SSnake

Augmented contour scoring snake for instance segmentation of placental separable villi

Yiming Liu, Xiang Tao, Yan Wang, Xia Jiang, Maxim Sergeevich Vonsky, Lubov Borisovna Mitrofanova, Qingli Li

November 2024, Biomedical Signal Processing and Control

This paper proposes SSnake which strengthens DeepSnake by auxiliary supervision task of Contour Scoring.

Note that this method can enhance interpretability and performance simultaneously!

• A picture for predicted contour scores is as follows.

• Architecture of SSnake

Installation

Please see INSTALL.md.

Quickstart by visualizing

We provide the checkpoint for downloading and a script for visualization. You can try this model for analyzing placental histopathology quantitatively.

The SSnake version with Augmented Contour Generator (ACG), Contour Scoring Network (CSN) and Score Recalibration (SR) is focused.

• The config file is documented at configs/ssnake-boxboxaug-shead_alpha006_minshift4_xy1-std.yaml .

• The checkpoint of first fold in cross validation can be downloaded here.

• An example image of placental histopathology is documented here.

To visualize the segmentation results, simply run:

python visualize.py --cfg_file $config_file --img_path $img_path --ckpt_path $ckpt_path

Results should be similar as follows:

Data Structure

The meta data of dataset like dataset_id is registered in lib/datasets/dataset_catalog.py.

Our dataset for recognizing separable villi is called as vcoco.

The dataset files are documented in the following structure.

data/dataset:
└── fold0
│   ├── annotations
│   │   ├── instances_test2017.json
│   │   ├── instances_train2017.json
│   │   └── instances_val2017.json
│   ├── train2017
│   │   └── img_name1.jpg
│   ├── test2017
│   │   └── img_name2.jpg
│   └── val2017
│       └── img_name2.jpg
└── other fold

Training and testing

For better comprehension of the training and testing process, we provide the related commands.

• Training in cross validation:

python train_net.py --cfg_file $config_file

• Testing in cross validation:

python train_net.py --cfg_file $config_file --test

Citation

If you find this code useful for your research, please use the following BibTeX entry.

@article{SSnake,
title = {Augmented contour scoring snake for instance segmentation of placental separable villi},
author = {Yiming Liu and Xiang Tao and Yan Wang and Xia Jiang and Maxim Sergeevich Vonsky and Lubov Borisovna Mitrofanova and Qingli Li},
journal = {Biomedical Signal Processing and Control},
volume = {97},
pages = {106713},
year = {2024},
issn = {1746-8094},
doi = {https://doi.org/10.1016/j.bspc.2024.106713},
}