Author:
Patel Niravkumar,Yan Jiawen,Li Gang,Monfaredi Reza,Priba Lukasz,Donald-Simpson Helen,Joy Joyce,Dennison Andrew,Melzer Andreas,Sharma Karun,Iordachita Iulian,Cleary Kevin
Abstract
This paper presents an intraoperative MRI-guided, patient-mounted robotic system for shoulder arthrography procedures in pediatric patients. The robot is designed to be compact and lightweight and is constructed with nonmagnetic materials for MRI safety. Our goal is to transform the current two-step arthrography procedure (CT/x-ray-guided needle insertion followed by diagnostic MRI) into a streamlined single-step ionizing radiation-free procedure under MRI guidance. The MR-conditional robot was evaluated in a Thiel embalmed cadaver study and healthy volunteer studies. The robot was attached to the shoulder using straps and ten locations in the shoulder joint space were selected as targets. For the first target, contrast agent (saline) was injected to complete the clinical workflow. After each targeting attempt, a confirmation scan was acquired to analyze the needle placement accuracy. During the volunteer studies, a more comfortable and ergonomic shoulder brace was used, and the complete clinical workflow was followed to measure the total procedure time. In the cadaver study, the needle was successfully placed in the shoulder joint space in all the targeting attempts with translational and rotational accuracy of 2.07 ± 1.22 mm and 1.46 ± 1.06 degrees, respectively. The total time for the entire procedure was 94 min and the average time for each targeting attempt was 20 min in the cadaver study, while the average time for the entire workflow for the volunteer studies was 36 min. No image quality degradation due to the presence of the robot was detected. This Thiel-embalmed cadaver study along with the clinical workflow studies on human volunteers demonstrated the feasibility of using an MR-conditional, patient-mounted robotic system for MRI-guided shoulder arthrography procedure. Future work will be focused on moving the technology to clinical practice.
Funder
National Institutes of Health
Subject
Artificial Intelligence,Computer Science Applications
Cited by
10 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献