Some codes for our experiments, you can use and change it by youself

Suport Method

• DeepLabV3: Rethinking Atrous Convolution for Semantic Image Segmentation

• DeepLabV3 Plus: Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation

• DANet: Dual Attention Network for Scene Segmentation

• EncNet: Context Encoding for Semantic Segmentation

• ESPNet V2: ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network

• UNet: U-Net: Convolutional Networks for Biomedical Image Segmentation

• PFSeg: Patch-Free 3D Medical Image Segmentation Driven by Super-Resolution Technique and Self-Supervised Guidance

• PSPNet: Pyramid Scene Parsing Network

• STDCNet: Rethinking BiSeNet For Real-time Semantic Segmentation

• SCS-Net: SCS-Net: A Scale and Context Sensitive Network for Retinal Vessel Segmentation

• SA-UNet: SA-UNet: Spatial Attention U-Net for Retinal Vessel Segmentation

• BiSeNetV2: BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation

• D2SF: Rethinking Dual-Stream Super-Resolution Semantic Learning in Medical Image Segmentation Official

• SuperVessel: SuperVessel: Segmenting High-resolution Vessel from Low-resolution Retinal Image Official

• SS-MAF Hard Exudate Segmentation Supplemented by Super-Resolution with Multi-scale Attention Fusion Module Offficial

• Learnable Ophthalmology SAM

• DPT: Vision Transformers for Dense Prediction

Usage

• Clone or download this repository to your computer

  git clone https://github.com/Qsingle/imed_vision.git
  

• Follow the samples at the configs directory to set the parameters used to train.

• Run the training script (e.g. vessel_segmentation_train.py).

  python  scripts/vessel_segmentation_train.py --json_path ./configs/iostar_vessel_segmentation.json
  

TODO

• [x] Update the structure for this repository
• [ ] Add the scripts for classification model train

Acknowledgment

Segment-Anything

PFSeg

Cititions

@inproceedings{hu2022supervessel,
  title={Supervessel: Segmenting high-resolution vessel from low-resolution retinal image},
  author={Hu, Yan and Qiu, Zhongxi and Zeng, Dan and Jiang, Li and Lin, Chen and Liu, Jiang},
  booktitle={Chinese Conference on Pattern Recognition and Computer Vision (PRCV)},
  pages={178--190},
  year={2022},
  organization={Springer Nature Switzerland Cham}
}

@inproceedings{zhang2022hard,
  title={Hard Exudate Segmentation Supplemented by Super-Resolution with Multi-scale Attention Fusion Module},
  author={Zhang, Jiayi and Chen, Xiaoshan and Qiu, Zhongxi and Yang, Mingming and Hu, Yan and Liu, Jiang},
  booktitle={2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)},
  pages={1375--1380},
  year={2022},
  organization={IEEE}
}

@article{qiu2023learnable,
  title={Learnable ophthalmology sam},
  author={Qiu, Zhongxi and Hu, Yan and Li, Heng and Liu, Jiang},
  journal={arXiv preprint arXiv:2304.13425},
  year={2023}
}

@article{qiu2023rethinking,
  title={Rethinking Dual-Stream Super-Resolution Semantic Learning in Medical Image Segmentation},
  author={Qiu, Zhongxi and Hu, Yan and Chen, Xiaoshan and Zeng, Dan and Hu, Qingyong and Liu, Jiang},
  journal={IEEE Transactions on Pattern Analysis \& Machine Intelligence},
  number={01},
  pages={1--14},
  year={2023},
  publisher={IEEE Computer Society}
}