A5101712095- Bat Biology and Ecology Studies
- Image Processing Techniques and Applications
- Cell Image Analysis Techniques
- Image and Signal Denoising Methods
- Wildlife Ecology and Conservation
- AI in cancer detection
- Advanced Image Processing Techniques
- Advanced Fluorescence Microscopy Techniques
- Greenhouse Technology and Climate Control
- Innovations in Aquaponics and Hydroponics Systems
- Water Quality Monitoring Technologies
- Entomological Studies and Ecology
- Light effects on plants
- Radiomics and Machine Learning in Medical Imaging
- Advanced Neural Network Applications
- Medical Image Segmentation Techniques
- Digital Holography and Microscopy
- Advanced Image and Video Retrieval Techniques
- Magnetic and Electromagnetic Effects
- Gene expression and cancer classification
- Insect and Arachnid Ecology and Behavior
- Solar-Powered Water Purification Methods
- Animal Behavior and Reproduction
- Single-cell and spatial transcriptomics
- Neurogenesis and neuroplasticity mechanisms
Humboldt-Universität zu Berlin
2010-2025
Max Planck Institute of Molecular Cell Biology and Genetics
2017-2021
Center for Systems Biology Dresden
2017-2020
Technical University of Darmstadt
2010-2019
Leibniz Institute of Freshwater Ecology and Inland Fisheries
2015
University of Bamberg
2014
University Medical Center Freiburg
2007-2012
University of Bonn
1977-2010
Universitätsklinikum Tübingen
2005-2006
Otto-von-Guericke University Magdeburg
2003
Many state-of-the-art image restoration approaches do not scale well to larger images, such as megapixel images common in the consumer segment. Computationally expensive optimization is often culprit. While efficient alternatives exist, they have reached same level of quality. The goal this paper develop an effective approach that offers both computational efficiency and high To end we propose shrinkage fields, a random field-based architecture combines model algorithm single unit....
Accurate detection and segmentation of cell nuclei in volumetric (3D) fluorescence microscopy datasets is an important step many biomedical research projects. Although automated methods for these tasks exist, they often struggle images with low signal-to-noise ratios and/or dense packing nuclei. It was recently shown 2D that issues can be alleviated by training a neural network to directly predict suitable shape representation (star-convex polygon) In this paper, we adopt extend approach 3D...
AEI Aquaculture Environment Interactions Contact the journal Facebook Twitter RSS Mailing List Subscribe to our mailing list via Mailchimp HomeLatest VolumeAbout JournalEditorsTheme Sections 7:179-192 (2015) - DOI: https://doi.org/10.3354/aei00146 A new concept for aquaponic systems improve sustainability, increase productivity, and reduce environmental impacts Werner Kloas1,3,4,*, Roman Groß1, Daniela Baganz2, Johannes Graupner1, Henrik Monsees1, Uwe Schmidt4, Georg Staaks2, Johanna...
Determining the spatial organization and morphological characteristics of molecularly defined cell types is a major bottleneck for characterizing architecture underpinning brain function. We developed Expansion-Assisted Iterative Fluorescence In Situ Hybridization (EASI-FISH) to survey gene expression in tissue, as well turnkey computational pipeline rapidly process large EASI-FISH image datasets. was optimized thick sections (300 μm) facilitate reconstruction spatio-molecular domains that...
Non-blind deblurring is an integral component of blind approaches for removing image blur due to camera shake. Even though learning-based methods exist, they have been limited the generative case and are computationally expensive. To this date, manually-defined models thus most widely used, limiting attained restoration quality. We address gap by proposing a discriminative approach non-blind deblurring. One key challenge that kernel in use at test time not known advance. this, we analyze...
This work addresses the task of non-blind image deconvolution. Motivated to keep up with constant increase in size, megapixel images becoming norm, we aim at pushing limits efficient FFT-based techniques. Based on an analysis traditional and more recent learning-based methods, generalize existing discriminative approaches by using powerful regularization, based convolutional neural networks. Additionally, propose a simple, yet effective, boundary adjustment method that alleviates problematic...
Abstract Deep learning-based approaches are revolutionizing imaging-driven scientific research. However, the accessibility and reproducibility of deep workflows for imaging scientists remain far from sufficient. Several tools have recently risen to challenge democratizing learning by providing user-friendly interfaces analyze new data with pre-trained or fine-tuned models. Still, few existing models interoperable between these tools, critically restricting a model’s overall utility...
