Extending XPos3R to MRI and ultrasound volumes

Extend XPos3R from CT-based volumetric inputs to MRI and ultrasound by determining the additional processing required, including possible modality translation or pseudo-CT generation through domain-transfer methods.

Background

The presented method uses CT because CT and X-ray have similar imaging physics, which enables large-scale digitally reconstructed radiograph synthesis. Extending the framework to MRI or ultrasound is unresolved because these modalities may require modality translation, such as generating pseudo-CT volumes through domain-transfer methods.

References

Extending XPos3R to other 3D modalities, such as MRI or ultrasound, may require additional steps such as modality translation, \ie, generating pseudo-CT using domain transfer methods. Despite these limitations, XPos3R demonstrate competitive performance and promising clinical values, and we leave these extensions to future work.

— XPos3R: Cross-Modal Transformer for Intraoperative 2D/3D Registration  (2609.10733 - Su et al., 9 Sep 2026) in Section Discussion, subsection Limitations