NucSegAI

Introduction

NucSegAI is a deep learning model for automated nuclear segmentation and classification in H&E-stained histology images. NucSegAI is based on the HoverNet backbone, and is trained on H&E-stained images with paired multiplex immunofluorescence imaging data. The current model is optimized for non-small-cell lung cancer images.

Environment setup

1. Create a new conda environment: conda create -n nucsegai python=3.9
2. Activate the environment: conda activate nucsegai
3. Install the library requirements for HoverNet: pip install src_hovernet/requirements.txt
4. Install the stain-tools: pip install staintools

Image preparation

1. Images can be either patches/tiles in .tiff or whole-slide images (WSI) in .svs.

2. Images are assumed to be captured at 40X magnification, with a resolution of 0.25 um/pixel. Improper resolution will generate suboptimal results.

3. During model development, we found staining normalization is critical for generating accurate and consistent results. It is highly recommended to normalize the H&E staining before applying the model:

• python3 stain_norm.py

Inference

1. Download the model weights ("NucSegAI_torch.tar") from the HuggingFace.

2. Place it into a model_bin/ folder under the same directory with this repo.

3. Run src_hovernet/run_tile.sh for tile inference or src_hovernet/run_wsi.sh for WSI inference.

Relevant repositories:

• StainTools: https://github.com/Peter554/StainTools
• HoverNet: https://github.com/vqdang/hover_net