AnyStar: Domain randomized universal star-convex 3D instance segmentation

TL;DR

AnyStar introduces a domain-randomized generative model that creates synthetic training data for star-convex 3D instance segmentation, enabling accurate segmentation across diverse biomedical imaging modalities without dataset-specific retraining.

Contribution

The paper presents a novel generative model that produces synthetic training data for universal star-convex 3D segmentation, reducing the need for manual annotation and domain-specific fine-tuning.

Findings

Networks trained on synthetic data accurately segment diverse biological structures.

The approach eliminates the need for retraining or domain adaptation across datasets.

Synthetic training data generalizes well to multiple imaging modalities.

Abstract

Star-convex shapes arise across bio-microscopy and radiology in the form of nuclei, nodules, metastases, and other units. Existing instance segmentation networks for such structures train on densely labeled instances for each dataset, which requires substantial and often impractical manual annotation effort. Further, significant reengineering or finetuning is needed when presented with new datasets and imaging modalities due to changes in contrast, shape, orientation, resolution, and density. We present AnyStar, a domain-randomized generative model that simulates synthetic training data of blob-like objects with randomized appearance, environments, and imaging physics to train general-purpose star-convex instance segmentation networks. As a result, networks trained using our generative model do not require annotated images from unseen datasets. A single network trained on our synthesized data accurately 3D segments C. elegans and P. dumerilii nuclei in fluorescence microscopy, mouse cortical nuclei in micro-CT, zebrafish brain nuclei in EM, and placental cotyledons in human fetal MRI, all without any retraining, finetuning, transfer learning, or domain adaptation. Code is available at https://github.com/neel-dey/AnyStar.