Deformation equivariant cross-modality image synthesis with paired non-aligned training data

• Published in: Medical image analysis 2023-12, Vol.90, p.102940-102940, Article 102940
• Main Authors: Honkamaa, Joel, Khan, Umair, Koivukoski, Sonja, Valkonen, Mira, Latonen, Leena, Ruusuvuori, Pekka, Marttinen, Pekka
• Format: Article
• Language: English
• Subjects: Cross-modality image synthesis; Image registration; Image-to-image translation
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2023.102940

Cross-modality image synthesis is an active research topic with multiple medical clinically relevant applications. Recently, methods allowing training with paired but misaligned data have started to emerge. However, no robust and well-performing methods applicable to a wide range of real world data sets exist. In this work, we propose a generic solution to the problem of cross-modality image synthesis with paired but non-aligned data by introducing new deformation equivariance encouraging loss functions. The method consists of joint training of an image synthesis network together with separate registration networks and allows adversarial training conditioned on the input even with misaligned data. The work lowers the bar for new clinical applications by allowing effortless training of cross-modality image synthesis networks for more difficult data sets. •A generic cross-modality image synthesis architecture trainable with misaligned data.•The introduced deformation equivariance encouraging loss functions result in a robust and easily trainable method.•The method can be applied even to data sets with domain specific geometrical differences.•The method allows robust input conditioned adversarial training even with misaligned training data.