{"slug": "camewm-a-causal-medical-world-model-for-generative-tumor-evolution-and-treatment", "title": "CaMeWM: A Causal Medical World Model for Generative Tumor Evolution and Treatment Planning", "summary": "Researchers introduced CaMeWM, a causal medical world model for simulating post-TACE tumor evolution and supporting treatment planning, according to an arXiv paper (2610.26923v1). Across 79 paired out-of-fold episodes, CaMeWM reached 0.618 SSIM and 0.528 change-map SSIM for follow-up CT transition and a 0.615 C-index on its generated-world survival pathway, with factual-only AIPW estimating a 9.60-week restricted mean survival time gain (95% CI 0.61–18.13) for the frozen policy. The model admits a queried regimen only for clinically comparable histories with empirical treatment support, then generates lesion-local follow-up worlds via an action-conditioned response descriptor, deterministic anchor and residual 3D diffusion transition before mapping them to RMST with a response-Cox utility model.", "body_md": "arXiv:2610.26923v1 Announce Type: new \nAbstract: Treatment planning is intrinsically counterfactual: clinicians must compare how a patient's tumor and survival would evolve under competing regimens, whereas a retrospective cohort records only the regimen received and one follow-up scan. A treatment-conditioned generator trained on such data can therefore reproduce historical treatment allocation rather than simulate the interventional disease dynamics needed to rank regimens. We introduce \\textbf{CaMeWM}, a causal medical world model for supported post-TACE tumor evolution and treatment planning. CaMeWM implements one intervention-to-decision loop rather than three independent prediction tasks: it first admits a queried regimen only for clinically comparable histories with empirical treatment support; it then generates a distribution of lesion-local follow-up worlds through an action-conditioned response descriptor, deterministic anchor, and residual 3D diffusion transition; and it finally maps sampled worlds to restricted mean survival time (RMST) with a response-Cox utility model. Under pre-treatment adjustment and positivity, this loop estimates a prespecified, support-restricted g-formula target; all selection and nuisance estimation are fold-local, and the induced policy is evaluated by cross-fitted, censoring-adjusted doubly robust estimation. Across 79 paired out-of-fold episodes, CaMeWM achieves 0.618 SSIM and 0.528 change-map SSIM for follow-up CT transition, while its generated-world survival pathway attains a 0.615 C-index; factual-only AIPW estimates a 9.60-week RMST gain (95\\% CI 0.61--18.13) for the frozen policy. CaMeWM provides an inspectable basis for retrospective, support-restricted treatment planning.", "url": "https://wpnews.pro/news/camewm-a-causal-medical-world-model-for-generative-tumor-evolution-and-treatment", "canonical_source": "https://arxiv.org/abs/2610.26923", "published_at": "2026-10-21 04:00:00+00:00", "updated_at": "2026-10-21 04:22:15.995436+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "generative-ai", "ai-research"], "entities": ["CaMeWM", "arXiv", "TACE"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/camewm-a-causal-medical-world-model-for-generative-tumor-evolution-and-treatment", "markdown": "https://wpnews.pro/news/camewm-a-causal-medical-world-model-for-generative-tumor-evolution-and-treatment.md", "text": "https://wpnews.pro/news/camewm-a-causal-medical-world-model-for-generative-tumor-evolution-and-treatment.txt", "jsonld": "https://wpnews.pro/news/camewm-a-causal-medical-world-model-for-generative-tumor-evolution-and-treatment.jsonld"}}