A Biophysically-Conditioned Generative Framework for 3D Brain Tumor MRI Synthesis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Biller, Valentin, Zimmer, Lucas, Erdur, Ayhan Can, Nagar, Sandeep, Rückert, Daniel, Bubeck, Niklas, Weidner, Jonas
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914187142234112
author Biller, Valentin
Zimmer, Lucas
Erdur, Ayhan Can
Nagar, Sandeep
Rückert, Daniel
Bubeck, Niklas
Weidner, Jonas
author_facet Biller, Valentin
Zimmer, Lucas
Erdur, Ayhan Can
Nagar, Sandeep
Rückert, Daniel
Bubeck, Niklas
Weidner, Jonas
contents Magnetic resonance imaging (MRI) inpainting supports numerous clinical and research applications. We introduce the first generative model that conditions on voxel-level, continuous tumor concentrations to synthesize high-fidelity brain tumor MRIs. For the BraTS 2025 Inpainting Challenge, we adapt this architecture to the complementary task of healthy tissue restoration by setting the tumor concentrations to zero. Our latent diffusion model conditioned on both tissue segmentations and the tumor concentrations generates 3D spatially coherent and anatomically consistent images for both tumor synthesis and healthy tissue inpainting. For healthy inpainting, we achieve a PSNR of 18.5, and for tumor inpainting, we achieve 17.4. Our code is available at: https://github.com/valentin-biller/ldm.git
format Preprint
id arxiv_https___arxiv_org_abs_2510_09365
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Biophysically-Conditioned Generative Framework for 3D Brain Tumor MRI Synthesis
Biller, Valentin
Zimmer, Lucas
Erdur, Ayhan Can
Nagar, Sandeep
Rückert, Daniel
Bubeck, Niklas
Weidner, Jonas
Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
Magnetic resonance imaging (MRI) inpainting supports numerous clinical and research applications. We introduce the first generative model that conditions on voxel-level, continuous tumor concentrations to synthesize high-fidelity brain tumor MRIs. For the BraTS 2025 Inpainting Challenge, we adapt this architecture to the complementary task of healthy tissue restoration by setting the tumor concentrations to zero. Our latent diffusion model conditioned on both tissue segmentations and the tumor concentrations generates 3D spatially coherent and anatomically consistent images for both tumor synthesis and healthy tissue inpainting. For healthy inpainting, we achieve a PSNR of 18.5, and for tumor inpainting, we achieve 17.4. Our code is available at: https://github.com/valentin-biller/ldm.git
title A Biophysically-Conditioned Generative Framework for 3D Brain Tumor MRI Synthesis
topic Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2510.09365