analiza aiqbench
Echo-POSED introduces a self-supervised framework that derives probe adjustment guidance directly from 2D ultrasound images, eliminating the dependency on expert-labeled views or tracked trajectories. By enforcing geometric equivariance to probe motions while remaining invariant to cardiac phase, the model achieves a mean angular error of 8.2 degrees in intra-patient simulations with cardiac motion. This approach offers a practical pathway for real-time transthoracic echocardiography guidance using routinely acquired 3D volumes.
Echo-POSED: Geometric Self-Distillation for Echocardiography Guidance
arXiv:2606.02634v1 Announce Type: cross Abstract: We introduce Echo-POSED, a self-supervised framework for real-time transthoracic echocardiography (TTE) guidance that recommends probe adjustments directly from 2D ultrasound images, without the need for expert-labelled views or tracked probe trajectories. Instead, it trains on 2D views sliced from routinely acquired 3D echocardiography volumes, enforcing equivariance to probe motions while remaining invariant to cardiac phase, yielding a pose representation on $\mathrm{SO}(3)\times\mathrm{SO}(3)$. Across a held-out split and public external 3D--TTE datasets (including vendor shift), Echo-POSED maintains geometric consistency under virtual perturbations and enables intra- and inter-patient guidance simulations, achieving a combined mean angular error of 8.2 degrees between the guided and target views in intra-patient simulations with cardiac motion.
Abstract:We introduce Echo-POSED, a self-supervised framework for real-time transthoracic echocardiography (TTE) guidance that recommends probe adjustments directly from 2D ultrasound images, without the need for expert-labelled views or tracked probe trajectories. Instead, it trains on 2D views sliced from routinely acquired 3D echocardiography volumes, enforcing equivariance to probe motions while remaining invariant to cardiac phase, yielding a pose representation on $\mathrm{SO}(3)\times\mathrm{SO}(3)$. Across a held-out split and public external 3D--TTE datasets (including vendor shift), Echo-POSED maintains geometric consistency under virtual perturbations and enables intra- and inter-patient guidance simulations, achieving a combined mean angular error of 8.2 degrees between the guided and target views in intra-patient simulations with cardiac motion.
Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI)
arXiv:2606.02634 [eess.IV]
(or arXiv:2606.02634v1 [eess.IV] for this version)
https://doi.org/10.48550/arXiv.2606.02634 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 207 cuvinte
Subiecte conexe
$\beta$-OPSD: Deriving with Policy Optimization, Training with Self-Distillation
arXiv · 31 iul. 2026
UltraWorld: Learning Interactive Ultrasound World Models from Untracked Clinical Videos with Acoustic Sampling Map
arXiv · 8 oct. 2026
PBSD: Privileged Bayesian Self-Distillation for Long-Horizon Credit Assignment
arXiv · 9 iun. 2026
CVSD-Reg: Cross-Modal Visual Semantic Prior Distillation for Robust LiDAR Registration
arXiv · 21 aug. 2026