Brain-WM: Brain Glioblastoma World Model

2026-03-08

AI Summary

This work addresses the challenge of modeling the complex co-evolutionary dynamics between glioblastoma and therapeutic responses under dynamic treatment interventions. To this end, the authors propose a unified world model for brain glioblastoma that jointly predicts subsequent treatment regimens and synthesizes future multimodal MRI scans using a Y-shaped hybrid Transformer architecture. The model integrates autoregressive treatment prediction, flow-based image generation, shared latent space encoding, and a multi-temporal masking alignment mechanism to effectively decouple heterogeneous tasks while ensuring consistent representations of both anatomical structures and tumor progression semantics. Evaluated on a multicenter dataset, the model achieves a treatment planning accuracy of 91.5% and SSIM scores of 0.8524, 0.8581, and 0.8404 for FLAIR, T1CE, and T2W sequences, respectively.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Precise prognostic modeling of glioblastoma (GBM) under varying treatment interventions is essential for optimizing clinical outcomes. While generative AI has shown promise in simulating GBM evolution, existing methods typically treat interventions as static conditional inputs rather than dynamic decision variables. Consequently, they fail to capture the complex, reciprocal interplay between tumor evolution and treatment response. To bridge this gap, we present Brain-WM, a pioneering brain GBM world model that unifies next-step treatment prediction and future MRI generation, thereby capturing the co-evolutionary dynamics between tumor and treatment. Specifically, Brain-WM encodes spatiotemporal dynamics into a shared latent space for joint autoregressive treatment prediction and flow-based future MRI generation. Then, instead of a conventional monolithic framework, Brain-WM adopts a novel Y-shaped Mixture-of-Transformers (MoT) architecture. This design structurally disentangles heterogeneous objectives, successfully leveraging cross-task synergies while preventing feature collapse. Finally, a synergistic multi-timepoint mask alignment objective explicitly anchors latent representations to anatomically grounded tumor structures and progression-aware semantics. Extensive validation on internal and external multi-institutional cohorts demonstrates the superiority of Brain-WM, achieving 91.5% accuracy in treatment planning and SSIMs of 0.8524, 0.8581, and 0.8404 for FLAIR, T1CE, and T2W sequences, respectively. Ultimately, Brain-WM offers a robust clinical sandbox for optimizing patient healthcare. The source code is made available at https://github.com/thibault-wch/Brain-GBM-world-model.

Problem

• glioblastoma
• treatment intervention
• tumor evolution
• prognostic modeling
• dynamic interaction

Innovation

• World Model
• Mixture-of-Transformers
• Dynamic Treatment Modeling
• Co-evolutionary Dynamics
• Latent Space Alignment