MedVSR: Medical Video Super-Resolution with Cross State-Space Propagation

2025-09-25

AI Summary

Medical video super-resolution (VSR) faces unique clinical challenges: severe degradation from hardware limitations, camera jitter and inter-frame discontinuities induced by physiological motion, and subtle yet anatomically critical tissue structures that are easily distorted or corrupted by artifacts in existing models. To address these, we propose MedVSR—a novel framework featuring Cross-State Space Propagation (CSSP), which leverages distant frames as control matrices to optimize long-range temporal feature alignment; and Intrinsic State Space Reconstruction (ISSR), a module jointly modeling long-range spatial dependencies and large-kernel local aggregation to faithfully recover fine tissue details while suppressing diagnostically disruptive artifacts. Evaluated on four real-world medical video datasets—including endoscopic and cataract surgery videos—MedVSR achieves state-of-the-art performance with superior PSNR/SSIM and reduced computational cost. The code is publicly available.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures

Abstract

High-resolution (HR) medical videos are vital for accurate diagnosis, yet are hard to acquire due to hardware limitations and physiological constraints. Clinically, the collected low-resolution (LR) medical videos present unique challenges for video super-resolution (VSR) models, including camera shake, noise, and abrupt frame transitions, which result in significant optical flow errors and alignment difficulties. Additionally, tissues and organs exhibit continuous and nuanced structures, but current VSR models are prone to introducing artifacts and distorted features that can mislead doctors. To this end, we propose MedVSR, a tailored framework for medical VSR. It first employs Cross State-Space Propagation (CSSP) to address the imprecise alignment by projecting distant frames as control matrices within state-space models, enabling the selective propagation of consistent and informative features to neighboring frames for effective alignment. Moreover, we design an Inner State-Space Reconstruction (ISSR) module that enhances tissue structures and reduces artifacts with joint long-range spatial feature learning and large-kernel short-range information aggregation. Experiments across four datasets in diverse medical scenarios, including endoscopy and cataract surgeries, show that MedVSR significantly outperforms existing VSR models in reconstruction performance and efficiency. Code released at https://github.com/CUHK-AIM-Group/MedVSR.

Problem

• Addresses alignment errors from camera shake in medical videos
• Reduces artifacts and distortions in tissue structure reconstruction
• Enhances super-resolution for continuous organ movements in endoscopy

Innovation

• Cross State-Space Propagation for frame alignment
• Inner State-Space Reconstruction to reduce artifacts
• Joint long-range and short-range feature aggregation