Refine
Has Fulltext
- yes (1554)
Is part of the Bibliography
- yes (1554) (remove)
Year of publication
- 2021 (1554) (remove)
Document Type
- Journal article (930)
- Doctoral Thesis (527)
- Complete part of issue (45)
- Book (12)
- Working Paper (8)
- Other (7)
- Conference Proceeding (5)
- Master Thesis (5)
- Book article / Book chapter (4)
- Preprint (4)
Language
- English (1140)
- German (410)
- Spanish (3)
- Multiple languages (1)
Keywords
- Wuerzburg (46)
- Wurzburg (45)
- Würzburg (45)
- University (44)
- Universität (44)
- COVID-19 (20)
- inflammation (17)
- boron (13)
- virtual reality (13)
- SARS-CoV-2 (11)
Institute
- Theodor-Boveri-Institut für Biowissenschaften (175)
- Graduate School of Life Sciences (124)
- Institut für Anorganische Chemie (73)
- Medizinische Klinik und Poliklinik I (71)
- Klinik und Poliklinik für Allgemein-, Viszeral-, Gefäß- und Kinderchirurgie (Chirurgische Klinik I) (61)
- Medizinische Klinik und Poliklinik II (60)
- Institut für Psychologie (56)
- Neurologische Klinik und Poliklinik (51)
- Institut für Organische Chemie (49)
- Institut für Informatik (46)
Schriftenreihe
Sonstige beteiligte Institutionen
- DFG Forschungsgruppe 2757 / Lokale Selbstregelungen im Kontext schwacher Staatlichkeit in Antike und Moderne (LoSAM) (2)
- Klinikum Fulda (2)
- Ökologische Station Fabrikschleichach (2)
- Airbus Defence and Space GmbH (1)
- Akademie der Wissenschaften und der Literatur, Mainz (1)
- Apotheke, Universitätsklinikum Würzburg (1)
- Biomedical Center Munich, Department of Physiological Chemistry, Ludwig-Maximilians-Universität München (1)
- Bundeswehr Institute of Radiobiology affiliated to the University of Ulm, Munich, Germany (1)
- Clinical Trial Center (CTC) / Zentrale für Klinische Studien Würzburg (ZKSW) (1)
- Cologne Game Lab (1)
Herpesviruses have mastered host cell modulation and immune evasion to augment productive infection, life-long latency and reactivation thereof 1,2. A long appreciated, yet elusively defined relationship exists between the lytic-latent switch and viral non-coding RNAs 3,4. Here, we identify miRNA-mediated inhibition of miRNA processing as a novel cellular mechanism that human herpesvirus 6A (HHV-6A) exploits to disrupt mitochondrial architecture, evade intrinsic host defense and drive the latent-lytic switch. We demonstrate that virus-encoded miR-aU14 selectively inhibits the processing of multiple miR-30 family members by direct interaction with the respective pri-miRNA hairpin loops. Subsequent loss of miR-30 and activation of miR-30/p53/Drp1 axis triggers a profound disruption of mitochondrial architecture, which impairs induction of type I interferons and is necessary for both productive infection and virus reactivation. Ectopic expression of miR-aU14 was sufficient to trigger virus reactivation from latency thereby identifying it as a readily drugable master regulator of the herpesvirus latent-lytic switch. Our results show that miRNA-mediated inhibition of miRNA processing represents a generalized cellular mechanism that can be exploited to selectively target individual members of miRNA families. We anticipate that targeting miR-aU14 provides exciting therapeutic options for preventing herpesvirus reactivations in HHV-6-associated disorders like myalgic encephalitis/chronic fatigue syndrome (ME/CFS) and Long-COVID.
Hematopoietic cell transplantation (HCT) is a curative therapy for the treatment of malignant and non-malignant bone marrow diseases. The major complication of this treatment is a highly inflammatory reaction called Graft-versus-Host Disease (GvHD). Here, transplanted donor T cells cause massive tissue destruction and inflammation in the main target organs liver, skin and the intestine. Currently, this inflammatory reaction can be treated successfully using strong immunosuppressive agents. One efficient group of immunosuppressants are calcineurin inhibitors such as Cyclosporin A (CsA) and Tacrolimus (FK506). These treatment strategies target all T lymphocytes subsets equally and do not separate GvH from the desirable Graft-versus-Leukemia (GvL) effect. Therefore, we aimed to find immunological targets on alloreactive T cells in order to develop novel treatment strategies, which selectively modulates alloreactive T cells without impairing the GvL effect or hematopoietic immune reconstitution.
