A5062748514- Cell Image Analysis Techniques
- Image Processing Techniques and Applications
- AI in cancer detection
- Advanced Fluorescence Microscopy Techniques
- Single-cell and spatial transcriptomics
- Generative Adversarial Networks and Image Synthesis
- Radiomics and Machine Learning in Medical Imaging
- Advanced Vision and Imaging
- Cellular Mechanics and Interactions
- Zebrafish Biomedical Research Applications
- Digital Imaging for Blood Diseases
- Medical Image Segmentation Techniques
- Developmental Biology and Gene Regulation
- Smart Agriculture and AI
- Medical Imaging Techniques and Applications
- Remote Sensing and LiDAR Applications
- COVID-19 diagnosis using AI
- Retinal Imaging and Analysis
- 3D Shape Modeling and Analysis
- Advanced Image and Video Retrieval Techniques
- Animal Genetics and Reproduction
- Neuroscience and Neural Engineering
- Advanced Neural Network Applications
- Advanced biosensing and bioanalysis techniques
- Neurobiology and Insect Physiology Research
Karlsruhe Institute of Technology
2014-2024
RWTH Aachen University
2018-2024
Maastricht University Medical Centre
2022
California Institute of Technology
2018
Institut für Assistenzsysteme und Qualifizierung
2017
Janelia Research Campus
2016
Helix (United States)
2016
Howard Hughes Medical Institute
2016
Heidelberg University
2013
University of Freiburg
2013
Highlighted Article: In situ hybridization chain reaction (HCR) v3.0 exploits automatic background suppression to enable multiplexed quantitative mRNA imaging and flow cytometry with dramatically enhanced performance ease of use.
Generalized nucleus segmentation techniques can contribute greatly to reducing the time develop and validate visual biomarkers for new digital pathology datasets. We summarize results of MoNuSeg 2018 Challenge whose objective was generalizable nuclei in pathology. The challenge an official satellite event MICCAI conference which 32 teams with more than 80 participants from geographically diverse institutes participated. Contestants were given a training set 30 images seven organs annotations...
Abstract Recent advances in computer vision have shown promising results image generation. Diffusion probabilistic models generated realistic images from textual input, as demonstrated by DALL-E 2, Imagen, and Stable Diffusion. However, their use medicine, where imaging data typically comprises three-dimensional volumes, has not been systematically evaluated. Synthetic may play a crucial role privacy-preserving artificial intelligence can also be used to augment small datasets. We show that...
Positional information is a central concept in developmental biology. In developing organs, positional can be idealized as local coordinate system that arises from morphogen gradients controlled by organizers at key locations. This offers plausible mechanism for the integration of molecular networks operating individual cells into spatially coordinated multicellular responses necessary organization emergent forms. Understanding how cues guide morphogenesis requires quantification gene...
Recent advances in computer vision have led to significant progress the generation of realistic image data, with denoising diffusion probabilistic models proving be a particularly effective method. In this study, we demonstrate that can effectively generate fully-annotated microscopy data sets through an unsupervised and intuitive approach, using rough sketches desired structures as starting point. The proposed pipeline helps reduce reliance on manual annotations when training deep...
Signaling filopodia, termed cytonemes, are dynamic actin-based membrane structures that regulate the exchange of signaling molecules and their receptors within tissues. However, how cytoneme formation is regulated remains unclear. Here, we show Wnt/planar cell polarity (PCP) autocrine controls emergence cytonemes subsequently control paracrine Wnt/β-catenin signal activation. Upon binding Wnt family member Wnt8a, receptor tyrosine kinase Ror2 becomes activated. Ror2/PCP leads to induction...
Abstract Accurate quantification of heartbeats in fish models is an important readout to study cardiovascular biology, disease states and pharmacology. However, dependence on anaesthesia, laborious sample orientation or requirement for fluorescent reporters have hampered the use high-throughput heartbeat analysis. To overcome these limitations, we established efficient screening assay employing automated label-free heart rate determination randomly oriented, non-anesthetized medaka ( Oryzias...
Background Clinicians consider both imaging and nonimaging data when diagnosing diseases; however, current machine learning approaches primarily from a single modality. Purpose To develop neural network architecture capable of integrating multimodal patient compare its performance to models incorporating modality for up 25 pathologic conditions. Materials Methods In this retrospective study, were extracted the Medical Information Mart Intensive Care (MIMIC) database an internal comprised...
Automated analysis of multi-dimensional microscopy images has become an integral part modern research in life science. Most available algorithms that provide sufficient segmentation quality, however, are infeasible for a large amount data due to their high complexity. In this contribution we present fast parallelized method is especially suited the extraction stained nuclei from images, e.g., developing zebrafish embryos. The idea transform input image based on gradient and normal directions...
