# DreamReg: Belief-Driven World Model for 2D-3D Ultrasound Registration

arXiv: https://arxiv.org/abs/2606.18825 · HTML: https://mlsanity.com/paper/2606.18825/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Medical and surgical world models, one of 11 directions in World models.
- Builds on 15 of its 32 references that are in the map.
- First posted on arXiv 17 Jun 2026.

## Placed in

- World models / [Medical and surgical world models](https://mlsanity.com/wm/medical-and-surgical-world-models/)

## Builds on (most cited first, 15 of 15)

- [Dream to Control: Learning Behaviors by Latent Imagination](https://mlsanity.com/paper/1912.01603/) (2019)
- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Mastering Atari with Discrete World Models](https://mlsanity.com/paper/2010.02193/) (2020)
- [DayDreamer: World Models for Physical Robot Learning](https://mlsanity.com/paper/2206.14176/) (2022)
- [Deep Learning for Segmentation using an Open Large-Scale Dataset in 2D Echocardiography](https://mlsanity.com/paper/1908.06948/) (2019)
- [Recurrent Memory Transformer](https://mlsanity.com/paper/2207.06881/) (2022)
- [GWM: Towards Scalable Gaussian World Models for Robotic Manipulation](https://mlsanity.com/paper/2508.17600/) (2025)
- [Medical World Model: Generative Simulation of Tumor Evolution for Treatment Planning](https://mlsanity.com/paper/2506.02327/) (2025)
- [Slice-to-volume medical image registration: a survey](https://mlsanity.com/paper/1702.01636/) (2017)
- [EHRWorld: A Patient-Centric Medical World Model for Long-Horizon Clinical Trajectories](https://mlsanity.com/paper/2602.03569/) (2026)
- [End-to-end Ultrasound Frame to Volume Registration](https://mlsanity.com/paper/2107.06449/) (2021)
- [Global Multi-modal 2D/3D Registration via Local Descriptors Learning](https://mlsanity.com/paper/2205.03439/) (2022)
- [Epicardium Prompt-guided Real-time Cardiac Ultrasound Frame-to-volume Registration](https://mlsanity.com/paper/2406.14534/) (2024)
- [Rigid Single-Slice-in-Volume registration via rotation-equivariant 2D/3D feature matching](https://mlsanity.com/paper/2410.18683/) (2024)
- [Identifying visible tissue in intraoperative ultrasound: a method and application](https://mlsanity.com/paper/2306.01190/) (2023)

## Built on by (most cited first, 4 of 4)

- [Medical world models in healthcare: foundations, applications, and challenges for trustworthy clinical translation](https://mlsanity.com/paper/2607.25242/) (2026)
- [AuricularWorld: Hierarchical Action-Guided World Modeling for Fine-Grained Auricular Structure Segmentation from CT Scans](https://mlsanity.com/paper/2607.28487/) (2026)
- [SCoPE-Reg: Efficient Rigid Ultrasound Slice-to-Volume Registration via State-Space Correlation and Closed-Form Pose Estimation](https://mlsanity.com/paper/2608.28715/) (2026)
- [Scanning While Imagining: A Scene-Graph World Model for Robotic Ultrasound Navigation](https://mlsanity.com/paper/2609.32837/) (2026)

## Abstract (arXiv)

Ultrasound (US) is widely used for surgical navigation, yet real-time registration between intraoperative 2D slices and preoperative 3D volumes remains challenging due to partial observability, speckle noise, and the action-dependent US acquisition. Existing methods are one-shot or short-horizon, making it hard for them to gather evidence over time or capture how surgeons adjust probe motion based on on-screen feedback. We propose DreamReg, a belief-driven world-model framework that formulates 2D-3D registration as belief updating over rigid transformations. DreamReg maintains a latent belief state that summarizes past observations and poses information, and continuously refines the transformation through learned dynamics as new slices arrive. During training, DreamReg is exposed to probe-motion trajectories that mimic clinical scanning behavior and learns to update its belief by conditioning pose refinement on the current US observation. During inference, DreamReg refines registration via internal imagination: it rolls out the learned world model to simulate candidate probe motions and their predicted observations, and integrates these imagined outcomes to converge to an accurate rigid transformation. Experiments on CAMUS and u-RegPro datasets demonstrate improved robustness and competitive registration accuracy for real-time guidance compared with state-of-the-art methods.