Instance segmentation and classification of nuclei is an important task in computational pathology. We show that StarDist, a deep learning method originally developed for fluorescence microscopy, can be extended successfully applied to histopathology images. This substantiated by conducting experiments on the Lizard dataset, through entering Colon Nuclei Identification Counting (CoNIC) challenge 2022, where our approach achieved first spot leaderboard both preliminary final test phase.
Nuclear detection, segmentation and morphometric profiling are essential in helping us further understand the relationship between histology patient outcome. To drive innovation this area, we setup a community-wide challenge using largest available dataset of its kind to assess nuclear cellular composition. Our challenge, named CoNIC, stimulated development reproducible algorithms for recognition with real-time result inspection on public leaderboards. We conducted an extensive...
Markov random fields (MRFs) are popular and generic probabilistic models of prior knowledge in low-level vision. Yet their generative properties rarely examined, while application-specific non-probabilistic learning gaining increased attention. In this paper we revisit the aspects MRFs, analyze quality common image priors a fully application-neutral setting. Enabled by general class MRFs with flexible potentials an efficient Gibbs sampler, find that do not capture statistics natural images...
Two clinical isolates of Vibrio alginolyticus from New Jersey are reported, one a mixed stump infection and the other grown in pure culture conjunctival discharge man with conjunctivitis. The biochemical characteristics antibiotic susceptibilities these two presented. Human infections caused by V. reviewed.
Fluorescence microscopy is a key driver of discoveries in the life-sciences, with observable phenomena being limited by optics microscope, chemistry fluorophores, and maximum photon exposure tolerated sample. These limits necessitate trade-offs between imaging speed, spatial resolution, light exposure, depth. In this work we show how image restoration based on deep learning extends range biological microscopy. On seven concrete examples demonstrate images can be restored even if 60-fold...
Conditional random fields (CRFs) are popular discriminative models for computer vision and have been successfully applied in the domain of image restoration, especially to denoising. For deblurring, however, approaches mostly lacking. We posit two reasons this: First, blur kernel is often only known at test time, requiring any approach cope with considerable variability. Second, given this variability it quite difficult construct suitable features prediction. To address these challenges we...
Removal of noise from fluorescence microscopy images is an important first step in many biological analysis pipelines. Current state-of-the-art supervised methods employ convolutional neural networks that are trained with clean (ground-truth) images. Recently, it was shown self-supervised image denoising blind spot achieves excellent performance even when ground-truth not available, as common microscopy. However, these approaches, e.g. Noise2Void (N2V), generally assume pixel-wise...
Live-cell imaging has revolutionized our understanding of dynamic cellular processes in bacteria and eukaryotes. Although similar techniques have been applied to the study halophilic archaea [1Bisson-Filho A.W. Zheng J. Garner E. Archaeal imaging: leading hunt for new discoveries.Mol. Biol. Cell. 2018; 29: 1675-1681Crossref PubMed Scopus (13) Google Scholar, 2Eun Y.J. Ho P.Y. Kim M. LaRussa S. Robert L. Renner L.D. Schmid A. Amir cells share common size control with despite noisier growth...
Conventional non-blind image deblurring algorithms involve natural priors and maximum a-posteriori (MAP) estimation. As a consequence of MAP estimation, separate pre-processing steps such as noise estimation training the regularization parameter are necessary to avoid user interaction. Moreover, estimates involving standard have been found lacking in terms restoration performance. To address these issues we introduce an integrated Bayesian framework that unifies thus freeing tediously...
To reduce the rock wool waste, present study is focused on evaluation of sheep wool, Sphagnum and hemp slabs, which may be used as replacement for growing substrate hydroponic tomato production. As such, physical chemical properties substrates, plant growth, yield, fruit characteristics, well primary secondary metabolites tomatoes were considered. The marketable yield plants grown in slabs (12.8 kg -1 ) was reduced to only a small extent compared produced by (13.8 ). Sheep (12.3 (10.4 ),...
Estimation of optical aberrations from volumetric intensity images is a key step in sensorless adaptive optics for 3D microscopy. Recent approaches based on deep learning promise accurate results at fast processing speeds. However, collecting ground truth microscopy data training the network typically very difficult or even impossible thereby limiting this approach practice. Here, we demonstrate that neural networks trained only simulated yield predictions real experimental images. We...
This is the first study who presents an approach to predict secondary metabolites content in tomatoes using multivariate time series classification of greenhouse sensor data, which includes climatic conditions as well photosynthesis and transpiration rates. The aim was find necessary a determine maximum metabolites, higher levels fruits can promote human health. For this, we defined multiple tasks derived suitable function. Cross-validated high accuracy results demonstrate effectiveness...