English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Ultrasound (US) is one of the most common medical imaging modalities since itis radiation-free, low-cost, and real-time. In freehand US examinations,sonographers often navigate a US probe to visualize standard examination planeswith rich diagnostic information. However, reproducibility and stability of theresulting images often suffer from intra- and inter-operator variation.Reinforcement learning (RL), as an interaction-based learning method, hasdemonstrated its effectiveness in visual navigating tasks; however, RL islimited in terms of generalization. To address this challenge, we propose asimulation-based RL framework for real-world navigation of US probes towardsthe standard longitudinal views of vessels. A UNet is used to provide binarymasks from US images; thereby, the RL agent trained on simulated binary vesselimages can be applied in real scenarios without further training. To accuratelycharacterize actual states, a multi-modality state representation structure isintroduced to facilitate the understanding of environments. Moreover,considering the characteristics of vessels, a novel standard view recognitionapproach based on the minimum bounding rectangle is proposed to terminate thesearching process. To evaluate the effectiveness of the proposed method, thetrained policy is validated virtually on 3D volumes of a volunteer's in-vivocarotid artery, and physically on custom-designed gel phantoms using roboticUS. The results demonstrate that proposed approach can effectively andaccurately navigate the probe towards the longitudinal view of vessels.

VesNet-RL: Simulation-based Reinforcement Learning for Real-World US Probe Navigation