BOrg
收藏BOrg: A Brain Organoid-Based Mitosis Dataset for Automatic Analysis of Brain Diseases
Overview
BOrg is a dataset designed to study mitotic events in the embryonic development of the brain using confocal microscopy images of brain organoids. The dataset utilizes an efficient annotation pipeline with sparse point annotations and techniques that minimize expert effort, overcoming limitations of standard deep learning approaches on sparse data. BOrg adapts and benchmarks state-of-the-art object detection and cell counting models for detecting and analyzing mitotic cells across prophase, metaphase, anaphase, and telophase stages.
Dataset
The dataset consists of a 2D transformation of 4D confocal microscopic images of brain organoids. It is provided in mmdetection format and can directly be used to train models in mmdetection.
Statistics
| Phases | Train | Validation | Total |
|---|---|---|---|
| Prophase | 282 | 82 | 364 |
| Metaphase | 146 | 62 | 208 |
| Anaphase | 69 | 24 | 93 |
| Telophase | 59 | 13 | 72 |
Projections
Two projection methods are employed to transform images from a 4D to a 2D format.
Training
This dataset can be used directly with mmdetection to train detection models.
Citation
bibtex @misc{awais2024BOrg, title={BOrg: A Brain Organoid-Based Mitosis Dataset for Automatic Analysis of Brain Diseases}, author={Mehaboobathunnisa Sahul Hameed and Muhammad Awais and Bidisha Bhattacharya and Orly Reiner and Rao Anwer}, year={2024}, eprint={}, archivePrefix={arXiv}, primaryClass={cs.CL} }




