Metraipolis › archive › dc_1149e4ef3dde

From Image Interpretation to Clinical Reasoning: Upstream Physician-Context-Aware Multimodal Learning with Causal Reinforcement Learning — arXiv:2609.38924

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.38924
License (as archived)
CC0-1.0
Archived
2026-10-07T14:45:38.653Z
SHA-256 of the archived text
3642a3215a63069490b35b788d856088c92e372f8b452081effdebb01c5ca0c4
Stable URL of this page
https://metraipolis.com/documento/dc_1149e4ef3dde

How to cite

From Image Interpretation to Clinical Reasoning: Upstream Physician-Context-Aware Multimodal Learning with Causal Reinforcement Learning — arXiv:2609.38924 — arXiv (en), https://arxiv.org/abs/2609.38924 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1149e4ef3dde (Metraipolis archive dc_1149e4ef3dde, sha256 3642a3215a63069490b35b788d856088c92e372f8b452081effdebb01c5ca0c4)
BibTeX
@misc{dc1149e4ef3dde,
  title = {From Image Interpretation to Clinical Reasoning: Upstream Physician-Context-Aware Multimodal Learning with Causal Reinforcement Learning — arXiv:2609.38924},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.38924},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1149e4ef3dde; sha256 3642a3215a63069490b35b788d856088c92e372f8b452081effdebb01c5ca0c4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38924. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Major adverse cardiovascular events (MACE) remain the leading cause of mortality worldwide. Opportunistic screening using routinely acquired clinical data offers a scalable approach for identifying high-risk individuals before acute events occur. Although chest X-rays (CXRs) capture latent cardiovascular biomarkers and clinical histories provide complementary patient context, existing medical vision-language models are primarily optimized for radiology interpretation rather than prognostic reasoning. We propose a causal reinforcement learning framework for multimodal clinical reasoning that integrates CXRs and physician-authored clinical histories for opportunistic MACE prediction. The framework introduces (1) a role-decoupled dual-LLM architecture that separates reasoning from risk prediction, (2) a dual-action causal reinforcement learning policy for evidence selection and reasoning optimization, and (3) causal token pruning to learn compact multimodal representations. Evaluated on an internal cohort, an emergency department cohort, and the external MIMIC dataset, the proposed framework consistently outperformed unimodal baselines and state-of-the-art medical vision-language models, achieving AUROCs of 0.720, 0.760, and 0.845, respectively. It also substantially improved reasoning quality, achieving higher GREEN scores and higher expert preference while maintaining robust predictive performance across diverse patient populations.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.38924 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T14:19:00.637Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).