---
license: cc-by-nc-sa-4.0
datasets:
- FOMO-MRI/FOMO260K
pipeline_tag: image-segmentation
---

# ResEnc UNet pretrained on FOMO260K using AMAES

This repository contains the checkpoint of the model from the [FOMO260K paper](https://arxiv.org/abs/2506.14432v2).

* Network: Resenc Unet B with 102M parameters ([model definition](https://github.com/Sllambias/asparagus/blob/72f333b9b59cf9c2a5689995b72a06d689707fec/asparagus/modules/networks/resenc_unet.py#L119))
* Pretext Task: [AMAES](https://arxiv.org/abs/2408.00640).
* Dataset: [FOMO260K](https://huggingface.co/datasets/FOMO-MRI/FOMO260K).
* Code: [Asparagus](https://github.com/Sllambias/asparagus).

## Finetuning the model for Segmentation tasks

To finetune the model with the configs used in the paper, simply use the following command
```
asp_finetune_seg \
    --config-name projects/datapaper/finetune/resenc_[dataset]_fewshot.yaml \
    checkpoint_path=path/to/checkpoint.ckpt
```
where `[dataset]` is one of `{cerebrum, wmh, isles, atlas, sbm}`.


## Installing Asparagus

For instructions on how to install asparagus, preprocess your data, and make your own configs, please refer to the [Asparagus Readme](https://github.com/Sllambias/asparagus).