This repository is the implementation of the paper Real-Time Instrument Segmentation in Robotic Surgery Using Auxiliary Supervised Deep Adversarial Learning.
MICCAI Surgical Instrument Segmentation Challenge 2017 dataset is used to conduct all the experiments in this paper. The dataset is split into train and validation set as:
Train Set: 1,2,3,4,5,6
Validation Set: 7,8
Trained model in type wise segmentation
The adversarial learning part is adopted from this repository
If you use this code for your research, please cite our paper.
@article{islam2019real,
title={Real-time instrument segmentation in robotic surgery using auxiliary supervised deep adversarial learning},
author={Islam, Mobarakol and Atputharuban, Daniel Anojan and Ramesh, Ravikiran and Ren, Hongliang},
journal={IEEE Robotics and Automation Letters},
volume={4},
number={2},
pages={2188--2195},
year={2019},
publisher={IEEE}
}