The aim of this thesis was to develop a predictive marker panel to track alloreactive T cells in the peripheral blood (PB) of murine allo-HCT recipients. In clinically relevant model of aGvHD we demonstrated that alloreactive T cells have a distinct surface marker expression profile and can be detected in the PB before aGvHD manifestation. Based on our data, we propose a combinatory panel consisting of 4 surface markers (a4b7 integrin, CD162E, CD162P und CD62L) on circulating CD8+ T cells to identify the risk of aGvHD after allo-HCT.
Since tumor necrosis factor receptor superfamily (TNFR SF) members are involved in several immunological processes, we did extensive surface marker expression analysis of several TNFR superfamily members and other immunomodulatory molecules on conventional and regulatory T cells (Tcons vs. Tregs) on different time points during aGvHD progression. The aim of this study was to find subset-specific immunomodulatory molecules on recently activated Tcons and Tregs. We found that GITR, 4-1BB and CD27 were highly expressed on alloreactive and naïve Tregs. In contrast, PD1 expression was highly upregulated on recently activated alloreactive Tcons. The data of this study serves as basis for future approaches, which aim to develop T cell subset specific therapeutic antibody fusion proteins.
a4b7 integrin and CD162P (P-Selectin ligand) are highly upregulated on alloreactive T cells and mediate the infiltration of these cells into GvHD target organs. We developed recombinant (antibody) fusion proteins to target these two homing molecules and could show that antibody-based fusion proteins are superior to ligand-based fusion proteins regarding production efficiency and binding affinity. Therefore, we propose for future studies to focus on the described antibody-based fusion proteins for the selective targeting of T cells.
Since the widely used calcineurin inhibitors are impairing the desirable GvL effect, we investigated if selective NFATc1 inhibition might be a novel strategy to prevent or reduce alloreactivity, while hopefully maintaining the GvL effect. In particular, we addressed the role of the isoform NFATc1 and inhibited its posttranslational modification by SUMO (Small Ubiquitin-related Modifier). Indeed, inhibition of NFATc1 SUMOylation resulted in reduced inflammation and increased Treg frequencies in a murine MHC major mismatch aGvHD model.
Conclusively, we showed that alloreactive T cells can be identified by their surface profile in the PB of allo-HCT recipients before aGvHD symptoms appeared. Furthermore, we introduced a approach to selectively target alloreactive T cells by antibody fusion proteins, which might serve as a novel strategy to separate GvH from GvL. Additionally, we demonstrated that averted posttranslational modification of NFATc1 by SUMOylation serves as potential target to reduce alloreactivity of T cells.
Purpose
To fully automatically derive quantitative parameters from late gadolinium enhancement (LGE) cardiac MR (CMR) in patients with myocardial infarction and to investigate if phase sensitive or magnitude reconstructions or a combination of both results in best segmentation accuracy.
Methods
In this retrospective single center study, a convolutional neural network with a U-Net architecture with a self-configuring framework (“nnU-net”) was trained for segmentation of left ventricular myocardium and infarct zone in LGE-CMR. A database of 170 examinations from 78 patients with history of myocardial infarction was assembled. Separate fitting of the model was performed, using phase sensitive inversion recovery, the magnitude reconstruction or both contrasts as input channels.
Manual labelling served as ground truth. In a subset of 10 patients, the performance of the trained models was evaluated and quantitatively compared by determination of the Sørensen-Dice similarity coefficient (DSC) and volumes of the infarct zone compared with the manual ground truth using Pearson’s r correlation and Bland-Altman analysis.
Results
The model achieved high similarity coefficients for myocardium and scar tissue. No significant difference was observed between using PSIR, magnitude reconstruction or both contrasts as input (PSIR and MAG; mean DSC: 0.83 ± 0.03 for myocardium and 0.72 ± 0.08 for scars). A strong correlation for volumes of infarct zone was observed between manual and model-based approach (r = 0.96), with a significant underestimation of the volumes obtained from the neural network.
