agenodrei / ms-tcn2-med

Adaption of the MS-TCN++ architecture for medical videos

MS-TCN2 for surgery recordings

Aim of this project is adapting the below action segementation for surgery phase recognition in medical recordings and specifically for cataract surgery:

• based on the great publication and code from the publication MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation (TPAMI 2020)
• MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation.

Datasets

• provide two datasets
• Adaption of the Cataract 101 dataset for this architecture
• Creation of a novel Small Incision Cataract Surgery (SICS) dataset
• Preprocessing and evaluation tools for both datasets

Based on:

@article{li2020ms,
   author={Shi-Jie Li and Yazan AbuFarha and Yun Liu and Ming-Ming Cheng and Juergen Gall},
    journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
    title={MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation}, 
    year={2020},
    volume={},
    number={},
    pages={1-1},
    doi={10.1109/TPAMI.2020.3021756},
}

@inproceedings{farha2019ms,
  title={Ms-tcn: Multi-stage temporal convolutional network for action segmentation},
  author={Farha, Yazan Abu and Gall, Jurgen},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  pages={3575--3584},
  year={2019}
}