Constantin Pape

The paper identifies and characterizes an aorta sensory organ in Drosophila that connects internal hemolymph state to central systems regulating feeding behavior and memory.

CochleaNet is a deep learning framework for analyzing cochlea imaging data, enabling segmentation and analysis of cochlear structures and gene therapy products.

The paper investigates the structure and function of otoferlin, showing how it binds Ca2+ and facilitates synaptic vesicle fusion in inner hair cells.

SynapseNet automatically segments synapses in electron micrographs using deep learning, enabling efficient analysis of synaptic structures.

The paper introduces IVEA, a deep-learning platform for automated exocytosis analysis in live-cell imaging, which is faster and more reliable than manual methods.

FeatureForest uses foundation model embeddings with random forests to enable fast, accurate segmentation of complex biological images with minimal labeling.

The paper develops a correlative cryo-light/electron microscopy workflow to target and image synapses with cryo-ET, revealing excitatory synapses.

micro_sam finetunes Segment Anything into a light microscopy generalist and a cell/neurite-focused electron microscopy model, plus napari tools for interactive 2d, volumetric and tracking annotation.

This study creates a spatiotemporal atlas of mouse gastrulation and early organogenesis using spatial transcriptomics and scRNA-seq, revealing gene expression dynamics and mesodermal fate decisions.

This study shows that complement signaling via C3r is critical for perinatal neuronal elimination but not synaptic pruning in the developing brain.

This paper describes OME-Zarr, a cloud-optimized bioimaging file format developed through community collaboration to improve scalability and data access.

MoBIE is a Fiji plugin that enables visualization, sharing, and management of large, multi-modal image data across platforms.

DeePiCt is a deep-learning framework for segmenting and localizing macromolecular complexes in cryo-electron tomograms, showing effectiveness in identifying low-abundance structures across species.

The BioImage Model Zoo provides a community standard, infrastructure, and open repository for sharing, testing, and deploying pre-trained deep learning microscopy models across multiple bioimage analysis tools.

This paper combines CNNs and Random Forests for domain adaptation in image segmentation, using pseudo-labels from a few annotations to improve performance across domains.

The paper identifies 18 cell types in sponges, including neuroid cells with synaptic gene expression, suggesting early neuro-immune system evolution.

The paper introduces next-generation file formats (NGFF) to improve the FAIRness of bioimaging data by addressing limitations of translation and conversion approaches.

The paper improves microscopy video segmentation by retraining a CNN with a Siamese network setup using unlabelled video frames.