Learning deep abdominal CT registration through adaptive loss weighting and synthetic data generation
Perez de Frutos, Javier; Pedersen, Andre; Pelanis, Egidijus; Bouget, David Nicolas Jean-Mar; Survarachakan, Shanmugapriya; Langø, Thomas; Elle, Ole Jakob; Lindseth, Frank
Peer reviewed, Journal article
Published version
Permanent lenke
https://hdl.handle.net/11250/3101452Utgivelsesdato
2023Metadata
Vis full innførselSamlinger
- Publikasjoner fra CRIStin - SINTEF AS [5654]
- SINTEF Digital [2388]
Sammendrag
Purpose: This study aims to explore training strategies to improve convolutional neural network-based image-to-image deformable registration for abdominal imaging.
Methods: Different training strategies, loss functions, and transfer learning schemes were considered. Furthermore, an augmentation layer which generates artificial training image pairs on-the-fly was proposed, in addition to a loss layer that enables dynamic loss weighting.
Results: Guiding registration using segmentations in the training step proved beneficial for deep-learning-based image registration. Finetuning the pretrained model from the brain MRI dataset to the abdominal CT dataset further improved performance on the latter application, removing the need for a large dataset to yield satisfactory performance. Dynamic loss weighting also marginally improved performance, all without impacting inference runtime.
Conclusion: Using simple concepts, we improved the performance of a commonly used deep image registration architecture, VoxelMorph. In future work, our framework, DDMR, should be validated on different datasets to further assess its value.