Benjamin Berkels

Ten numbers from the Laplace-Beltrami spectrum facilitate training-free classification of protein structures based on surface shape

The paper proposes using the first 10 Laplace-Beltrami eigenvalues to classify protein structures by surface shape, achieving high accuracy on benchmark datasets.

CellSium - Versatile Cell Simulator for Microcolony Ground Truth Generation

CellSium generates synthetic microcolony images for training deep learning segmentation models in microfluidic live-cell imaging.