_patients_with_macular_dystrophy SN - 1468-4233 SN - 0301-0066 VL - 42 SP - 72 EP - 73 ER - TY - THES A1 - Auer, Simon T1 - Musculoskeletal models in highly dynamic motion: effects of model parameters and mental stress N2 - The analysis and understanding of highly dynamic movements is a fundamental part of biomechanics. Since sports injuries often involve the lower extremities and muscles, musculoskeletal models can help to prevent them. These models allow the calculation of ground and joint reaction forces as well as muscle forces and activities for individual muscle strands. One goal of this work is to use musculoskeletal models to investigate the influence of mental stress on lower extremity loading. Moreover, the models themselves are evaluated for highly dynamic movements and practical recommendations for action will be derived. For this purpose, fast movements of youth competitive and amateur athletes will be recorded using different measurement systems. Subsequently, the models calculate the target parameters using inverse dynamics. Furthermore, measured and calculated muscle activities of the lower extremities are compared and artificial balancing forces (residuals) in the models are analyzed and minimization approaches are presented. The investigation of muscle and joint loading under mental stress has shown that the response to mental stress is highly individual. Athletes may experience a significant increase in muscle and knee forces with a simultaneous decrease in performance. The comparison of measured and calculated muscle activity proved the reliability of the models also for highly dynamic movements. With the frequently used default settings in the model and optical and inertial motion capture, the muscle activities in the model could be calculated reliably. The residual forces were highest, when the model transitioned from foot-ground contact to no contact and vice versa. By adjusting the settings of the kinematic filter and the ground reaction force prediction, the residuals were reduced by up to 54%. The analysis of musculoskeletal loading under mental stress has shown that the models can make a valuable contribution to the biomechanical analysis of highly dynamic movements. Subsequently, the models have also proven to be a reliable tool for the analysis of highly dynamic movements when the calculated parameters as well as the model-specific optimization options are reviewed. With this in mind, these models can contribute to further understand highly dynamic movements and prevent muscle injuries in athletes. N2 - Die Analyse und das Verständnis hochdynamischer Bewegungen ist ein fundamentaler Bestandteil der Biomechanik. Da Sportverletzungen häufig die unteren Extremitäten und Muskeln betreffen, können muskuloskelettale Modelle dazu beitragen, sie zu vermeiden. Diese Modelle ermöglichen die Berechnung von Boden- und Gelenkreaktionskräften sowie von Muskelkräften und -aktivitäten für einzelne Muskelstränge. Ein Ziel dieser Arbeit ist es, mit Hilfe von muskuloskelettalen Modellen den Einfluss von mentalen Belastungen auf die Belastung der unteren Extremitäten zu untersuchen. Darüber hinaus werden die Modelle selbst für hochdynamische Bewegungen evaluiert und praktische Handlungsempfehlungen abgeleitet. Zu diesem Zweck werden schnelle Bewegungen von jugendlichen Leistungs- und Freizeitsportlern mit verschiedenen Messsystemen aufgezeichnet. Anschließend berechnen die Modelle mittels inverser Dynamik die Zielparameter. Weiterhin werden gemessene und berechnete Muskelaktivitäten der unteren Extremitäten verglichen sowie künstliche Ausgleichskräfte (Residuen) in den Modellen analysiert und Minimierungsansätze vorgestellt. Die Untersuchung der Muskel- und Gelenkbelastung unter mentalem Stress hat gezeigt, dass die Reaktion auf mentalen Stress sehr individuell ist. Bei einzelnen Personen kann es zu einem deutlichen Anstieg der Muskel- und Kniegelenkreaktionskräfte bei gleichzeitiger Leistungsminderung kommen. Der Vergleich von gemessener und berechneter Muskelaktivität beweist die Zuverlässigkeit der Modelle auch bei hochdynamischen Bewegungen. Mit den häufig verwendeten Standardeinstellungen im Modell und der vielseitigen Bewegungserfassung