Publication Server of Zuse Institute Berlin (ZIB)
Not a member yet
6648 research outputs found
Sort by
Completion of Missing Parts in Medical Images Using Generative Adversarial Networks and Transformers
No full textImage inpainting (image completion) is a technique to remove undesirable elements or fill missing (repair damaged) sections in an image [1]. In medical images, inpainting can be applied for various applications. Issues like image artifacts (e.g., metal artifacts in CT and MRI) can introduce alterations to medical images. In simpler terms, these artifacts may introduce unwanted elements or anomalies into the medical images. These unwanted elements can be replaced with appropriate intensity in the images. In the context of defect reconstruction, which involves repairing skull defects, medical images are taken. To design an implant for a defective skull, image inpainting can be applied to an image containing the defective skull. This process generates an image with skull tissue, which can then be utilized for implant design. Additionally, in situations involving a restricted field of view, such as when only accessing the lower part of the full-body MRI, predicting the missing information (e.g., the upper part of the full-body MRI) can be beneficial for subsequent tasks, such as extracting bone structures for 3D shape modeling studies. Another application of inpainting is observed during actual stereo-fluoroscopic X-ray measurements. In these measurements, the calibration grid, which is beneficial for correcting distortions, introduces unwanted black dots into the image content. These black dots need to be removed from the image and replaced with the appropriate intensity.
This thesis extends an existing architecture named Mask-aware Transformer (MAT) to fill in missing information in medical scans. The proposed method utilizes Generative Adversarial Networks and a Mask-aware Swin Transformer. Various adjustments, including changes in data type, model architecture, and adaptation for multinode training, have been implemented to address missing information in three different datasets. In one of our datasets, where black dots disrupt the image content, we applied our method to fill in the missing information. The results indicate that our approach outperforms other model-driven methods. In the second dataset, we employed the German National Cohort full-body MRI dataset to train a network for predicting the upper part of the full-body MRI. The outcomes indicate that the generated 2D slices exhibit realism and perform well in both pixel-wise and perceptual metrics. However, when these 2D slices are stacked into a 3D volume, inconsistencies between different slices become apparent. Furthermore, when we compare our 2D-aware image inpainting method to 3D inpainting algorithms used in the Brain Tumor Segmentation 2023 challenge, the results confirm the findings of the earlier experiment. The 2D slices produced are realistic, but there is inconsistency between them.
In summary, theMAT demonstrates its ability to capture global dependencies by utilizing an adapted Swin Transformer for X-ray images, inherently 2D images. The results for 2D slices from a 3D MRI volume are also reasonable because the network is trained specifically on 2D slices. However, due to the lack of awareness regarding this architecture in the third dimension, inconsistencies between slices are observed
Research data for "Fabrication uncertainty guided design optimization of a photonic crystal cavity by using Gaussian processes"
No full textCharacterising information gains and losses when collecting multiple epidemic model outputs
No full textCollaborative comparisons and combinations of epidemic models are used as policy-relevant evidence during epidemic outbreaks. In the process of collecting multiple model projections, such collaborations may gain or lose relevant information. Typically, modellers contribute a probabilistic summary at each time-step. We compared this to directly collecting simulated trajectories. We aimed to explore information on key epidemic quantities; ensemble uncertainty; and performance against data, investigating potential to continuously gain information from a single cross-sectional collection of model results. Methods We compared July 2022 projections from the European COVID-19 Scenario Modelling Hub. Five modelling teams projected incidence in Belgium, the Netherlands, and Spain. We compared projections by incidence, peaks, and cumulative totals. We created a probabilistic ensemble drawn from all trajectories, and compared to ensembles from a median across each model’s quantiles, or a linear opinion pool. We measured the predictive accuracy of individual trajectories against observations, using this in a weighted ensemble. We repeated this sequentially against increasing weeks of observed data. We evaluated these ensembles to reflect performance with varying observed data. Results. By collecting modelled trajectories, we showed policy-relevant epidemic characteristics. Trajectories contained a right-skewed distribution well represented by an ensemble of trajectories or a linear opinion pool, but not models’ quantile intervals. Ensembles weighted by performance typically retained the range of plausible incidence over time, and in some cases narrowed this by excluding some epidemic shapes. Conclusions. We observed several information gains from collecting modelled trajectories rather than quantile distributions, including potential for continuously updated information from a single model collection. The value of information gains and losses may vary with each collaborative effort’s aims, depending on the needs of projection users. Understanding the differing information potential of methods to collect model projections can support the accuracy, sustainability, and communication of collaborative infectious disease modelling efforts. Data availability All code and data available on Github: https://github.com/covid19-forecast-hub-europe/aggregation-info-los
A Vision for Data Management Plans in the NFDI
