CellSeg1: Robust Cell Segmentation with One Training Image
This provides an efficient solution for cell segmentation with minimal annotation effort, benefiting researchers in biomedical imaging by reducing data requirements.
The paper tackles the problem of requiring large annotated datasets for cell segmentation by introducing CellSeg1, which achieves robust segmentation with only one training image containing a few dozen annotations, matching the performance of models trained on over 500 images with an average mAP of 0.81 at 0.5 IoU on 19 diverse datasets.
Recent trends in cell segmentation have shifted towards universal models to handle diverse cell morphologies and imaging modalities. However, for continuously emerging cell types and imaging techniques, these models still require hundreds or thousands of annotated cells for fine-tuning. We introduce CellSeg1, a practical solution for segmenting cells of arbitrary morphology and modality with a few dozen cell annotations in 1 image. By adopting Low-Rank Adaptation of the Segment Anything Model (SAM), we achieve robust cell segmentation. Tested on 19 diverse cell datasets, CellSeg1 trained on 1 image achieved 0.81 average mAP at 0.5 IoU, performing comparably to existing models trained on over 500 images. It also demonstrated superior generalization in cross-dataset tests on TissueNet. We found that high-quality annotation of a few dozen densely packed cells of varied sizes is key to effective segmentation. CellSeg1 provides an efficient solution for cell segmentation with minimal annotation effort.