Self-supervised Learning of Dense Hierarchical Representations for Medical Image Segmentation

This repository is the implementation of the paper 'Self-supervised Learning of Dense Hierarchical Representations for Medical Image Segmentation' submitted to ISBI 2024.

Current version of code is fully-functional but still requiers adaptation for easy usage, writing documentation and some grooming.

Code is based on vox2vec repo.

Pretrained models

Pretrained weights for CT and MRI models.

Usage

Pretraining

Downstream task