ItsMaxNorm/ultralivecell

Cell Segmentation Masks Dataset

This repository contains NumPy mask files generated using AutoSeg-SAM2 for cell segmentation experiments.

Dataset Structure

The dataset is organized by experiment type and image number:

{experiment_type}/
  └── image_{number}/
      └── cached_masks.npy

Experiments Included

Cdx2_Gata6_Oct4

Images: image_2, image_3, image_4

Sox2_Sox17

Images: image_1, image_2, image_3

Usage

To load the NumPy mask files:

import numpy as np

# Example for loading a mask file
masks = np.load('Cdx2_Gata6_Oct4/image_1/cached_masks.npy', allow_pickle=True)

File Format

Each .npy file contains cell segmentation masks generated using AutoSeg-SAM2, a modification of the Segment Anything Model 2 (SAM2) for automated cell segmentation.

Citation

If you use this dataset in your research, please cite:

@misc{yourarticle2025,
    title={Automated Cell Segmentation using SAM2},
    author={Kumar, Komal and [Other Authors]},
    year={2025},
    publisher={ArXiv},
    howpublished={\url{https://arxiv.org/abs/xxxx.xxxxx}}
}

License

This dataset is provided under [specify license]. Please refer to the LICENSE file for more details.

Contact

For questions or inquiries, please contact Komal Kumar.