Master Thesis Creating Synthetic Motion in MR Images for Evaluation of Novel Approaches to Motion Management on Unity

Elekta

Stockholm · Posted 3h ago

Apply

At a glance

Location: Stockholm, Sweden
Work type: Onsite

Master thesis to create synthetic motion in MR images for evaluating motion-management approaches on Elekta Unity, generating 3D/2D MR datasets with ground truth and benchmarking tracking algorithms.

What they are looking for

ProgrammingPythonMedical image processingQuantitative analysis

Location: Stockholm

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

Radiotherapy targets and surrounding anatomy can move during treatment. The Elekta Unity combines an MR scanner with a linear accelerator, permitting anatomy to be visualised during radiation delivery, and enabling correction based on live information.

This project will investigate how synthetic MR image data can be used to accelerate the development and evaluation of motion-management technologies on the Elekta Unity. The student will generate anatomically plausible motion within MR images, create realistic Unity-like cine datasets with known ground truth, and develop quantitative methods for assessing tracking accuracy. The resulting framework will provide a reusable platform for benchmarking current and future motion-tracking algorithms.

Background

A radiotherapy treatment is designed with millimeter precision to ensure that high doses of radiation can be delivered to the target, whilst avoiding critical anatomic structures. However, both the target and the surrounding anatomy may move during treatment, due to a range of physiological processes such as respiration, peristalsis, and organ drift. The Elekta Unity can visualize and respond to this motion in real-time, increasing dosimetric accuracy, and thus improving patient outcomes.

The current Automatic Position Management (APM) framework acquires an orthogonal pair of 2D cine images at a rate of 5 Hz. Each image is compared to a reference using rigid registration to estimate target motion from cine MR images. This method is challenged in cases where there is deformation of anatomy or so called “through-plane motion” i.e. anatomical motion normal to the imaging plane.

It may be possible to improve the performance of APM, i.e. by accounting for deformable motion or by modifying the imaging planes. However, it is not possible to verify these improvements without a reliable ground truth. Therefore, we propose creating a series of 3D images with anatomically plausible motion and extracting 2D images from them. This will allow algorithms to be tested on the 2D images and compared to a 3D ground truth.

Problem Statement

The project will address three linked problems

Generate paired 3D and 2D MR image datasets containing artificial but anatomically plausible motion. Pretreatment MR images will be transformed using plausible deformation vector fields, and synthetic sagittal and coronal cine images will be generated with characteristics representative of Unity acquisitions.

Establish baseline performance for the current APM algorithm. The synthetic cine images will be processed using the Motion Management Replay Tool (MMRP), with quantitative metrics defined for displacement accuracy and known failure modes including jitter, through-plane motion and anatomical deformation.

Explore one or more approaches for improving target tracking performance, for example by incorporating deformable motion information into the tracking process.

Output

A synthetic dataset of paired 3D MR images and Unity-like 2D cine images for future testing.

A defined set of quantitative evaluation metrics and a baseline measurement of current APM tracking accuracy.

A preliminary evaluation of alternative tracking approaches incorporating deformable motion information.

Prototype scripts and reproducible processing pipelines.

A written student report and final presentation.

Who are we looking for?

We are looking for a master’s student in medical physics, computer science, applied mathematics, engineering physics, or a related scientific discipline.