Conclusion
The self-configuring nnU-net achieves predictions with strong agreement compared to manual segmentation, proving the potential as a promising tool to provide fully automatic quantitative evaluation of LGE-CMR.
Facial masks have become and may remain ubiquitous. Though important for preventing infection, they may also serve as a reminder of the risks of disease. Thus, they may either act as cues for threat, priming avoidance-related behavior, or as cues for a safe interaction, priming social approach. To distinguish between these possibilities, we assessed implicit and explicit evaluations of masked individuals as well as avoidance bias toward relatively unsafe interactions with unmasked individuals in an approach-avoidance task in an online study. We further assessed Covid19 anxiety and specific attitudes toward mask-wearing, including mask effectiveness and desirability, hindrance of communication from masks, aesthetic appeal of masks, and mask-related worrying. Across one sample of younger (18–35 years, N = 147) and one of older adults (60+ years, N = 150), we found neither an average approach nor avoidance bias toward mask-wearing compared to unmasked individuals in the indirect behavior measurement task. However, across the combined sample, self-reported mask-related worrying correlated with reduced avoidance tendencies toward unmasked individuals when Covid19 anxiety was low, but not when it was high. This relationship was specific to avoidance tendencies and was not observed in respect to explicit or implicit preference for mask-wearing individuals. We conclude that unsafe interaction styles may be reduced by targeting mask-related worrying with public interventions, in particular for populations that otherwise have low generalized Covid19 anxiety.
Descriptors play an important role in point cloud registration. The current state-of-the-art resorts to the high regression capability of deep learning. However, recent deep learning-based descriptors require different levels of annotation and selection of patches, which make the model hard to migrate to new scenarios. In this work, we learn local registration descriptors for point clouds
in a self-supervised manner. In each iteration of the training, the input of the network is merely one unlabeled point cloud. Thus, the whole training requires no manual annotation and manual selection of patches. In addition, we propose to involve keypoint sampling into the pipeline, which further improves the performance of our model. Our experiments demonstrate the capability of our self-supervised local descriptor to achieve even better performance than the supervised model, while being easier to train and requiring no data labeling.
Two dipolar merocyanines consisting of the same π‐conjugated chromophore but different alkyl substituents adopt very different packing arrangements in their respective solid state with either H‐ or J‐type exciton coupling, leading to ultranarrow absorption bands at 477 and 750 nm, respectively, due to exchange narrowing. The social self‐sorting behavior of these push‐pull chromophores in their mixed thin films is evaluated and the impact on morphology as well as opto‐electronical properties is determined. The implementation of this well‐tuned two‐component material with tailored optical features allows to optimize planar heterojunction organic photodiodes with fullerene (C\(_{60}\)) with either dual or single wavelength selectivity in the blue and NIR spectral range with ultranarrow bandwidths of only 11 nm (200 cm\(^{-1}\)) and an external quantum efficiency of up to 18% at 754 nm under 0 V bias. The application of these photodiodes as low‐power consuming heart rate monitors is demonstrated by a reflectance‐mode photoplethysmography (PPG) sensor.
Progressive deterioration of the central nervous system (CNS) is commonly associated with aging. An important component of the neurovasculature is the blood-brain barrier (BBB), majorly made up of endothelial cells joined together by intercellular junctions. The relationship between senescence and changes in the BBB has not yet been thoroughly explored. Moreover, the lack of in vitro models for the study of the mechanisms involved in those changes impede further and more in-depth investigations in the field. For this reason, we herein present an in vitro model of the senescent BBB and an initial attempt to identify senescence-associated alterations within.
Sensitivity analysis for interpretation of machine learning based segmentation models in cardiac MRI
(2021)
Background
Image segmentation is a common task in medical imaging e.g., for volumetry analysis in cardiac MRI. Artificial neural networks are used to automate this task with performance similar to manual operators. However, this performance is only achieved in the narrow tasks networks are trained on. Performance drops dramatically when data characteristics differ from the training set properties. Moreover, neural networks are commonly considered black boxes, because it is hard to understand how they make decisions and why they fail. Therefore, it is also hard to predict whether they will generalize and work well with new data. Here we present a generic method for segmentation model interpretation. Sensitivity analysis is an approach where model input is modified in a controlled manner and the effect of these modifications on the model output is evaluated. This method yields insights into the sensitivity of the model to these alterations and therefore to the importance of certain features on segmentation performance.