No full textAt present, data management plans (DMPs) are still often perceived as mere documents for funding agencies providing clarity on how research data will be handled during a funded project, but are not usually actively involved in the processes. However, they contain a great deal of information that can be shared automatically to facilitate active research data management (RDM) by providing metadata to research infrastructures and supporting communication between all involved stakeholders. This position paper brings together a number of ideas developed and collected during interdisciplinary workshops of the Data Management Planning Working Group (infra-dmp), which is part of the section Common Infrastructures of the National Research Data Infrastructure (NFDI) in Germany. We present our vision of a possible future role of DMPs, templates, and tools in the upcoming NFDI service architecture
Optimal discrete pipe sizing for tree-shaped CO2 networks
No full textFor industries like the cement industry, switching to a carbon-neutral production process is impossible. They must rely on carbon capture, utilization and storage (CCUS) technologies to reduce their production processes’ inevitable carbon dioxide (CO2) emissions. For transporting continuously large amounts of CO2, utilizing a pipeline network is the most effective solution; however, building such a network is expensive. Therefore minimizing the cost of the pipelines to be built is extremely important to make the operation financially feasible. In this context, we investigate the problem of finding optimal pipeline diameters from a discrete set of diameters for a tree-shaped network transporting captured CO2 from multiple sources to a single sink. The general problem of optimizing arc capacities in potential-based fluid networks is already a challenging mixed-integer nonlinear optimization problem. The problem becomes even more complex when adding the highly sensitive nonlinear behavior of CO2 regarding temperature and pressure changes. We propose an iterative algorithm splitting the problem into two parts: a) the pipe-sizing problem under a fixed supply scenario and temperature distribution and b) the thermophysical modeling, including mixing effects, the Joule-Thomson effect, and heat exchange with the surrounding environment. We demonstrate the effectiveness of our approach by applying our algorithm to a real-world network planning problem for a CO2 network in Western Germany. Further, we show the robustness of the algorithm by solving a large artificially created set of network instances
An interpretable data-driven prediction model to anticipate scoliosis in spinal muscular atrophy in the era of (gene-) therapies
No full text5q-spinal muscular atrophy (SMA) is a neuromuscular disorder (NMD) that has become one of the first 5% treatable rare diseases. The efficacy of new SMA therapies is creating a dynamic SMA patient landscape, where disease progression and scoliosis development play a central role, however, remain difficult to anticipate. New approaches to anticipate disease progression and associated sequelae will be needed to continuously provide these patients the best standard of care. Here we developed an interpretable machine learning (ML) model that can function as an assistive tool in the anticipation of SMA-associated scoliosis based on disease progression markers. We collected longitudinal data from 86 genetically confirmed SMA patients. We selected six features routinely assessed over time to train a random forest classifier. The model achieved a mean accuracy of 0.77 (SD 0.2) and an average ROC AUC of 0.85 (SD 0.17). For class 1 ‘scoliosis’ the average precision was 0.84 (SD 0.11), recall 0.89 (SD 0.22), F1-score of 0.85 (SD 0.17), respectively. Our trained model could predict scoliosis using selected disease progression markers and was consistent with the radiological measurements. During post validation, the model could predict scoliosis in patients who were unseen during training. We also demonstrate that rare disease data sets can be wrangled to build predictive ML models. Interpretable ML models can function as assistive tools in a changing disease landscape and have the potential to democratize expertise that is otherwise clustered at specialized centers
Kissing polytopes
No full textWe investigate the following question: How close can two disjoint lattice polytopes contained in a fixed hypercube be? This question stems from various contexts where the minimal distance between such polytopes appears in complexity bounds of optimization algorithms. We provide nearly matching bounds on this distance and discuss its exact computation. We also give similar bounds for disjoint rational polytopes whose binary encoding length is prescribed
Varying water activity and momentum transfer on comet 67P/Churyumov-Gerasimenko from its non-gravitational forces and torques
No full textAims. We investigate the ability of a simultaneous fitting of comet 67P/Churyumov-Gerasimenko’s non-gravitational forces, torques, and total water-outgassing rate, as observed by Rosetta, to constrain complex thermophysical models of cometary material.
Methods. We extend the previous work of fitting geographically defined surface outgassing models to the Rosetta observations by testing the effects of a more detailed geomorphological mapping, the resolution of the shape-model used, self-heating by neighbouring facets on the shape-model, thermal inertia in the outgassing solution, and the variation in the momentum coupling between the gas and the nucleus. We also directly compare the non-gravitational acceleration curves available in the literature.
Results. We correct an error in the calculation of pole-orientation in the previous paper. We find that, under the assumptions of the model, non-gravitational forces and torques are driven by water sublimation from the nucleus, thermal inertia and self-heating have only minor effects, spatially uniform activity cannot explain 67P’s non-gravitational dynamics, spatially uniform momentum transfer cannot explain 67P’s non-gravitational dynamics, and different terrain types have different instantaneous responses to insolation.
Conclusions. Consolidated terrain facing south on 67P/Churyumov-Gerasimenko has a high outgassing flux, a steep response to insolation, and a large gas momentum transfer coefficient. Instead, that facing north behaves differently, producing little to no water outgassing, and with a lower momentum transfer efficiency. Dusty terrain also has a lower outgassing rate and momentum transfer efficiency, and either depletes its volatile component or is buried in fall-back as the comet approaches the Sun. Momentum transfer appears correlated with insolation, likely due to an increased enhancement in the gas temperature as the dust it flows through is heated