English

Ultrasound-Guided Robotic Navigation with Deep Reinforcement Learning

Machine Learning 2020-04-08 v2 Robotics Machine Learning

Abstract

In this paper we introduce the first reinforcement learning (RL) based robotic navigation method which utilizes ultrasound (US) images as an input. Our approach combines state-of-the-art RL techniques, specifically deep Q-networks (DQN) with memory buffers and a binary classifier for deciding when to terminate the task. Our method is trained and evaluated on an in-house collected data-set of 34 volunteers and when compared to pure RL and supervised learning (SL) techniques, it performs substantially better, which highlights the suitability of RL navigation for US-guided procedures. When testing our proposed model, we obtained a 82.91% chance of navigating correctly to the sacrum from 165 different starting positions on 5 different unseen simulated environments.

Keywords

Cite

@article{arxiv.2003.13321,
  title  = {Ultrasound-Guided Robotic Navigation with Deep Reinforcement Learning},
  author = {Hannes Hase and Mohammad Farid Azampour and Maria Tirindelli and Magdalini Paschali and Walter Simson and Emad Fatemizadeh and Nassir Navab},
  journal= {arXiv preprint arXiv:2003.13321},
  year   = {2020}
}

Comments

Submitted for IROS 2020

R2 v1 2026-06-23T14:31:36.677Z