PhySe-RPO: Physics and Semantics Guided Relative Policy Optimization for Diffusion-Based Surgical Smoke Removal

• URL: http://arxiv.org/abs/2603.22844v2
• Date: Wed, 25 Mar 2026 11:20:24 GMT
• Title: PhySe-RPO: Physics and Semantics Guided Relative Policy Optimization for Diffusion-Based Surgical Smoke Removal
• Abstract summary: Surgical smoke severely degrades intraoperative video quality, obscuring anatomical structures and limiting surgical perception.<n>We propose PhySe-RPO, a diffusion restoration framework optimized through Physics- and Semantics-Guided Relative Policy Optimization.<n>A physics-guided reward imposes illumination and color consistency, while a visual-concept semantic reward learned from CLIP-based surgical concepts promotes smoke-free and anatomically coherent restoration.
• Score: 11.399795585157863
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Surgical smoke severely degrades intraoperative video quality, obscuring anatomical structures and limiting surgical perception. Existing learning-based desmoking approaches rely on scarce paired supervision and deterministic restoration pipelines, making it difficult to perform exploration or reinforcement-driven refinement under real surgical conditions. We propose PhySe-RPO, a diffusion restoration framework optimized through Physics- and Semantics-Guided Relative Policy Optimization. The core idea is to transform deterministic restoration into a stochastic policy, enabling trajectory-level exploration and critic-free updates via group-relative optimization. A physics-guided reward imposes illumination and color consistency, while a visual-concept semantic reward learned from CLIP-based surgical concepts promotes smoke-free and anatomically coherent restoration. Together with a reference-free perceptual constraint, PhySe-RPO produces results that are physically consistent, semantically faithful, and clinically interpretable across synthetic and real robotic surgical datasets, providing a principled route to robust diffusion-based restoration under limited paired supervision.