konnten die Muskelaktivitäten im Modell zuverlässig berechnet werden. Die Residualkräfte waren am höchsten, wenn das Modell von Fuß-Boden-Kontakt zu keinem Kontakt und umgekehrt überging. Durch Anpassung der Einstellungen des kinematischen Filters und der Berechnung der Bodenreaktionskraft konnten die Residualkräfte um bis zu 54% reduziert werden. Die Analyse der muskuloskelettalen Belastung unter psychischer Beanspruchung hat gezeigt, dass die Modelle einen wertvollen Beitrag zur biomechanischen Analyse von hochdynamischen Bewegungen leisten können. In der Folge haben sich die Modelle auch bei der Überprüfung der berechneten Parameter sowie der modellspezifischen Optimierungsmöglichkeiten als zuverlässiges Werkzeug für die Analyse hochdynamischer Bewegungen erwiesen. Daher können die Modelle dazu beitragen, hochdynamische Bewegungen besser zu verstehen und Muskelverletzungen bei Sportlern vorzubeugen. Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:355-epub-551061 CY - Regensburg ER - TY - CHAP A1 - Zürner, Christian ED - Bedford-Strohm, Heinrich ED - Höhne, Florian ED - Reitmeier, Tobias T1 - What needs saying and how? Reflections on content and methodology in the profile of Public Theology T2 - Contextuality and Intercontextuality in Public Theology. - Proceedings from the Bamberg Conference 23.-25.06.2011. - Bamberg 23.06.2011-25.06.2011. - (Theologie in der Öffentlichkeit : 4) Y1 - 2013 SN - 978-3-643-90189-7 SP - 335 EP - 338 PB - Lit Verlag CY - Berlin ER - TY - THES A1 - Souza Jr., Luis Antonio de T1 - Computer-assisted diagnosis of Barrett's esophagus using machine learning techniques T1 - Auxílio ao diagnóstico automático do esôfago de Barrett utilizando aprendizado de máquina N2 - Esophageal adenocarcinoma is an illness that is usually hard to detect at the early stages in the presence of Barrett’s esohagus. The development of automatic evaluation systems of such illness may be very useful, thus assisting the experts in the neoplastic region detection. With the strong growth of machine learning techniques aiming to improve the effectivess of medical diagnosis, the use of such approaches characterizes a strong scenario to be explored for the early diagnosis of esophageal adenocarcinoma. Barrett’s esophagus as a predecessor of adenocarcinoma can be explained by some risk factors, such as obesity, smoking, and late medical diagnosis. This project proposes the development of new computer vision and machine learning techniques to assist the automatic diagnosis of the esophageal adenocarcinioma based on the evaluation of two kind of features: (i) handcrafted features, calculated by means of human knowledge using some image processing technique and; (ii) deeply-learnable features, calculated exclusively based on deep learning techniques. From the extensive application of global and local protocols for the models proposed in this work, the description of cancer-affected images and Barrett’s esophagus-affected samples were generalized and deeply evaluated using, for example, classifiers such as Support Vector Machines, ResNet-50 and the combination of descriptions by handcrafted and deeply-learnable features. Also, the behavior of the automatic definition of key-points within the evaluated techniques was observed, something of a paramount importance nowadays to guarantee transparency and reliability in the decisions made by computational techniques. Thus, this project contributes to both the computational and medical fields, introducing new classifiers, approaches and interpretation of the class generalization process, in addition to proposing fast and precise manners to define cancer, delivering important and novel results concerning the accurate identification of cancer in samples affected by Barrett’s esophagus, showing values around 95% of correct identification rates and arranged in a collection of scientific works developed by the author during the research period and submitted/published to date. KW - Machine Learning KW - Barrett's esophagus KW - Deep Learning KW - handcrafted features KW - deeply-learnable features KW - convolutional neural networks KW - interpretability Y1 - 2022 PB - Universidade Federal de São Carlos ER -