README.md

---
license: cc-by-nc-4.0
---
## FLARE Task2 Laptop Seg Dataset

![20220309-FLARE22-Pictures-2.png](https://s2.loli.net/2024/03/14/mJoTYKNxUG9Pbe8.png)

## Data Description
This is the dataset for [MICCAI FLARE 2024-2025 Task2: Abdominal CT Organ Segmentation on Laptop](https://www.codabench.org/competitions/2320/). 
The training set includes 2050 cases, where 50 cases have ground-truth labels from the FLARE22 dataset, 
and the remaining 2000 cases have pseudo labels generated by the FLARE 2022 winning solution.
The old validation set and testing set are merged as a new validation set with 250 cases. 
For those participants who are constrained by computing resources, we also provide an unlabeled core set to develop the methods, 
where 50 unlabeled CT scans are sampled from the original pseudo training set.

### Data Structure

**coreset_train_50_random：**
50 unlabeled CT scans sampled from the train_pseudo_label.

**train_gt_label：**
50 CT scans with ground-truth labels.

**train_pseudo_label：**
2000 CT scans with pseudo labels generated by the FLARE 2022 winning solution.

**validation：**
200 hidden validation set and 50 public validation set.

FLARE-Task2-LaptopSeg/  
├── coreset_train_50_random/  
├── train_gt_label/  
│    ├── imagesTr/  
│    ├── labelsTr/  
│    └── dataset.json  
├── train_pseudo_label/  
│    ├── imagesTr/  
│    ├── pseudo_label_aladdin5_flare22.7z  
│    └── pseudo_label_blackbean_flare22.zip  
├── validation/  
│    ├── Validation-Hidden-Images/  
│    ├── Validation-Public-Images/  
│    └── Validation-Public-Labels/  
└── README.md  

### Dataset Download Instructions

Participants can download the complete dataset using the following Python script:

```python
from huggingface_hub import snapshot_download

local_dir = "./FLARE-Task2-LaptopSeg"
snapshot_download(
    repo_id="FLARE-MedFM/FLARE-Task2-LaptopSeg",
    repo_type="dataset",
    local_dir=local_dir,
    local_dir_use_symlinks=False,
    resume_download=True,
)