Deformation-Compensated Learning for Image Reconstruction Without Ground Truth

• Published in: IEEE transactions on medical imaging 2022-09, Vol.41 (9), p.2371-2384
• Main Authors: Gan, Weijie, Sun, Yu, Eldeniz, Cihat, Liu, Jiaming, An, Hongyu, Kamilov, Ulugbek S.
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Convolutional neural networks; Deep learning; Image processing; Image Processing, Computer-Assisted - methods; Image quality; Image reconstruction; Imaging; Inverse problems; Learning; Machine learning; Magnetic Resonance Imaging; magnetic resonance imaging (MRI); Medical imaging; Neural networks; Neural Networks, Computer; Noise measurement; Strain; Training
• ISSN: 0278-0062; 1558-254X; 1558-254X
• DOI: 10.1109/TMI.2022.3163018

Deep neural networks for medical image reconstruction are traditionally trained using high-quality ground-truth images as training targets. Recent work on Noise2Noise (N2N) has shown the potential of using multiple noisy measurements of the same object as an alternative to having a ground-truth. However, existing N2N-based methods are not suitable for learning from the measurements of an object undergoing nonrigid deformation. This paper addresses this issue by proposing the deformation-compensated learning (DeCoLearn) method for training deep reconstruction networks by compensating for object deformations. A key component of DeCoLearn is a deep registration module, which is jointly trained with the deep reconstruction network without any ground-truth supervision. We validate DeCoLearn on both simulated and experimentally collected magnetic resonance imaging (MRI) data and show that it significantly improves imaging quality.