TumorFlow: Physics-Guided Longitudinal MRI Synthesis of Glioblastoma Growth

• URL: http://arxiv.org/abs/2603.04058v2
• Date: Thu, 05 Mar 2026 06:55:34 GMT
• Title: TumorFlow: Physics-Guided Longitudinal MRI Synthesis of Glioblastoma Growth
• Authors: Valentin Biller, Niklas Bubeck, Lucas Zimmer, Ayhan Can Erdur, Sandeep Nagar, Anke Meyer-Baese, Daniel Rückert, Benedikt Wiestler, Jonas Weidner,
• Abstract summary: Glioblastoma exhibits diverse, infiltrative, and patient-specific growth patterns that are only partially visible on routine MRI.<n>We present a biophysically-conditioned generative framework that synthesizes biologically realistic 3D brain MRI volumes from estimated, spatially continuous tumor-concentration fields.
• Score: 2.3743336223695057
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Glioblastoma exhibits diverse, infiltrative, and patient-specific growth patterns that are only partially visible on routine MRI, making it difficult to reliably assess true tumor extent and personalize treatment planning and follow-up. We present a biophysically-conditioned generative framework that synthesizes biologically realistic 3D brain MRI volumes from estimated, spatially continuous tumor-concentration fields. Our approach combines a generative model with tumor-infiltration maps that can be propagated through time using a biophysical growth model, enabling fine-grained control over tumor shape and growth while preserving patient anatomy. This enables us to synthesize consistent tumor growth trajectories directly in the space of real patients, providing interpretable, controllable estimation of tumor infiltration and progression beyond what is explicitly observed in imaging. We evaluate the framework on longitudinal glioblastoma cases and demonstrate that it can generate temporally coherent sequences with realistic changes in tumor appearance and surrounding tissue response. These results suggest that integrating mechanistic tumor growth priors with modern generative modeling can provide a practical tool for patient-specific progression visualization and for generating controlled synthetic data to support downstream neuro-oncology workflows. In longitudinal extrapolation, we achieve a consistent 75% Dice overlap with the biophysical model while maintaining a constant PSNR of 25 in the surrounding tissue. Our code is available at: https://github.com/valentin-biller/lgm.git

Related papers

• Mechanistic Learning with Guided Diffusion Models to Predict Spatio-Temporal Brain Tumor Growth [2.1289132027940605]
  We train our model on the BraTS adult and pediatric glioma and evaluate on 60 axial slices of in-house longitudinal pediatric diffuse midline glioma (DMG) cases.<n>Our framework generates realistic follow-up scans based on spatial similarity metrics.<n>It also introduces tumor growth probability maps, which capture both clinically relevant extent and directionality of tumor growth.
  arXiv  Detail & Related papers  (2025-09-11T16:52:09Z)
• Patient-specific prediction of glioblastoma growth via reduced order modeling and neural networks [0.0]
  We present a proof-of-concept for a mathematical model of GBL growth, enabling real-time prediction and patient-specific parameter identification.<n>A neural network surrogate learns the inverse mapping from tumor evolution to model parameters, achieving significant computational speed-up.
  arXiv  Detail & Related papers  (2024-12-04T18:46:05Z)
• Physics-Informed Latent Diffusion for Multimodal Brain MRI Synthesis [43.82741134285203]
  We present a physics-informed generative model capable of synthesizing a variable number of brain MRI modalities. Our approach utilizes latent diffusion models and a two-step generative process. Experiments demonstrate the efficacy of this approach in generating unseen MR contrasts and preserving physical plausibility.
  arXiv  Detail & Related papers  (2024-09-20T14:21:34Z)
• Towards Generalizable Tumor Synthesis [48.45704270448412]
  Tumor synthesis enables the creation of artificial tumors in medical images, facilitating the training of AI models for tumor detection and segmentation. This paper made a progressive stride toward generalizable tumor synthesis by leveraging a critical observation. We have ascertained that generative AI models, e.g., Diffusion Models, can create realistic tumors generalized to a range of organs even when trained on a limited number of tumor examples from only one organ.
  arXiv  Detail & Related papers  (2024-02-29T18:57:39Z)
• Integration of Graph Neural Network and Neural-ODEs for Tumor Dynamic Prediction [4.850774880198265]
  We propose a graph encoder that utilizes a bipartite Graph Convolutional Neural network (GCN) combined with Neural Ordinary Differential Equations (Neural-ODEs) We first show that the methodology is able to discover a tumor dynamic model that significantly improves upon an empirical model. Our findings indicate that the methodology holds significant promise and offers potential applications in pre-clinical settings.
  arXiv  Detail & Related papers  (2023-10-02T06:39:08Z)
• Treatment-aware Diffusion Probabilistic Model for Longitudinal MRI Generation and Diffuse Glioma Growth Prediction [0.5667953366862138]
  We present a novel end-to-end network capable of future predictions of tumor masks and multi-parametric magnetic resonance images (MRI) of how the tumor will look at any future time points for different treatment plans.<n>Our approach is based on cutting-edge diffusion probabilistic models and deep-segmentation neural networks.
  arXiv  Detail & Related papers  (2023-09-11T12:12:52Z)
• Improved Prognostic Prediction of Pancreatic Cancer Using Multi-Phase CT by Integrating Neural Distance and Texture-Aware Transformer [37.55853672333369]
  This paper proposes a novel learnable neural distance that describes the precise relationship between the tumor and vessels in CT images of different patients. The developed risk marker was the strongest predictor of overall survival among preoperative factors.
  arXiv  Detail & Related papers  (2023-08-01T12:46:02Z)
• Generating counterfactual explanations of tumor spatial proteomes to discover effective strategies for enhancing immune infiltration [44.99833362998488]
  The tumor microenvironment (TME) significantly impacts cancer prognosis due to its immune composition. Here, we formulate T-cell infiltration prediction as a self-supervised machine learning problem. We apply our framework to melanoma, colorectal cancer liver metastases, and breast tumor data, discovering perturbations predicted to support T-cell infiltration.
  arXiv  Detail & Related papers  (2022-11-08T05:46:02Z)
• Breast Cancer Induced Bone Osteolysis Prediction Using Temporal Variational Auto-Encoders [65.95959936242993]
  We develop a deep learning framework that can accurately predict and visualize the progression of osteolytic bone lesions. It will assist in planning and evaluating treatment strategies to prevent skeletal related events (SREs) in breast cancer patients.
  arXiv  Detail & Related papers  (2022-03-20T21:00:10Z)
• Learn-Morph-Infer: a new way of solving the inverse problem for brain tumor modeling [1.1214822628210914]
  We introduce a methodology for inferring patient-specific spatial distribution of brain tumor from T1Gd and FLAIR MRI medical scans. Coined as itLearn-Morph-Infer, the method achieves real-time performance in the order of minutes on widely available hardware.
  arXiv  Detail & Related papers  (2021-11-07T13:45:35Z)
• Scale-Space Autoencoders for Unsupervised Anomaly Segmentation in Brain MRI [47.26574993639482]
  We show improved anomaly segmentation performance and the general capability to obtain much more crisp reconstructions of input data at native resolution. The modeling of the laplacian pyramid further enables the delineation and aggregation of lesions at multiple scales.
  arXiv  Detail & Related papers  (2020-06-23T09:20:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.