A single-stage generative framework that brings Flow Matching into a fixed, invertible wavelet domain: fast to sample, controllable, and anatomically faithful.</p>\n","updatedAt":"2026-06-11T09:14:34.658Z","author":{"_id":"62f920854588fe31f4385f1a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62f920854588fe31f4385f1a/coi_1TNFSX7PI3KlBjY_j.png","fullname":"Danilo Danese","name":"danesed","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8230371475219727},"editors":["danesed"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/62f920854588fe31f4385f1a/coi_1TNFSX7PI3KlBjY_j.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2601.05212","authors":[{"_id":"6a292be741060df46307ad90","user":{"_id":"62f920854588fe31f4385f1a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62f920854588fe31f4385f1a/coi_1TNFSX7PI3KlBjY_j.png","isPro":false,"fullname":"Danilo Danese","user":"danesed","type":"user","name":"danesed"},"name":"Danilo Danese","status":"claimed_verified","statusLastChangedAt":"2026-06-11T08:40:19.612Z","hidden":false},{"_id":"6a292be741060df46307ad91","name":"Angela Lombardi","hidden":false},{"_id":"6a292be741060df46307ad92","name":"Matteo Attimonelli","hidden":false},{"_id":"6a292be741060df46307ad93","name":"Giuseppe Fasano","hidden":false},{"_id":"6a292be741060df46307ad94","name":"Tommaso Di Noia","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/62f920854588fe31f4385f1a/JFD_9BxTRM5vGbT_Y8Euz.png"],"publishedAt":"2026-06-08T00:00:00.000Z","submittedOnDailyAt":"2026-06-11T00:00:00.000Z","title":"FlowLet: Conditional 3D Brain MRI Synthesis using Wavelet Flow Matching","submittedOnDailyBy":{"_id":"62f920854588fe31f4385f1a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62f920854588fe31f4385f1a/coi_1TNFSX7PI3KlBjY_j.png","isPro":false,"fullname":"Danilo Danese","user":"danesed","type":"user","name":"danesed"},"summary":"Brain Magnetic Resonance Imaging (MRI) plays a central role in studying neurological development, aging, and diseases. One key application is Brain Age Prediction (BAP), which estimates an individual's biological brain age from MRI data. Effective BAP models require large, diverse, and age-balanced datasets, whereas existing 3D MRI datasets are demographically skewed, limiting fairness and generalizability. Acquiring new data is costly and ethically constrained, motivating generative data augmentation. Current generative methods are often based on latent diffusion models, which operate in learned low dimensional latent spaces to address the memory demands of volumetric MRI data. However, these methods are typically slow at inference, may introduce artifacts due to latent compression, and are rarely conditioned on age, thereby affecting the BAP performance. In this work, we propose FlowLet, a conditional generative framework that synthesizes age-conditioned 3D MRIs by leveraging flow matching within an invertible 3D wavelet domain, helping to avoid reconstruction artifacts and reducing computational demands. Experiments show that FlowLet generates high-fidelity volumes with few sampling steps. Training BAP models with data generated by FlowLet improves performance for underrepresented age groups, and region-based analysis confirms preservation of anatomical structures.","upvotes":1,"discussionId":"6a292be741060df46307ad95","projectPage":"https://danesed.github.io/flowlet-page/","githubRepo":"https://github.com/sisinflab/FlowLet","githubRepoAddedBy":"user","ai_summary":"FlowLet is a conditional generative framework that synthesizes age-conditioned 3D MRIs using flow matching in an invertible 3D wavelet domain, improving brain age prediction performance for underrepresented age groups.","ai_keywords":["brain age prediction","generative data augmentation","latent diffusion models","flow matching","invertible 3D wavelet domain","age-conditioned generation","anatomical structure preservation"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":1,"organization":{"_id":"68ac08e1de1b20624ba6bcf5","name":"sisinflab-ai","fullname":"SisInfLab","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/66950209dff241076cb08f5d/LoblBb8r63vdV41qleoZT.webp"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6a2ae6c2e36bc84d91b6e7cc","avatarUrl":"/avatars/abf4b4c0020f9332b6827952cc53163e.svg","isPro":false,"fullname":"mmgood","user":"mmgood","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"68ac08e1de1b20624ba6bcf5","name":"sisinflab-ai","fullname":"SisInfLab","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/66950209dff241076cb08f5d/LoblBb8r63vdV41qleoZT.webp"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2601/2601.05212.md"}">
FlowLet: Conditional 3D Brain MRI Synthesis using Wavelet Flow Matching
Abstract
FlowLet is a conditional generative framework that synthesizes age-conditioned 3D MRIs using flow matching in an invertible 3D wavelet domain, improving brain age prediction performance for underrepresented age groups.
Brain Magnetic Resonance Imaging (MRI) plays a central role in studying neurological development, aging, and diseases. One key application is Brain Age Prediction (BAP), which estimates an individual's biological brain age from MRI data. Effective BAP models require large, diverse, and age-balanced datasets, whereas existing 3D MRI datasets are demographically skewed, limiting fairness and generalizability. Acquiring new data is costly and ethically constrained, motivating generative data augmentation. Current generative methods are often based on latent diffusion models, which operate in learned low dimensional latent spaces to address the memory demands of volumetric MRI data. However, these methods are typically slow at inference, may introduce artifacts due to latent compression, and are rarely conditioned on age, thereby affecting the BAP performance. In this work, we propose FlowLet, a conditional generative framework that synthesizes age-conditioned 3D MRIs by leveraging flow matching within an invertible 3D wavelet domain, helping to avoid reconstruction artifacts and reducing computational demands. Experiments show that FlowLet generates high-fidelity volumes with few sampling steps. Training BAP models with data generated by FlowLet improves performance for underrepresented age groups, and region-based analysis confirms preservation of anatomical structures.
Community
A single-stage generative framework that brings Flow Matching into a fixed, invertible wavelet domain: fast to sample, controllable, and anatomically faithful.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2601.05212 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2601.05212 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.