{"id":28807058,"url":"https://github.com/agr78/prlx-gan","last_synced_at":"2026-04-29T19:33:19.213Z","repository":{"id":296178419,"uuid":"960581301","full_name":"agr78/PRLx-GAN","owner":"agr78","description":"Generative modeling and latent projection label denoising approach to create synthetic rim lesions on QSM","archived":false,"fork":false,"pushed_at":"2025-05-29T10:13:12.000Z","size":7455,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-05-29T11:10:58.832Z","etag":null,"topics":["cornell-university","cvpr","cvpr2025","deep-learning","generative-adversarial-network","generative-ai","generative-model","label-noise","label-noise-learning","mri","multiple-sclerosis","paramagnetic-rim-lesion","pytorch","qsm","quantitative-susceptibility-mapping","rim-classifier","stylegan2-ada","synthetic-data","weill"],"latest_commit_sha":null,"homepage":"https://openreview.net/pdf?id=wFkiqB5spT","language":"Shell","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/agr78.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-04-04T17:22:42.000Z","updated_at":"2025-05-29T10:13:15.000Z","dependencies_parsed_at":"2025-06-03T02:03:44.879Z","dependency_job_id":null,"html_url":"https://github.com/agr78/PRLx-GAN","commit_stats":null,"previous_names":["agr78/prlx-gan"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/agr78/PRLx-GAN","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agr78%2FPRLx-GAN","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agr78%2FPRLx-GAN/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agr78%2FPRLx-GAN/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agr78%2FPRLx-GAN/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/agr78","download_url":"https://codeload.github.com/agr78/PRLx-GAN/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/agr78%2FPRLx-GAN/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":260540818,"owners_count":23024894,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cornell-university","cvpr","cvpr2025","deep-learning","generative-adversarial-network","generative-ai","generative-model","label-noise","label-noise-learning","mri","multiple-sclerosis","paramagnetic-rim-lesion","pytorch","qsm","quantitative-susceptibility-mapping","rim-classifier","stylegan2-ada","synthetic-data","weill"],"created_at":"2025-06-18T11:06:36.717Z","updated_at":"2026-04-29T19:33:19.202Z","avatar_url":"https://github.com/agr78.png","language":"Shell","funding_links":[],"categories":[],"sub_categories":[],"readme":"# PRLx-GAN\nRepository for [_Synthetic Generation and Latent Projection Denoising of Rim Lesions in Multiple Sclerosis_](https://openreview.net/forum?id=wFkiqB5spT\u0026referrer=%5Bthe%20profile%20of%20Alexandra%20Grace%20Roberts%5D(%2Fprofile%3Fid%3D~Alexandra_Grace_Roberts1)) published in [Synthetic Data](https://syndata4cv.github.io/) at [CVPR 2025](https://cvpr.thecvf.com/). \n## Summary\nParamagnetic rim lesions (PRLs) are a rare but highly prognostic lesion subtype in multiple sclerosis, visible only on susceptibility ($\\chi$) contrasts. [This work](https://huggingface.co/papers/2505.23353) presents a generative framework to: \u003cbr/\u003e\n* Synthesize new rim lesion maps that address class imbalance in training data \u003cbr/\u003e\n* Enable a novel denoising method to resolve radiologist disagreements on noisy labels, \"ambiguous rim lesions\" \u003cbr/\u003e\n\n\u003cbr/\u003e\n\u003cbr/\u003e\n\n\u003cp align=\"center\"\u003e\n   \u003cimg src=\"./gif/proj_4x.gif\"/\u003e\u003c/br\u003e\n   \u003ci\u003eLatent projection iterations to identify the denoised rim lesion from a \"noisy\" rim lesion.\u003c/i\u003e\n\u003c/p\u003e\n\n\u003cbr/\u003e\n\n## Contents\nUncurated synthetic rim lesion [susceptibilities](https://huggingface.co/datasets/agr78/PRLx-GAN-synthetic-rim) can be found in [`png`](https://github.com/agr78/PRLx-GAN/tree/main/png) \u003cbr/\u003e\nPretrained [weights](https://huggingface.co/agr78/PRLx-GAN) are located in [`net`](https://github.com/agr78/PRLx-GAN/tree/main/net) \u003cbr/\u003e\n\n## Preliminaries\nTo download the pretrained weights, ensure [Git Large File Service](https://git-lfs.com/) is installed\n```\nsudo apt-get install git-lfs\ngit lfs install\n```\nThe `main.sh` script will skip retraining unless `\\your\\QSM\\data` is replaced by a valid path\n\n## Installation\nClone the repository with\n```\ngit clone https://github.com/agr78/PRLx-GAN.git\n```\nNavigate to the repository\n```\ncd PRLx-GAN\n```\nRun the setup script\n```\nsource ./src/main.sh\n```\nWait...then check the generated and denoised outputs\n```\ncd ./out\n```\n\n## Publications\nIf this code is used, please cite the following:\n\u003e [Conference Paper](https://openreview.net/forum?id=wFkiqB5spT\u0026noteId=wFkiqB5spT): A. G. Roberts et al., \"Synthetic Generation and Latent Projection Denoising of Rim Lesions in Multiple Sclerosis,\" Synthetic Data for Computer Vision at CVPR, 2025.\n\u003e \n\n## BibTex\n```bibtex\n@inproceedings{\nroberts2025synthetic,\ntitle={Synthetic Generation and Latent Projection Denoising of Rim Lesions in Multiple Sclerosis},\nauthor={Alexandra Grace Roberts and Ha Manh Luu and Mert Sisman and Alexey V. Dimov and Ceren Tozlu and Ilhami Kovanlikaya and Susan Gauthier and Thanh D. Nguyen and Yi Wang},\nbooktitle={Synthetic Data for Computer Vision Workshop @ CVPR 2025},\nyear={2025},\nurl={https://openreview.net/forum?id=wFkiqB5spT}\n}\n```\n## Acknowledgements\nThis method relies on the [StyleGAN2-ADA](https://github.com/NVlabs/stylegan2-ada-pytorch) architecture developed by [`@tkarras`](https://github.com/tkarras).\n## Contact\nPlease direct questions to [Alexandra G. Roberts](https://github.com/agr78) at agr78@cornell.edu.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fagr78%2Fprlx-gan","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fagr78%2Fprlx-gan","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fagr78%2Fprlx-gan/lists"}