This repository contains the official code for the FOMO25 Challenge of the team FOMO2JOMO. For more information on the challenge, please visit the FOMO25 Challenge website.

This codebase supports three tasks:

1. Task 1: Infarct Detection - Binary classification
2. Task 2: Meningioma Segmentation - Binary segmentation
3. Task 3: Brain Age Regression - Regression

Data for the challenge includes:

• Pretraining Data: 11,187 subjects, 13,900 sessions, 60,529 scans
• Finetuning Data: Limited few-shot data (~20-200 cases per task)

Install required dependencies:

# Install basic dependencies
pip install -e .

# For development
pip install -e ".[dev]"

# For testing
pip install -e ".[test]"

# For all dependencies
pip install -e ".[dev,test]"

While the data included in this challenge is already preprocessed (co-registered, transposed to RAS orientation and defaced/skull-stripped), to run this code, one needs to further preprocess with the following highly opinionated preprocessing steps.

This "Opinionated Preprocessing" can be done in the following way

For preprocessing the pretraining (FOMO60K) data:

python src/data/fomo-60k/preprocess.py --in_path=/path/to/raw/pretrain/data --out_path=/path/to/output/preprocessed/data

This will:

1. Store each tensor in numpy format for easy loading.
2. Treat each scan as a separate datapoint which can be sampled iid.
3. Crop to the minimum bounding box.
4. Z-normalize on a per-volume level.
5. Resample to isotropic (1mm, 1mm, 1mm) spacing.

For preprocessing the finetuning data for tasks 1-3:

python src/data/preprocess/run_preprocessing.py --taskid=1 --source_path=/path/to/raw/finetuning/data

Replace --taskid=1 with --taskid=2 or --taskid=3 for the other tasks.

This will apply a preprocessing akin to the one of the pre-trained data:

1. Assemble each session into a single 4D tensor and store it as a numpy array for easy loading.
2. Treat each scan as a separate datapoint which can be sampled iid.
3. Crop to the minimum bounding box.
4. Z-normalize on a per-volume level.
5. Resample to isotropic (1mm, 1mm, 1mm) spacing.

To pretrain a model using the proposed framework solution:

python src/pretrain.py \
    --save_dir=/path/to/save/models \
    --pretrain_data_dir=/path/to/preprocessed/pretrain/data \
    --model_name=mmunetvae \
    --patch_size=96 \
    --batch_size=2 \
    --epochs=100 \
    --warmup_epochs=5 \
    --num_workers=64 \
    --augmentation_preset=all

Key pretraining parameters:

• --model_name: Supported models include unet_b_lw_dec, unet_xl_lw_dec, etc.
• --patch_size: Size of 3D patches (must be divisible by 8)
• --mask_patch_size: Size of masking unit for MAE (default is 4)
• --mask_ratio: Ratio of patches to mask (default is 0.6)
• --augmentation_preset: Choose from all, basic, or none

To finetune a pretrained model on one of the three tasks:

python src/finetune.py \
    --data_dir=/path/to/preprocessed/data \
    --save_dir=/path/to/save/finetuned/models \
    --pretrained_weights_path=/path/to/pretrained/checkpoint.pth \
    --model_name=mmunetvae \
    --patch_size=96 \
    --taskid=1 \
    --batch_size=2 \
    --epochs=500 \
    --train_batches_per_epoch=100 \
    --augmentation_preset=basic

Key finetuning parameters:

• --taskid: Task ID (1: Infarct Detection, 2: Meningioma Segmentation, 3: Brain Age Regression)
• --model_name: Must match the architecture of the pretrained checkpoint
• --pretrained_weights_path: Path to the pretrained model checkpoint
• --augmentation_preset: Choose from all, basic, or none

The reference implementation was pretrained on 1 A100 GPU with 80GB of memory. Depending on your hardware, you may need to adjust batch sizes and patch sizes accordingly.

If you use this code, please cite:

@article{llambias2024yucca,
  title={Yucca: A deep learning framework for medical image analysis},
  author={Llambias, Sebastian N{\o}rgaard and Machnio, Julia and Munk, Asbj{\o}rn and Ambsdorf, Jakob and Nielsen, Mads and Ghazi, Mostafa Mehdipour},
  journal={arXiv preprint arXiv:2407.19888},
  year={2024}
}

@article{munk2024amaes,
  title={AMAES: Augmented Masked Autoencoder Pretraining on Public Brain MRI Data for 3D-Native Segmentation},
  author={Munk, Asbjørn and Ambsdorf, Jakob and Llambias, Sebastian and Nielsen, Mads},
  journal={MICCAI Workshop on Advancing Data Solutions in Medical Imaging AI (ADSMI 2024)},
  year={2024}
}

Our work is currently under review. Full citation details will be available once published.

@article{brainfm2025,
  title={TITLE},
  author={[Authors]},
  journal={Under Review},
  year={2025}
}