Image registration of an electromagnetic tracking enabled afterloader and CT using a phantom for the quality control of implant reconstruction

TL;DR

This study demonstrates effective registration of electromagnetic tracking data with CT images of a phantom, achieving sub-millimeter accuracy across different environments for improved implant reconstruction quality control.

Contribution

It introduces a method for registering EMT and CT reference frames using rigid registration and multiple approaches, ensuring high accuracy in various clinical environments.

Findings

Median registration errors below 1 mm in all environments

Environment did not significantly affect measurement accuracy

Registration errors were within clinically acceptable thresholds

Abstract

Electromagnetic tracking (EMT) shows great potential for automating implant reconstruction in brachytherapy. One of the challenges of this technology is that it does not intrinsically share the same reference frame as the patient's medical imaging. The purpose of this work is to register the reference frames of an EMT-enabled afterloader and a CT scan image of a rigid phantom for quality control of EMT-based implant reconstruction. The geometry of twelve 6F catheters and a tandem and ring gynecological applicator were reconstructed using an EMT sensor attached to an afterloader's check cable. All EMT reconstructions were done in three different environments: disturbance free, CT-on-rails brachytherapy suite and MRI brachytherapy suite. Implants were reconstructed using two acquisition methods: step-and-record and continuous motion. A CT scan of the phantom was obtained and manually reconstructed in the treatment planning system, with the help of the solid model of the applicator. Rigid registration algorithms were used to register EMT and CT scan reference frames using corresponding points from the EMT and CT based reconstructions of the phantom, following three approaches: one gynecological applicator, four catheters inside four calibration plates having and S-shaped path, and four ceramic marbles within the calibration plates. The absolute median deviation from the expected value for EMT measurements in all three environments significantly lie below the sensor's expected accuracy of 0.70 mm (p < 0.001), suggesting that the environment did not have a significant impact on the measurements, given that care is taken in the immediate surroundings. In all three environments, the two acquisition and three registration approaches have a mean and median registration errors (perpendicular distance) that lie at or below 1 mm, which is lower than the clinical acceptable threshold of 2 mm.