ProtoKD

KO | EN

gitlite — search

#python #ai #llm #go #machine-learning #rust #deep-learning #linux #docker #macos #awesome #awesome-list #mcp #open-source

(★ 60)

[ICASSP 2023] Prototype Knowledge Distillation for Medical Segmentation with Missing Modality

ProtoKD Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# ProtoKD
This repo includes:  
1. PyTorch official implementation of __Prototype Knowledge Distillation for Medical Segmentation with Missing Modality__  
by __[Shuai Wang](https://scholar.google.com/citations?user=UbGMEyQAAAAJ&hl=en)__, [__Zipei Yan__](https://yanzipei.github.io/), [__Daoan Zhang__](https://dwan.ch/), __Haining Wei__, __Zhongsen Li__, [__Rui Li__](https://scholar.google.com/citations?user=zTByNnsAAAAJ&hl=en&oi=ao).  
IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2023).  
[__arXiv__](https://arxiv.org/abs/2303.09830), [__IEEEXplore__](https://ieeexplore.ieee.org/abstract/document/10095014)  
2. Paper collection of missing modality in medical image segmentation.

## Overview
![Methods](https://raw.githubusercontent.com/SakurajimaMaiii/ProtoKD/master/figures/framework.png)

## Dependencies
```
python==3.8.13
torch==1.12.0
numpy==1.22.3
medpy==0.4.0
simpleitk==2.2.1
tensorboard==2.12.0
tqdm==4.65.0
```
## Dataset
Please download BraTS2018 training set from [here](https://www.med.upenn.edu/sbia/brats2018.html).
For preprocessing
```bash
cd code
python preprocess.py
```
the preprocessed data will be organzied as:
```
--code
--data
  --brats2018  ##pre-processed data, npy format
    --Brats18_2013_0_1.npy
  --train_list.txt
  --val_list.txt
  --test_list.txt
```
## Training
First, train a teacher model
```bash
python pretrain.py --log_dir ../log/teachermodel
```
For baseline (unimodal in paper),
```bash
python train_baseline.py --log_dir ../log/unimodal_modality0 --modality 0
```
where modality=0,1,2,3 denotes using T1/T2/T1ce/Falir images for training.  
For ProtoKD (our method)
```bash
python train_protokd.py --modality 0 --log_dir ../log/protokd_modality0
```
## Test
```
python evaluate.py --model_path ../log/protokd_modality0/model/best_model.pth \
                   --test_modality 0 \
                   --output_path protokd_modality0_outputs
```
If you want to save visualization results (`nii.gz` format, you can open it using [ITK-Snap](http://www.itksnap.org/pmwiki/pmwiki.php) or [3D-slicer](https://www.slicer.org/)), please add `--save_vis`.
```
python evaluate.py --model_path ../log/protokd_modality0/model/best_model.pth \
                   --test_modality 0 \
                   --output_path protokd_modality0_outputs \
                   --save_vis
```
## Results
![results](https://raw.githubusercontent.com/SakurajimaMaiii/ProtoKD/master/figures/results.png)

## Pretrained models
We provide models for teacher, baseline T1, protokd T1 in [Google Drive](https://drive.google.com/drive/folders/1DhCBMn5Z002TzsfRwFzu_pvXjYc4BoCn?usp=sharing).

## Citation
If this code or ProtoKD is useful for your research, please consider citing our paper:
```bibtex
@inproceedings{Wang2023,
  year = {2023},
  author = {Shuai Wang and Zipei Yan and Daoan Zhang and Haining Wei and Zhongsen Li and Rui Li},
  title = {Prototype Knowledge Distillation for Medical Segmentation with Missing Modality},
  booktitle = {ICASSP}
}
```

