SelfAdapt: Unsupervised Domain Adaptation of Cell Segmentation Models

TL;DR

SelfAdapt is an unsupervised method that adapts pre-trained cell segmentation models to new domains without labels, significantly improving performance and enhancing existing supervised models.

Contribution

It introduces a label-free unsupervised domain adaptation technique for cell segmentation, combining student-teacher training with L2-SP regularization and stopping criteria.

Findings

Achieved up to 29.64% improvement in AP0.5 on benchmark datasets.

Effective in adapting models without requiring annotated data.

Can enhance models that were previously fine-tuned with supervision.

Abstract

Deep neural networks have become the go-to method for biomedical instance segmentation. Generalist models like Cellpose demonstrate state-of-the-art performance across diverse cellular data, though their effectiveness often degrades on domains that differ from their training data. While supervised fine-tuning can address this limitation, it requires annotated data that may not be readily available. We propose SelfAdapt, a method that enables the adaptation of pre-trained cell segmentation models without the need for labels. Our approach builds upon student-teacher augmentation consistency training, introducing L2-SP regularization and label-free stopping criteria. We evaluate our method on the LiveCell and TissueNet datasets, demonstrating relative improvements in AP0.5 of up to 29.64% over baseline Cellpose. Additionally, we show that our unsupervised adaptation can further improve models that were previously fine-tuned with supervision. We release SelfAdapt as an easy-to-use extension of the Cellpose framework. The code for our method is publicly available at https: //github.com/Kainmueller-Lab/self_adapt.