Subphase-Labeled Mitotic Dataset for AI-powered Cell Cycle Analysis - Code for Model Training, Inference and Evaluation

1. Create a virtual environment with python3.8: conda create -n mmdet python=3.8
2. Install requirements.txt: pip install -r requirements.txt
3. Install MMDetection-related packages:
   • pip install -U openmim
   • mim install mmengine
   • mim install "mmcv==2.1.0"
   • mim install mmpretrain==1.2.0
   • mim install mmdet
4. Clone the mmdetection repository for the network config files: git clone https://github.com/open-mmlab/mmdetection.git
5. Change the value of the _base_ variable to your config file path in configs/convnext_base_fold1.py
6. Download the model checkpoint file from the shared drive (midog2025/models directory)
7. Run predict.py

Example

python predict.py --config_file configs/convnext_base_fold1.py --model_path models/convnext_base_fold1.py --image_dir path/to/your/images --out_dir path/to/output