🙋 Participant Guide

Participants need to submit a Docker image (in the form of link) in the competition platform we provide, which contains a single model capable of multitasking carotid plaque. We have provided participants with baseline model files and submitted Docker image templates. Participants can obtain the baseline and the Docker submission template (including necessary guidelines) through the following link:

🧱 Baseline File Structure

    Baseline/      
    ├── data/                                         
    │   └── train/
    │       ├── images/        # .h5 image files (long_img & trans_img)                          
    │       └── labels/        # .h5 image files (long_mask, trans_mask, cls)                               
    ├── pretrain/                                             
    │   └── echocare_encoder.pth
    ├── dataset/
    │   ├── csv.py                                        
    │   └── transform.py  
    ├── model/
    │   ├── Echocare.py                                        
    │   └── unet.py
    ├── util/                                       
    │   └── utils.py                                                 
    ├── README.md 
    ├── inference.py       
    ├── requirements.txt 
    ├── split_train_vaild_fold.py                                                   
    └── train.py  
                

For the usage of Baseline, please refer to the readme.md file in the root directory of the project documentation.

🐋 Docker File Structure

    docker_submission_template/      
    ├── models/                                         
    │   └── model.py                                         
    ├── weights/                                              
    │   └── best_model.pth                                  
    ├── Dockerfile
    ├── requirements.txt                                                   
    └── run_infer.py 
                

🎫 Submission Portal

Participants can access through the CSV Challenge platform (http://119.29.231.17:8000/login) and submit the Docker image link. The submitted docker containers will be evaluated on an Ubuntu 20.04 desktop. Detailed information is listed as follows:

• CPU: Intel® Core™ Ultra 7 265K
• RAM: 32G
• NVIDIA GeForce RTX 4090 D (24G)
• CUDA Version: 12.4

‼️ Important Notice

1. During the Validation phase, participants can submit up to 30 times in total.
2. During the testing phase, each participant can submit up to 5 times strictly following the submission format we provide (Note that the entire testing phase can only submit 5 times).
3. Every registered participant is required to make at least 1 submission before the Challenge deadline.
4. We will select the best score from all submissions of each participant in each phase as the participant's score for that phase. The scores and rankings will be updated after submission.