Results
We present an open-source Python library (misas), that facilitates the use of sensitivity analysis with arbitrary data and models. We show that this method is a suitable approach to answer practical questions regarding use and functionality of segmentation models. We demonstrate this in two case studies on cardiac magnetic resonance imaging. The first case study explores the suitability of a published network for use on a public dataset the network has not been trained on. The second case study demonstrates how sensitivity analysis can be used to evaluate the robustness of a newly trained model.
Conclusions
Sensitivity analysis is a useful tool for deep learning developers as well as users such as clinicians. It extends their toolbox, enabling and improving interpretability of segmentation models. Enhancing our understanding of neural networks through sensitivity analysis also assists in decision making. Although demonstrated only on cardiac magnetic resonance images this approach and software are much more broadly applicable.
Perinatal inflammatory stress is strongly associated with adverse pulmonary outcomes after preterm birth. Antenatal infections are an essential perinatal stress factor and contribute to preterm delivery, induction of lung inflammation and injury, pre-disposing preterm infants to bronchopulmonary dysplasia. Considering the polymicrobial nature of antenatal infection, which was reported to result in diverse effects and outcomes in preterm lungs, the aim was to examine the consequences of sequential inflammatory stimuli on endogenous epithelial stem/progenitor cells and vascular maturation, which are crucial drivers of lung development. Therefore, a translational ovine model of antenatal infection/inflammation with consecutive exposures to chronic and acute stimuli was used. Ovine fetuses were exposed intra-amniotically to Ureaplasma parvum 42 days (chronic stimulus) and/or to lipopolysaccharide 2 or 7 days (acute stimulus) prior to preterm delivery at 125 days of gestation. Pulmonary inflammation, endogenous epithelial stem cell populations, vascular modulators and morphology were investigated in preterm lungs. Pre-exposure to UP attenuated neutrophil infiltration in 7d LPS-exposed lungs and prevented reduction of SOX-9 expression and increased SP-B expression, which could indicate protective responses induced by re-exposure. Sequential exposures did not markedly impact stem/progenitors of the proximal airways (P63+ basal cells) compared to single exposure to LPS. In contrast, the alveolar size was increased solely in the UP+7d LPS group. In line, the most pronounced reduction of AEC2 and proliferating cells (Ki67+) was detected in these sequentially UP + 7d LPS-exposed lambs. A similar sensitization effect of UP pre-exposure was reflected by the vessel density and expression of vascular markers VEGFR-2 and Ang-1 that were significantly reduced after UP exposure prior to 2d LPS, when compared to UP and LPS exposure alone. Strikingly, while morphological changes of alveoli and vessels were seen after sequential microbial exposure, improved lung function was observed in UP, 7d LPS, and UP+7d LPS-exposed lambs. In conclusion, although sequential exposures did not markedly further impact epithelial stem/progenitor cell populations, re-exposure to an inflammatory stimulus resulted in disturbed alveolarization and abnormal pulmonary vascular development. Whether these negative effects on lung development can be rescued by the potentially protective responses observed, should be examined at later time points.
Sequential optimality conditions for cardinality-constrained optimization problems with applications
(2021)
Recently, a new approach to tackle cardinality-constrained optimization problems based on a continuous reformulation of the problem was proposed. Following this approach, we derive a problem-tailored sequential optimality condition, which is satisfied at every local minimizer without requiring any constraint qualification. We relate this condition to an existing M-type stationary concept by introducing a weak sequential constraint qualification based on a cone-continuity property. Finally, we present two algorithmic applications: We improve existing results for a known regularization method by proving that it generates limit points satisfying the aforementioned optimality conditions even if the subproblems are only solved inexactly. And we show that, under a suitable Kurdyka–Łojasiewicz-type assumption, any limit point of a standard (safeguarded) multiplier penalty method applied directly to the reformulated problem also satisfies the optimality condition. These results are stronger than corresponding ones known for the related class of mathematical programs with complementarity constraints.