TumorFlow: Physics-Guided Longitudinal MRI Synthesis of Glioblastoma Growth

2026-03-04

AI Summary

This study addresses the challenge of highly individualized glioblastoma growth and its incomplete visualization in conventional MRI, which hinders precise treatment planning. To overcome this limitation, the work proposes a novel framework that integrates a mechanistic biophysical tumor growth model with 3D generative AI to synthesize biologically plausible and temporally consistent longitudinal MRI sequences within real patient-specific brain anatomy. By leveraging spatially continuous tumor concentration fields, the method enables fine-grained control over tumor morphology and infiltration dynamics while ensuring consistency between generated images and underlying tumor kinetics. Experimental results demonstrate that extrapolated tumor regions achieve a Dice overlap of 75% with the biophysical model, and surrounding tissue exhibits stable image fidelity with a PSNR of approximately 25, reflecting realistic temporal evolution and tissue response characteristics.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Natural Language Processing: Generation

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

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

Problem

• glioblastoma
• tumor growth
• MRI synthesis
• tumor infiltration
• longitudinal imaging

Innovation

• physics-guided generative modeling
• glioblastoma growth simulation
• longitudinal MRI synthesis
• biophysical tumor modeling
• tumor infiltration mapping