ABSTRACT In situ hybridization based on the mechanism of chain reaction (HCR) has addressed multi-decade challenges to imaging mRNA expression in diverse organisms, offering a unique combination multiplexing, quantitation, sensitivity, resolution, and versatility. Here, with third-generation HCR, we augment these capabilities using probes amplifiers that combine provide automatic background suppression throughout protocol, ensuring even if reagents bind non-specifically within sample they...
The quantitative analysis of cellular membranes helps understanding developmental processes at the level. Particularly 3D microscopic image data offers valuable insights into cell dynamics, but error-free automatic segmentation remains challenging due to huge amount generated and strong variations in intensities. In this paper, we propose a new approach that combines discriminative power convolutional neural networks (CNNs) for preprocessing investigates performance three watershed-based...
When clinicians assess the prognosis of patients in intensive care, they take imaging and non-imaging data into account. In contrast, many traditional machine learning models rely on only one these modalities, limiting their potential medical applications. This work proposes evaluates a transformer-based neural network as novel AI architecture that integrates multimodal patient data, i.e., (chest radiographs) (clinical data). We evaluate performance our model retrospective study with 6,125...
The nanoscale arrangement of ligands can have a major effect on the activation membrane receptor proteins and thus cellular communication mechanisms. Here we report technological development use tailored DNA origami-based molecular rulers to fabricate "Multiscale Origami Structures As Interface for Cells" (MOSAIC), enable systematic investigation spacing epidermal growth factor (EGF) EGF (EGFR). MOSAIC-based analyses revealed that distances about 30–40 nm led highest response in EGFR...
Abstract A new era in developmental biology has been ushered by recent advances the quantitative imaging of all-cell morphogenesis living organisms. Here we have developed a light-sheet fluorescence microscopy-based framework with single-cell resolution for identification and characterization subtle phenotypical changes millimeter-sized Such comparative study requires analyses entire ensembles to be able distinguish sample-to-sample variations from definitive changes. We present kinetic...
Increasing data set sizes of 3D microscopy imaging experiments demand for an automation segmentation processes to be able extract meaningful biomedical information. Due the shortage annotated image that can used machine learning-based approaches, approaches are required robust and generalize well unseen data. The Cellpose approach proposed by Stringer et al. [1] proved such a generalist cell instance tasks. In this paper, we extend improve accuracy on further show how formulation gradient...
Logical image understanding involves interpreting and reasoning about the relationships consistency within an image's visual content. This capability is essential in applications such as industrial inspection, where logical anomaly detection critical for maintaining high-quality standards minimizing costly recalls. Previous research (AD) has relied on prior knowledge designing algorithms, which often requires extensive manual annotations, significant computing power, large amounts of data...
We analyze the capabilities of foundation models addressing tedious task generating annotations for animal tracking. Annotating a large amount data is vital and can be make-or-break factor robustness tracking model. Robustness particularly crucial in tracking, as accurate over long time horizons essential capturing behavior animals. However, additional using counterproductive, quality just important. Poorly annotated introduce noise inaccuracies, ultimately compromising performance accuracy...
Understanding cross-subject and cross-device consistency in visual fixation prediction is essential for advancing eye-tracking applications, including attention modeling neuroprosthetics. This study evaluates using an embedded eye tracker integrated into regular-sized glasses, comparing its performance with high-end standalone systems. Nine participants viewed 300 images from the MIT1003 dataset subjective experiments, allowing us to analyze variations patterns various evaluation metrics....
Abstract Summary: The Insight Toolkit offers plenty of features for multidimensional image analysis. Current implementations, however, often suffer either from a lack flexibility due to hard-coded C++ pipelines certain task or by slow execution times, e.g. caused inefficient implementations multiple read/write operations separate filter execution. We present an XML-based wrapper application the that combines performance pure implementation with easy-to-use graphical setup dynamic analysis...
Cranial neural crest (NC) contributes to the developing vertebrate eye. By multidimensional, quantitative imaging, we traced origin of ocular NC cells two distinct populations that differ in maintenance sox10 expression, Wnt signalling, origin, route, mode and destination migration. The first population migrates proximal second cell group populates distal (anterior) part analysing zebrafish pax6a/b compound mutants presenting anterior segment dysgenesis, demonstrate Pax6a/b guide...
The intrinsic genetic program of a cell is not sufficient to explain all the cell's activities. External mechanical stimuli are increasingly recognized as determinants behavior. In epithelial folding event that constitutes beginning gastrulation in Drosophila, future mesoderm leads establishment contractile actomyosin network triggers apical constriction cells and thereby tissue folding. However, some do constrict but instead stretch, even though they share same their constricting neighbors....
State-of-the-art light-sheet and confocal microscopes allow recording of entire embryos in 3D over time (3D+t) for many hours. Fluorescently labeled structures can be segmented tracked automatically these terabyte-scale 3D+t images, resulting thousands cell migration trajectories that provide detailed insights to large-scale tissue reorganization at the cellular level. Here we present EmbryoMiner, a new interactive open-source framework suitable in-depth analyses comparisons embryos,...