Longitudinal NSCLC Treatment Progression via Multimodal Generative Models

2026-03-06

AI Summary

This study addresses the prediction of longitudinal tumor evolution in non-small cell lung cancer (NSCLC) during radiotherapy, driven jointly by anatomical changes and delivered radiation dose. To this end, the authors propose a virtual treatment (VT) framework that, for the first time, integrates incremental radiation dose with multimodal clinical variables into a conditional generative model. Tumor evolution is formulated as a dose-aware, multimodal image-to-image translation task capable of synthesizing follow-up CT images at arbitrary time points. The approach is implemented using both GANs and diffusion models and evaluated in 2D and 2.5D configurations. Experiments on a dataset comprising 222 patients and 895 CT scans demonstrate that the diffusion-based model generates more stable and anatomically plausible evolutionary trajectories, significantly outperforming GAN baselines and offering a promising tool for personalized adaptive radiotherapy.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Multimodal Learning
• Natural Language Processing: Generation

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Predicting tumor evolution during radiotherapy is a clinically critical challenge, particularly when longitudinal changes are driven by both anatomy and treatment. In this work, we introduce a Virtual Treatment (VT) framework that formulates non-small cell lung cancer (NSCLC) progression as a dose-aware multimodal conditional image-to-image translation problem. Given a CT scan, baseline clinical variables, and a specified radiation dose increment, VT aims to synthesize plausible follow-up CT images reflecting treatment-induced anatomical changes. We evaluate the proposed framework on a longitudinal dataset of 222 stage III NSCLC patients, comprising 895 CT scans acquired during radiotherapy under irregular clinical schedules. The generative process is conditioned on delivered dose increments together with demographic and tumor-related clinical variables. Representative GAN-based and diffusion-based models are benchmarked across 2D and 2.5D configurations. Quantitative and qualitative results indicate that diffusion-based models benefit more consistently from multimodal, dose-aware conditioning and produce more stable and anatomically plausible tumor evolution trajectories than GAN-based baselines, supporting the potential of VT as a tool for in-silico treatment monitoring and adaptive radiotherapy research in NSCLC.

Problem

• tumor evolution
• radiotherapy
• longitudinal imaging
• non-small cell lung cancer
• treatment-induced changes

Innovation

• dose-aware generative modeling
• multimodal conditional synthesis
• longitudinal tumor progression
• diffusion-based image translation
• virtual treatment simulation