Published on 16.03.2026

NEW on CONNECT: Evaluating Motion and Registration in Adaptive Radiotherapy

Composite medical visualization showing two related views of human anatomy. On the left, a 3D rendering of a human skeleton (skull, spine, ribs, and shoulders) with numerous colored vectors or arrows around the chest and rib cage indicating motion or force directions. On the right, two CT scan slices of the upper chest: a frontal view at the top and a transverse (cross-sectional) view at the bottom, with colored overlays highlighting anatomical structures around the lungs, heart, and shoulder area.

Image: Hendrik Teske | info

Modern radiotherapy can deliver radiation with remarkable precision. However, in areas such as the thorax, breathing motion causes tumors and surrounding organs to move in complex ways, making accurate imaging and dose delivery more challenging.

In a new blog post on the Helmholtz Imaging CONNECT platform, Richard Häcker explores how these challenges affect adaptive radiotherapy and the evaluation of medical image registration methods. Using lung radiotherapy as an example, the article highlights how even small misalignments between CT and MRI scans can influence dose calculations and biomarker analysis.

The post also discusses how learned image representations and new evaluation metrics could improve the assessment of motion and registration algorithms when ground truth data is unavailable.

Read the full blog post on CONNECT

This work was carried out within the Helmholtz Imaging project CLARITY.