## Contact
If you have any question, please contact bit.ybws@gmail.com
## 📜 Papers of missing modality in medical image analysis
We have compiled a collection of papers addressing the issue of missing modality in medical image analysis. This resource aims to assist researchers and practitioners interested in this topic. **Papers with publicly available code have been highlighted.** If you're aware of any significant papers that we may have overlooked, please don't hesitate to inform us.
* __Learning with Privileged Multimodal Knowledge for Unimodal Segmentation, [IEEE TMI 2021](https://ieeexplore.ieee.org/document/9567675), [code](https://github.com/cchen-cc/PMKL)__  
* __D2-Net: Dual Disentanglement Network for Brain Tumor Segmentation with Missing Modalities, [IEEE TMI 2022](https://ieeexplore.ieee.org/document/9567675), [code](https://github.com/CityU-AIM-Group/D2Net)__ 
* __mmFormer: Multimodal Medical Transformer for Incomplete Multimodal Learning of Brain Tumor Segmentation, MICCAI 2022, [arXiv](https://arxiv.org/abs/2206.02425), [code](https://github.com/YaoZhang93/mmFormer)__
* __Robust Multimodal Brain Tumor Segmentation via Feature Disentanglement and Gated Fusion, MICCAI 2019, [arXiv](https://arxiv.org/abs/2002.09708), [code](https://github.com/cchen-cc/Robust-Mseg)__  
* __ACN: Adversarial Co-training Network for Brain Tumor Segmentation with Missing Modalities, MICCAI 2021, [arXiv](https://arxiv.org/abs/2106.14591), [code](https://github.com/Wangyixinxin/ACN)__  
* __Hetero-Modal Variational Encoder-Decoder for Joint Modality Completion and Segmentation, MICCAI 2019, [arXiv](https://arxiv.org/abs/1907.11150), [code](https://github.com/ReubenDo/U-HVED)__  
* Knowledge distillation from multi-modal to mono-modal segmentation networks, MICCAI 2020, [arXiv](https://arxiv.org/abs/2106.09564)  
* HeMIS: Hetero-Modal Image Segmentation, MICCAI 2016, [arXiv](https://arxiv.org/abs/1607.05194)  
* Brain Tumor Segmentation on MRI with Missing Modalities, IPMI 2019, [arXiv](https://arxiv.org/abs/1904.07290)  
* Latent Correlation Representation Learning for Brain Tumor Segmentation With Missing MRI Modalities, [IEEE TIP 2021](https://ieeexplore.ieee.org/document/9399263)  
* SMU-Net: Style matching U-Net for brain tumor segmentation with missing modalities, [arXiv](https://arxiv.org/abs/2204.02961)  
* Learning Cross-Modality Representations From Multi-Modal Images, [IEEE TMI 2018](https://ieeexplore.ieee.org/document/8456579)  
* Multi-Domain Image Completion for Random Missing Input Data, [IEEE TMI 2020](https://ieeexplore.ieee.org/abstract/document/9302720)
* Medical Image Segmentation on MRI Images with Missing Modalities: A Review, [arXiv 2022](https://arxiv.org/abs/2203.06217)
* Disentangle First, Then Distill: A Unified Framework for Missing Modality Imputation and Alzheimer’s Disease Diagnosis, [IEEE TMI 2023](https://ieeexplore.ieee.org/abstract/document/10184044)
* __Discrepancy and Gradient-Guided Multi-modal Knowledge Distillation for Pathological Glioma Grading, [MICCAI 2022](https://link.springer.com/chapter/10.1007/978-3-031-16443-9_61), [code](https://github.com/CityU-AIM-Group/MultiModal-learning)__
* __Gradient modulated contrastive distillation of low-rank multi-modal knowledge for disease diagnosis, [Medical Image Analysis 2023](https://www.sciencedirect.com/science/article/abs/pii/S1361841523001342), [code](https://github.com/CityU-AIM-Group/MultiModal-learning)__
* __M3AE: Multimodal Representation Learning for Brain Tumor Segmentation with Missing Modalities, [arXiv 2023](https://arxiv.org/abs/2303.05302), [code](https://github.com/ccarliu/m3ae)__
* __Fundus-Enhanced Disease-Aware Distillation Model for Retinal Disease Classification from OCT Images, MICCAI 2023 [arXiv](https://arxiv.org/abs/2308.00291), [code](https://github.com/xmed-lab/FDDM)__
* __Multi-modal Learning with Missing Modality via Shared-Specific Feature Modelling, [CVPR 2023](https://openaccess.thecvf.com/content/CVPR2023/papers/Wang_Multi-Modal_Learning_With_Missing_Modality_via_Shared-Specific_Feature_Modelling_CVPR_2023_paper.pdf), [code](https://github.com/billhhh/ShaSpec)__
* __M$`^2`$FTrans: Modality-Masked Fusion Transformer for Incomplete Multi-Modality Brain Tumor Segmentation, [IEEE JBHI 2023](https://ieeexplore.ieee.org/document/10288381), [code](https://github.com/Jun-Jie-Shi/M2FTrans).__
* Bootstrapping Chest CT Image Understanding by Distilling Knowledge from X-ray Expert Models, [CVPR 2024](https://openaccess.thecvf.com/content/CVPR2024/html/Cao_Bootstrapping_Chest_CT_Image_Understanding_by_Distilling_Knowledge_from_X-ray_CVPR_2024_paper.html)
* PASSION: Towards Effective Incomplete Multi-Modal Medical Image Segmentation with Imbalanced Missing Rates, [ACM MM 2024](https://dl.acm.org/doi/abs/10.1145/3664647.3681543)
* A Vision Transformer-Based Framework for Knowledge Transfer From Multi-Modal to Mono-Modal Lymphoma Subtyping Models, [IEEE JBHI 2024](https://ieeexplore.ieee.org/abstract/document/10543042)
* MedMAP: Promoting Incomplete Multi-modal Brain Tumor Segmentation with Alignment, [arXiv 2024](https://arxiv.org/abs/2408.09465)

ProtoKD

ProtoKD Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

ProtoKD

ProtoKD Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README