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

Ultrasound (US) is one of the most common medical imaging modalities since it is radiation-free, low-cost, and real-time. In freehand US examinations, sonographers often navigate a US probe to visualize standard examination planes with rich diagnostic information. However, reproducibility and stability of the resulting images often suffer from intra- and inter-operator variation. Reinforcement learning (RL), as an interaction-based learning method, has demonstrated its effectiveness in visual navigating tasks; however, RL is limited in terms of generalization. To address this challenge, we propose a simulation-based RL framework for real-world navigation of US probes towards the standard longitudinal views of vessels. A UNet is used to provide binary masks from US images; thereby, the RL agent trained on simulated binary vessel images can be applied in real scenarios without further training. To accurately characterize actual states, a multi-modality state representation structure is introduced to facilitate the understanding of environments. Moreover, considering the characteristics of vessels, a novel standard view recognition approach based on the minimum bounding rectangle is proposed to terminate the searching process. To evaluate the effectiveness of the proposed method, the trained policy is validated virtually on 3D volumes of a volunteer’s in-vivo carotid artery, and physically on custom-designed gel phantoms using robotic US. The results demonstrate that proposed approach can effectively and accurately navigate the probe towards the longitudinal view of vessels.

I. Introduction

In The field of medical imaging, ultrasound (US) is one of the most popular diagnostic tools for medical examinations of internal organs. Compared to computed tomography (CT) and magnetic resonance imaging (MRI) examinations, US is real-time, low cost and radiation free [1]. For vascular medicine, in particular, US plays a critical role in everyday practice, namely, for the diagnostics, image-guided interventions and therapy assessment of diseases. In carotid ultrasonography, the optimal acquisition of the longitudinal view of the carotid artery (see Fig. 1) is required for evaluation of the intima-medial thickness (IMT) [2], the plaque morphology [3], or the peak systolic velocity of the blood over plaques [4]. As for real-time US-guided femoral arterial access, longitudinal views of the target vessel provide a clear visualization of the needle path and the real-time guidance of the guidewire in the vessel of interest [5].