Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Biomedical Engineering

Supervisor

Peters, Terry M.

Affiliation

Robarts Research Institute

Abstract

Transcatheter cardiac interventions are characterized by their percutaneous nature, increased patient safety, and low hospitalization times. Transcatheter procedures involve two major stages: navigation towards the target site and the positioning of tools to deliver the therapy, during which the interventionalists face the challenge of visualizing the anatomy and the relative position of the tools such as a guidewire. Fluoroscopic and transesophageal ultrasound (TEE) imaging are the most used techniques in cardiac procedures; however, they possess the disadvantage of radiation exposure and suboptimal imaging. This work explores the potential of intracardiac ultrasound (ICE) within an image guidance system (IGS) to facilitate the two stages of cardiac interventions. First, a novel 2.5D side-firing, conical Foresight ICE probe (Conavi Medical Inc., Toronto) is characterized, calibrated, and tracked using an electromagnetic sensor. The results indicate an acceptable tracking accuracy within some limitations. Next, an IGS is developed for navigating the vessels without fluoroscopy. A forward-looking, tracked ICE probe is used to reconstruct the vessel on a phantom which mimics the ultrasound imaging of an animal vena cava. Deep learning methods are employed to segment the complex vessel geometry from ICE imaging for the first time. The ICE-reconstructed vessel showed a clinically acceptable range of accuracy. Finally, a guidance system was developed to facilitate the positioning of tools during a tricuspid valve repair. The designed system potentially facilitates the positioning of the TriClip at the coaptation gap by pre-mapping the corresponding site of regurgitation in 3D tracking space.

Summary for Lay Audience

Heart surgeries have evolved from high-risk open-heart surgeries to much safer minimally invasive procedures. Transcatheter interventions often involve repairing a structural heart disease by accessing the heart through veins and arteries. Thin, wire-like tools such as a guidewire are inserted in the body via limbs and are then used to traverse the vessels using live x-ray technology called fluoroscopy. Once the tools reach the heart, they are then positioned at the pathological, target site and deployed to deliver therapy. This positioning of tools is often facilitated using an external ultrasound probe. In this work, we explore the potential of using a novel intracardiac ultrasound probe (ICE) to assist transcatheter procedures in an image-guided system (IGS). We augment a Foresight ICE probe (Conavi Medical Inc.) with an electromagnetic tracking sensor so the probe’s position can be always tracked in 3D space. Calibration methods to track the exact location of the ICE image are described as well. The second objective is to demonstrate the feasibility of using a tracked ICE probe in order to generate a vascular roadmap which can then be followed by a tracked guidewire to navigate the vessels. We designed an ultrasound-realistic vessel phantom and reconstructed the vessel in real-time using deep learning methods. The results indicate that ultrasound technology can be used instead of fluoroscopy to visualize and traverse the vessels. The third objective is to develop an IGS to assist the positioning of a therapeutic device during tricuspid valve repair surgery. Current imaging standards produce suboptimal imaging of the tricuspid valve and it can be challenging to identify the site of tricuspid valve regurgitation. We designed an algorithm to automatically detect the location of the regurgitation site from the color Doppler imaging on a tracked ICE probe. This method helps pre-map the location the clinicians have to target with the therapeutic device. This work demonstrates some of the ways an ICE ultrasound technology can improve and assist the existing procedural workflows by providing more information to the clinicians safely and accurately.

Download

Share

COinS