Master Thesis Automatic CT/MR image registration

Are you a current Elekta employee?

Please click here to apply through our internal career site Find Jobs - Elekta.

We don’t just build technology. We build hope for everyone dealing with Cancer.

Elekta is shaping the standard of care, together with healthcare providers, by developing precision radiation medicine that brings advanced radiotherapy into routine clinical practice. Our solutions support consistent, efficient and personalized care, addressing evolving patient needs. Through adaptive radiotherapy, precision treatment delivery and integrated workflows, Elekta supports care teams across a range of clinical settings. Each year more than 2 million patients are treated using Elekta solutions across over 130 countries. Elekta is headquartered in Stockholm, Sweden, with over 4,000 employees and offices in more than 40 countries. The company is listed on Nasdaq Stockholm. For more information, visit elekta.com.

Introduction

Leksell Gamma Plan (LGP) is a successful treatment planning system for stereotactic neuro radiotherapy performed by the Leksell Gamma Knife. Both are made by Elekta in Stockholm.

LGP is used by neurosurgeons and radio oncologists worldwide to analyse brain imaging and identify targets and choose doses of gamma radiation, as well as planning the application of the required doses to these targets by the Leksell Gamma Knife over a series of one or more treatment fractions.

Background

An important step in each treatment is correctly determining the position of structures shown in CT/MR brain imaging in a co-ordinate system relative to the machine. A mechanical frame with fiducial marks is firmly attached to the patient’s head prior to imaging, and the marks are visible in the resulting images. The frame is firmly attached to the Gamma Knife during treatment, allowing the correct calculation of positions of structures in the brain in a known co-ordinate system (calculation of the transformation from image co-ordinates to machine co-ordinates is known as registration).

Problem Statement

Present image registration algorithm for tomographic (MR/CT) images is semi-automatic and we would like to automate it and see if better registration transforms can be found.

Output

This thesis project aims to:

a) Propose and prototype method(s) for fully automatic detection of fiducial marks in MR/CT images.

b) Propose and prototype method(s) for fully automatic image registration based on fiducials.

c) Propose relevant metric(s) for quality assurance (e.g. mean and max errors).

d) Evaluate robustness and accuracy for proposed method(s) compared to current fiducial based

registration in LGP (for >100 representative and varying cases).

Who are we looking for?

Masters student, probably specialized in image handling/graphics.

No prior knowledge of Gamma Knife treatment planning is required. Domain experts from Elekta will provide introductions to the relevant aspects of the software architecture and workflow. The student will work closely with experts in software development and clinical applications throughout the project.

Practicalities

Location: Elekta office, Stockholm

Supervisor: Ethan Palmiere, [email protected]

Start date: Early 2027