MorphSeek: Fine-grained Latent Representation-Level Policy Optimization for Deformable Image Registration Article Swipe

MAKE IMPACT

Zhang, Runxun , Liu, Yizhou , Dongrui Li , Xu Bo , Wei Jingwei · YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2511.17392

Deformable image registration (DIR) remains a fundamental yet challenging problem in medical image analysis, largely due to the prohibitively high-dimensional deformation space of dense displacement fields and the scarcity of voxel-level supervision. Existing reinforcement learning frameworks often project this space into coarse, low-dimensional representations, limiting their ability to capture spatially variant deformations. We propose MorphSeek, a fine-grained representation-level policy optimization paradigm that reformulates DIR as a spatially continuous optimization process in the latent feature space. MorphSeek introduces a stochastic Gaussian policy head atop the encoder to model a distribution over latent features, facilitating efficient exploration and coarse-to-fine refinement. The framework integrates unsupervised warm-up with weakly supervised fine-tuning through Group Relative Policy Optimization, where multi-trajectory sampling stabilizes training and improves label efficiency. Across three 3D registration benchmarks (OASIS brain MRI, LiTS liver CT, and Abdomen MR-CT), MorphSeek achieves consistent Dice improvements over competitive baselines while maintaining high label efficiency with minimal parameter cost and low step-level latency overhead. Beyond optimizer specifics, MorphSeek advances a representation-level policy learning paradigm that achieves spatially coherent and data-efficient deformation optimization, offering a principled, backbone-agnostic, and optimizer-agnostic solution for scalable visual alignment in high-dimensional settings.

Type

preprint

Landing Page

http://arxiv.org/abs/2511.17392

PDF

https://arxiv.org/pdf/2511.17392

OA Status

green

OpenAlex ID

https://openalex.org/W4416598215

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4416598215

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2511.17392

Digital Object Identifier

Title

MorphSeek: Fine-grained Latent Representation-Level Policy Optimization for Deformable Image Registration

Work title

Type

preprint

OpenAlex work type

Publication year

2025

Year of publication

Publication date

2025-11-21

Full publication date if available

Authors

Zhang, Runxun, Liu, Yizhou, Dongrui Li, Xu Bo, Wei Jingwei

List of authors in order

Landing page

https://arxiv.org/abs/2511.17392

Publisher landing page

PDF URL

https://arxiv.org/pdf/2511.17392

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2511.17392

Direct OA link when available

Cited by

0

Total citation count in OpenAlex

Full payload