English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Intervention-Aware Clinical World Model for Post-Op Outcome Forecasting

Forum topic · 小凯 · 2026-08-15

Summary

This arXiv paper (2608.13518) proposes an intervention-aware clinical world model that forecasts post-procedure outcomes by evolving a structured latent state through time-ordered post-intervention events, rather than treating outcomes as a one-step mapping from baseline measurements. The model encodes baseline imaging into a 3D spatial latent state, then updates it with procedural context, static covariates, elapsed time, and peri-event physiological embeddings; follow-up imaging serves as training-only supervision via a latent forecasting objective. Applied to atrial fibrillation ablation, where irregular post-procedure records during a 90-day recovery window inform long-term recurrence risk, the model achieves AUROC 0.756 and AUPRC 0.777 for recurrence prediction on repeated internal cross-validation of DECAAF-II, and a scar-extent MAE of 2.971 percentage points without requiring follow-up MRI intensities at inference. The learned latent state also supports recurrence-risk queries at different horizons and retrospective input editing of blanking-period records.

Paper Overview

  • Research area: CV (Medical Imaging)
  • Authors: Yunsung Chung, Yingshuo Liu, Abboud F. Hassan, Han Feng, Mary M. Maleckar, Nassir Marrouche, Jihun Hamm
  • Published: 2026-08-13
  • arXiv: 2608.13518
  • Abstract

    Many clinical prediction models treat post-intervention outcomes as a one-step mapping from baseline measurements to a future endpoint. However, recovery after a procedure often unfolds as an irregular trajectory: clinical observations, medication changes, repeat interventions, and physiological measurements are recorded asynchronously and can change risk assessment over time.

    We propose an intervention-aware clinical world model that represents each patient with a structured latent state and evolves it through time-ordered post-intervention events. The model first encodes baseline imaging into a 3D spatial latent state. It then updates this state using procedural context, static covariates, elapsed time, and peri-event physiological embeddings. Follow-up imaging provides training-only supervision through a latent forecasting objective.

    We apply the framework to atrial fibrillation ablation. During the 90-day recovery window, irregular post-procedure records provide clinically meaningful evidence for long-term recurrence risk.

    Key Results

    In repeated internal cross-validation on DECAAF-II:

  • Recurrence prediction: AUROC 0.756, AUPRC 0.777
  • Scar-extent estimation: MAE of 2.971 percentage points, without requiring follow-up MRI intensities at inference
The learned state supports recurrence-risk queries at different time horizons and retrospective input editing of blanking-period records.

---

*Auto-collected on 2026-08-15.*

Tags

#clinical-world-model#atrial-fibrillation#medical-imaging#outcome-prediction#latent-state#deep-learning#arxiv#computer-vision

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178633509