From design through to clinical validation of a soft end-effector for robot-assisted ultrasound imaging

Abstract

Sonographers are exposed to an increased risk of work-related musculoskeletal disorders due to performing diagnostic ultrasound procedures. Existing research therefore aims at developing robotic ultrasound solutions which could be integrated into clinics to improve workplace ergonomics.

This dissertation addresses limitations in state-of-the-art ultrasound robots by designing and clinically validating a soft robotic system for ultrasound imaging. It employs the benefits of soft robotics technologies to establish a safe interaction between the ultrasound probe and patient. Specifications for the robot are established from clinical data obtained from prenatal foetal ultrasound scans. The 3DOF parallel, soft robotic end-effector (SEE) is characterised and its imaging capabilities validated by a trained sonographer on a foetal ultrasound phantom. A kinetostatic model capable of accurately estimating the pose of the end-effector is derived and validated. Further, the contact force on the ultrasound probe is estimated according to internal pressure variations in the fluidic actuators. The model is applied to the SEE in real-time. To regulate contact pressure during the imaging procedure, a force controller based solely on the fluidic pressure of the actuators is proposed. The performance of the controller is validated for both static and dynamic contacts.

The system is evaluated in a clinical volunteer study with healthy participants. Standard views are obtained for abdominal ultrasound scans in which participants are either still or exhibiting pronounced respiratory motion. It is shown that the system is able to acquire high-quality images and eliminate forces on the sonographer without causing discomfort to the participant. The SEE could therefore provide a cost-effective means to address the prevalent risk of musculoskeletal disorders in sonographers.

Date of Award	1 Apr 2020
Original language	English
Awarding Institution	King's College London
Supervisor	Hongbin Liu (Supervisor) & Kawal Rhode (Supervisor)