DiPro Checkpoints
This repository contains the trained models for Multimodal Disease Progression Modeling via Spatiotemporal Disentanglement and Multiscale Alignment.
Our paper is available at: [https://arxiv.org/pdf/2510.11112]
The code is available at: [https://github.com/Chenliu-svg/DiPro]
The model files are large (.ckpt format) and are managed using Git LFS.
Models are organized in folders named according to the task: disease_progression, mortality, and length_of_stay.
Within each task folder, models trained with three different random seeds are provided.
Download the Models
Use Git with Git LFS to clone the full repository:
# Install Git LFS (if not already installed)
git lfs install
# Clone the repository
git clone https://huggingface.co/Chenhihihi/DiPro_model.git