Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/alawryaguila/conddiff

Pytorch code for "Conditional diffusion models for guided anomaly detection in brain MRI using fluid-driven anomaly randomization"

https://github.com/alawryaguila/conddiff

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (6.5%) to scientific vocabulary

Keywords

anomaly-detection diffusion-models fluid-driven-anomaly-randomization generative-models
Last synced: 6 months ago · JSON representation

Repository

Pytorch code for "Conditional diffusion models for guided anomaly detection in brain MRI using fluid-driven anomaly randomization"

Basic Info
  • Host: GitHub
  • Owner: alawryaguila
  • License: mit
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 999 KB
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
anomaly-detection diffusion-models fluid-driven-anomaly-randomization generative-models
Created 7 months ago · Last pushed 7 months ago
Metadata Files
Readme License

README.md

This repository contains official code for the DGM4MICCAI 2025 paper "Conditional diffusion models for guided anomaly detection in brain MRI using fluid-driven anomaly randomization".

To run the scripts it is necessary to edit the dataloaders and configuration files by editing the example paths. The code in this repository was run using Python 3.11.

Install the required libraries

bash bash ./install.sh

First stage model: AutoencoderKL training

bash python ./scripts/training_autoencoderKL.py which uses the parameters in aekl_ad_3d.yaml

Second stage model: CondDiff training

bash python ./scripts/train_ddpm_pl_cunet.py --config ./conddiff/configs/precalc/train_conddiff_healthy_synthetic.yaml

LDM training

bash python ./scripts/train_ddpm_pl_unet.py --config ./conddiff/configs/healthy/train_unet_healthy.yaml

cLDM training

bash python ./scripts/train_ddpm_pl_cunet.py --config ./conddiff/configs/precalc/train_condunet_healthy_synthetic.yaml

ICDM-3D training

bash python ./scripts/train_ddpm_pl_cunet.py --config ./conddiff/configs/cond_baseline/train_condunet.yaml

VAE training

bash python ./scripts/training_vaebaseline.py which uses the parameters in aekl_ad_3d_vae.yaml

Owner

  • Login: alawryaguila
  • Kind: user

GitHub Events

Total
  • Push event: 4
  • Create event: 2
Last Year
  • Push event: 4
  • Create event: 2

Dependencies

autoencoders/pyproject.toml pypi
  • hydra-core *
  • matplotlib *
  • numpy *
  • ordereddict *
  • pandas *
  • pytest ^5.4.1
  • python ^3.6
  • pytorch-lightning *
  • schema *
  • scipy *
  • torch 2.4.*
  • torchvision *