Exploring Strategies for Personalized Radiation Therapy Part II Predicting Tumor Drift Patterns with Diffusion Models

2025-06-20

AI Summary

Conventional radiotherapy for brain cancer struggles with spatially and temporally heterogeneous tumor responses, leading to overtreatment or undertreatment. Method: We propose a diffusion-model-based framework for personalized tumor drift prediction to enable early dynamic decision-making in personalized ultra-fractionated stereotactic adaptive radiotherapy (PULSAR). For the first time in radiotherapy response modeling, we introduce Denoising Diffusion Implicit Models (DDIM), implementing both single-step and iterative denoising pathways for end-to-end prediction of longitudinal MRI evolution from baseline to follow-up. Integrated with deformable image registration and dosiomics analysis, our method precisely identifies treatment-response–associated regions. Contribution/Results: The framework significantly improves biologically conformal dose delivery and temporal sensitivity while avoiding overtreatment and undertreatment. It overcomes the limitations of conventional radiomics in modeling temporal dynamics, establishing a novel, interpretable, and generalizable spatiotemporal prediction paradigm for adaptive radiotherapy.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Multimodal Learning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Responsible Web: Consent frameworks and practices on the web

Abstract

Radiation therapy outcomes are decided by two key parameters, dose and timing, whose best values vary substantially across patients. This variability is especially critical in the treatment of brain cancer, where fractionated or staged stereotactic radiosurgery improves safety compared to single fraction approaches, but complicates the ability to predict treatment response. To address this challenge, we employ Personalized Ultra-fractionated Stereotactic Adaptive Radiotherapy (PULSAR), a strategy that dynamically adjusts treatment based on how each tumor evolves over time. However, the success of PULSAR and other adaptive approaches depends on predictive tools that can guide early treatment decisions and avoid both overtreatment and undertreatment. However, current radiomics and dosiomics models offer limited insight into the evolving spatial and temporal patterns of tumor response. To overcome these limitations, we propose a novel framework using Denoising Diffusion Implicit Models (DDIM), which learns data-driven mappings from pre to post treatment imaging. In this study, we developed single step and iterative denoising strategies and compared their performance. The results show that diffusion models can effectively simulate patient specific tumor evolution and localize regions associated with treatment response. The proposed strategy provides a promising foundation for modeling heterogeneous treatment response and enabling early, adaptive interventions, paving the way toward more personalized and biologically informed radiotherapy.

Problem

• Predict tumor drift patterns for personalized radiation therapy
• Overcome limitations of current radiomics and dosiomics models
• Simulate patient-specific tumor evolution using diffusion models

Innovation

• Uses DDIM for tumor evolution prediction
• Implements single and iterative denoising strategies
• Adapts PULSAR for dynamic treatment adjustment