SUPR
Deformable Image Registration for Improving Precision Radiotherapy
Dnr:

NAISS 2025/22-169

Type:

NAISS Small Compute

Principal Investigator:

Onur Ali Zeybekoglu

Affiliation:

Uppsala universitet

Start Date:

2025-03-10

End Date:

2026-04-01

Primary Classification:

10210: Artificial Intelligence

Webpage:

Allocation

Abstract

The overarching objective of this project is to enhance the precision and efficacy of radiation therapy (RT) by harnessing cutting-edge image registration technologies. At its core, the project seeks to innovate within the field of deformable image registration (DIR) — a technique vital for accurately assessing radiation doses to tumors and healthy tissues. By integrating DIR with the sophisticated computational capabilities of deep learning (DL), the project aims to develop a system that can determine the accumulated delivered dose, taking the variations in the patient anatomy into account. Moreover, the project places emphasis on uncertainty estimation — a process that quantifies the confidence in the predicted accumulated dose. The project intends to incorporate measures that can assess and communicate the degree of certainty in DIR and/or dose. This not only will foster greater trust in the system but will also provide clinicians with valuable information when making a treatment decision.