header

Medical Imaging & Tomographic Reconstruction (2022–)

Cone-beam CT is central to image-guided radiation therapy, but patient motion and long scan times make high-quality, time-resolved imaging difficult in practice.

Research collaboration between ZHAW and Varian Medical Imaging. Design, development and validation of deep learning methods for tomographic reconstruction and motion management. CNN architectures for time-resolved 4D reconstruction, artifact reduction, and sparse-view imaging.

Much of the work involved coming up with new approaches that combine classical reconstruction with trainable convolutional neural networks. Developing synthethic data generation pipelines and coming up with evaluation methods for tasks where real-world ground-truth is generally not available.

Publications

  • Herzig, I., Paysan, P., Barco, D., Stadelmann, M. A., Schilling, F.-P., Peterlik, I., Walczak, M., Aryananda, L., Ahn, W. S., Füchslin, R. M., & Lichtensteiger, L. (2026). Dual-domain U-Nets with embedded back projection operators for motion-resolved 4D CBCT reconstruction. arXiv arXiv:2608.03430. https://doi.org/10.48550/arXiv.2608.03430

  • Barco, D., Stadelmann, M., Oswald, M., Herzig, I., Lichtensteiger, L., Paysan, P., Peterlik, I., Walczak, M., Menze, B., & Schilling, F. (2026). MInDI-3D: Iterative Deep Learning in 3D for Sparse-View Cone Beam Computed Tomography. IEEE Access, 14, 6438–6449. https://doi.org/10.1109/ACCESS.2026.3652627

  • Amirian, M., Barco, D., Herzig, I., & Schilling, F. (2024). Artifact Reduction in 3D and 4D Cone-Beam Computed Tomography Images With Deep Learning: A Review. IEEE Access, 12, 10281–10295. https://doi.org/10.1109/ACCESS.2024.3353195

  • Amirian, M., Montoya-Zegarra, J. A., Herzig, I., Eggenberger Hotz, P., Lichtensteiger, L., Morf, M., Züst, A., Paysan, P., Peterlik, I., Scheib, S., Füchslin, R. M., Stadelmann, T., & Schilling, F. (2023). Mitigation of motion-induced artifacts in cone beam computed tomography using deep convolutional neural networks. Medical Physics, 50(10), 6228–6242. https://doi.org/10.1002/mp.16405

  • Herzig, I., Paysan, P., Scheib, S., Züst, A., Schilling, F.-P., Montoya, J., Amirian, M., Stadelmann, T., Eggenberger Hotz, P., Füchslin, R. M., & Lichtensteiger, L. (2022). Deep learning-based simultaneous multi-phase deformable image registration of sparse 4D-CBCT [Conference poster]. American Association of Physicists in Medicine. https://doi.org/10.21256/zhaw-25181