Related papers

• IADD-TR: Intervention-Aware Dynamics Decoupling with Targeted Regularization for Model-Based Reinforcement Learning [57.16513298507272]
  We propose IADD-TR, a unified framework combining Intervention-Aware Dynamics Decoupling (IADD) and Targeted Regularization (TR)<n>IADD factorizes transitions into an action-intervention stage and an action-free natural evolution stage, using a zero-action anchor to resolve the non-uniqueness of this two-stage factorization for robust generalization.<n>For policy learning, we derive TR from the efficient influence function of a replay-state policy-gradient functional.
  arXiv  Detail & Related papers  (2026-08-11T08:20:02Z)
• Next-Acceleration-Scale Prediction for Autoregressive MRI Reconstruction [52.32112533846212]
  MRI reconstruction is an inherently ill-posed inverse problem, since incomplete measurements admit many plausible solutions.<n>We address this limitation by moving reconstruction to discrete multi-scale latent space and posing it as autoregressive next-acceleration-scale prediction.<n>We show that our approach delivers improved reconstruction performance across diverse sampling patterns under extreme undersampling.
  arXiv  Detail & Related papers  (2026-05-19T04:40:50Z)
• Regulating Anatomy-Aware Rewards via Trajectory-Integral Feedback for Volumetric Computed Tomography Analysis [30.439440028190944]
  We introduce textbfClinical Abnormality Benchmarking Substrate (CABS), a structured system that decomposes radiology reports into verifiable clinical semantic units.<n>We propose textbfTrajectory-Integral FeedbackO (TIF-GRPO), a novel framework integrating control-theoretic principles into policy optimization.
  arXiv  Detail & Related papers  (2026-05-19T04:33:27Z)
• Unveiling the Backdoor Mechanism Hidden Behind Catastrophic Overfitting in Fast Adversarial Training [66.02119132131321]
  Fast Adversarial Training (FAT) has attracted significant attention due to its efficiency in enhancing neural network robustness against adversarial attacks.<n>FAT is prone to catastrophic overfitting (CO), wherein models overfit to the specific attack used during training and fail to generalize to others.<n>We propose a systematic and intuitive explanation of CO through the lens of backdoor.
  arXiv  Detail & Related papers  (2026-04-27T11:44:42Z)
• Are Natural-Domain Foundation Models Effective for Accelerated Cardiac MRI Reconstruction? [9.952061326734741]
  We investigate whether natural-domain foundation models can serve as effective image priors for accelerated cardiac MRI reconstruction.<n>We show that while task-specific state-of-the-art reconstruction models such as E2E-VarNet achieve superior performance in standard in-distribution settings, foundation-model-based approaches remain competitive.
  arXiv  Detail & Related papers  (2026-04-24T13:52:25Z)
• Investigating a Policy-Based Formulation for Endoscopic Camera Pose Recovery [13.321458523588626]
  Vision-based navigation systems seek to replicate endoscopic surgery by recovering camera pose directly from endoscopic video.<n>Most approaches do not embody the same principles of reasoning about new frames that makes surgeons successful.<n>We propose a policy-based formulation of endoscopic camera pose recovery that seeks to imitate experts in estimating trajectories conditioned on the previous camera state.
  arXiv  Detail & Related papers  (2026-03-20T15:30:59Z)
• Percept-Aware Surgical Planning for Visual Cortical Prostheses with Vascular Avoidance [6.649753747542209]
  Cortical visual prostheses aim to restore sight by electrically stimulating neurons in early visual cortex (V1)<n>Existing strategies emphasize visual field coverage and anatomicals but do not directly optimize predicted perceptual outcomes under safety constraints.<n>We present a percept-aware framework for surgical planning of cortical visual prostheses that formulates electrode placement as a constrained optimization problem in anatomical space.
  arXiv  Detail & Related papers  (2026-02-27T22:59:18Z)
• Bidirectional Reward-Guided Diffusion for Real-World Image Super-Resolution [79.35296000454694]
  Diffusion-based super-resolution can synthesize rich details, but models trained on synthetic paired data often fail on real-world LR images.<n>We propose Bird-SR, a reward-guided diffusion framework that formulates super-resolution as trajectory-level preference optimization.<n>Experiments on real-world SR benchmarks demonstrate that Bird-SR consistently outperforms state-of-the-art methods in perceptual quality.
  arXiv  Detail & Related papers  (2026-02-05T19:21:45Z)
• A Koopman-Bayesian Framework for High-Fidelity, Perceptually Optimized Haptic Surgical Simulation [0.0]
  We introduce a unified framework that combines nonlinear dynamics, perceptual psychophysics and high frequency haptic rendering to enhance realism in surgical simulation.<n>For various simulated surgical tasks such as palpation, incision, and bone milling, the proposed system attains an average rendering latency of 4.3 ms, a force error of less than 2.8% and a 20% improvement in perceptual discrimination.
  arXiv  Detail & Related papers  (2026-01-04T20:43:37Z)
• Moving Beyond Diffusion: Hierarchy-to-Hierarchy Autoregression for fMRI-to-Image Reconstruction [65.67001243986981]
  We propose MindHier, a coarse-to-fine fMRI-to-image reconstruction framework built on scale-wise autoregressive modeling.<n>MindHier achieves superior semantic fidelity, 4.67x faster inference, and more deterministic results than the diffusion-based baselines.
  arXiv  Detail & Related papers  (2025-10-25T15:40:07Z)
• EndoUFM: Utilizing Foundation Models for Monocular depth estimation of endoscopic images [7.350425834778092]
  EndoUFM is an unsupervised monocular depth estimation framework.<n>It enhances the depth estimation performance by leveraging the powerful pre-learned priors.<n>This work contributes to augmenting surgeons' spatial perception during minimally invasive procedures.
  arXiv  Detail & Related papers  (2025-08-25T11:33:05Z)
• Benchmarking Laparoscopic Surgical Image Restoration and Beyond [54.28852320829451]
  In laparoscopic surgery, a clear and high-quality visual field is critical for surgeons to make accurate decisions.<n> persistent visual degradation, including smoke generated by energy devices, lens fogging from thermal gradients, and lens contamination pose risks to patient safety.<n>We introduce a real-world open-source surgical image restoration dataset covering laparoscopic environments, called SurgClean.
  arXiv  Detail & Related papers  (2025-05-25T14:17:56Z)
• RSFR: A Coarse-to-Fine Reconstruction Framework for Diffusion Tensor Cardiac MRI with Semantic-Aware Refinement [10.620138922340258]
  We introduce RSFR (Reconstruction, Fusion & Refinement), a novel framework for cardiac diffusion-weighted image reconstruction.<n> RSFR employs a coarse-to-fine strategy, leveraging zero-shot semantic priors via the Segment Anything Model and a robust Vision Mamba-based reconstruction backbone.<n>Our framework integrates semantic features effectively to mitigate artefacts and enhance fidelity, achieving state-of-the-art reconstruction quality and accurate DT parameter estimation under high under rates.
  arXiv  Detail & Related papers  (2025-04-25T17:41:14Z)
• Safe Deep RL for Intraoperative Planning of Pedicle Screw Placement [61.28459114068828]
  We propose an intraoperative planning approach for robotic spine surgery that leverages real-time observation for drill path planning based on Safe Deep Reinforcement Learning (DRL) Our approach was capable of achieving 90% bone penetration with respect to the gold standard (GS) drill planning.
  arXiv  Detail & Related papers  (2023-05-09T11:42:53Z)
• Real-time landmark detection for precise endoscopic submucosal dissection via shape-aware relation network [51.44506007844284]
  We propose a shape-aware relation network for accurate and real-time landmark detection in endoscopic submucosal dissection surgery. We first devise an algorithm to automatically generate relation keypoint heatmaps, which intuitively represent the prior knowledge of spatial relations among landmarks. We then develop two complementary regularization schemes to progressively incorporate the prior knowledge into the training process.
  arXiv  Detail & Related papers  (2021-11-08T07:57:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.