Nico Scherf

AUTOENCODIX provides a standardized framework for training and evaluating autoencoders, demonstrating its utility across multi-modal biological data and offering insights into representation learning trade-offs.

This study uses explainable AI to link brain-age prediction models with neuroimaging features of aging, revealing correlations between voxel relevance and structural brain markers.

This work benchmarks large language models for generating Python code in bio-image analysis, evaluating 18 models on 57 tasks with reference solutions and unit-tests.

linus provides an interactive web-based tool for exploring and sharing large-scale biological trajectory data.

TraCurate curates and annotates cell tracking data from time-lapse microscopy, improving accuracy with minimal user input.

This paper develops an ontological framework for cellular genealogies to support integration of biological knowledge from single-cell experiments.

This paper develops a method combining deep learning and active contours for automated segmentation of neural cell bodies from light-sheet microscopy data.

The paper uses single-cell tracking to show that HSCs from infected mice exhibit heterogeneous movement patterns compared to controls, with increased persistence possibly aiding survival in altered niche environments.

The paper analyzes cell dynamics across all three germlayers in zebrafish gastrulation, showing that cell movement is position-dependent rather than germlayer-specific.