Improving 3D deep learning segmentation with biophysically motivated cell synthesis

TL;DR

This paper introduces a biophysically motivated framework for generating realistic 3D cell training data, improving segmentation accuracy in biomedical imaging by integrating shape modeling and a novel GAN scheme.

Contribution

The authors develop a new synthetic data generation method that incorporates biophysical modeling and a GAN scheme to produce high-quality training data for 3D cell segmentation.

Findings

Synthetic data outperforms manual annotations

Biophysical modeling enhances data realism

Improved segmentation accuracy achieved

Abstract

Biomedical research increasingly relies on 3D cell culture models and AI-based analysis can potentially facilitate a detailed and accurate feature extraction on a single-cell level. However, this requires for a precise segmentation of 3D cell datasets, which in turn demands high-quality ground truth for training. Manual annotation, the gold standard for ground truth data, is too time-consuming and thus not feasible for the generation of large 3D training datasets. To address this, we present a novel framework for generating 3D training data, which integrates biophysical modeling for realistic cell shape and alignment. Our approach allows the in silico generation of coherent membrane and nuclei signals, that enable the training of segmentation models utilizing both channels for improved performance. Furthermore, we present a new GAN training scheme that generates not only image data but also matching labels. Quantitative evaluation shows superior performance of biophysical motivated synthetic training data, even outperforming manual annotation and pretrained models. This underscores the potential of incorporating biophysical modeling for enhancing synthetic training data quality.