https://github.com/m1dsolo/echocardmae
EchoCardMAE: Video Masked Auto-Encoders Customized for Echocardiography
https://github.com/m1dsolo/echocardmae
camus deep-learning echonet echonet-dynamic mae python3 pytorch videomae
Last synced: 2 months ago
JSON representation
EchoCardMAE: Video Masked Auto-Encoders Customized for Echocardiography
- Host: GitHub
- URL: https://github.com/m1dsolo/echocardmae
- Owner: m1dsolo
- Created: 2025-02-24T01:58:52.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-05-15T04:24:31.000Z (about 1 year ago)
- Last Synced: 2025-05-15T05:20:58.165Z (about 1 year ago)
- Topics: camus, deep-learning, echonet, echonet-dynamic, mae, python3, pytorch, videomae
- Language: Python
- Homepage:
- Size: 20.5 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# EchoCardMAE: Video Masked Auto-Encoders Customized for Echocardiography
## Introduction
## Visualization
### Reconstruction
EchoCardMAE reconstruction results on the EchoNet-Dynamic dataset.
### Segmentation
Segmentation results on the EchoNet-Dynamic and CAMUS dataset.
## Installation
```bash
# remove GIT_LFS_SKIP_SMUDGE=1 if you want to download the pretraining weights
GIT_LFS_SKIP_SMUDGE=1 git clone https://github.com/m1dsolo/EchoCardMAE.git
cd EchoCardMAE
conda create -n EchoCardMAE python=3.10
conda activate EchoCardMAE
pip install -r requirements.txt
git submodule add --depth=1 https://github.com/m1dsolo/yangdl.git yangdl
cd yangdl
pip install -e .
```
Experimental environment:
- PyTorch 2.5.1
- Python 3.10.15
- GPU memory 24GB
## Usage
### Data Preparation
1. EchoNet-Dynamic: [Download](https://echonet.github.io/dynamic/index.html#dataset) to `EchoCardMAE/dataset/EchoNet-Dynamic`
2. CAMUS: [Download](https://humanheart-project.creatis.insa-lyon.fr/database/#collection/6373703d73e9f0047faa1bc8) to `EchoCardMAE/dataset/CAMUS`
3. HMC-QU: [Download](https://www.kaggle.com/datasets/aysendegerli/hmcqu-dataset) to `EchoCardMAE/dataset/hmcqu-dataset`
### Data preprocessing
1. Ejection fraction (EF) prediction:
```bash
python -m echonet.avi2npy
```
2. Segmentation:
```bash
python -m echonet.avi2edes_npy
```
### Pre-training
You can use [pretraining weights](EchoCardMAE.pt) provided by us.
Or you can pretrain the model by yourself:
```bash
python pretrain.py
```
### Fine-tuning
1. EF prediction:
```bash
python -m echonet.train_ef
python -m echonet.val_ef
```
2. Segmentation:
```bash
python -m echonet.train_seg
```
## TODO
- [ ] upload the code of CAMUS and HMC-QU
## Citation
TODO