Introduction This project focuses on medical image segmentation using Flow Matching techniques. It's capable of processing medical image data of various modalities and dimensions, such as MoNuSeg (nuclei segmentation) and Prostate MRI (2D slice segmentation). The project provides a complete pipeline including 3D original data to 2D slice preprocessing, data augmentation, model training, and performance evaluation.
1.Clone the repository:
git clone https://github.com/Elzat-hub/FlowSeg.git2.Create a Conda environment
conda create -n flowseg python=3.8.12
conda activate flowseg3.Install dependencies:
pip install -r requirements.txt