A Meta-Learning Approach for Medical Image Registration

Non-rigid registration is a necessary but challenging task in medical imaging studies. Recently, unsupervised registration models have shown good performance, but they often require a large-scale training dataset and long training time. Therefore, in real world application where only dozens to hundreds of image pairs are available, existing models cannot be practically used. To address these limitations, we propose a novel unsupervised registration model which is integrated with a gradient-based meta learning framework. In particular, we train a meta learner which finds an optimal initialization point of parameters by utilizing various registration datasets. To quickly adapt to diverse tasks, the meta learner was updated to get close to the center of parameters which are fine-tuned for each registration task. Thereby, our model can adapt to unseen domain tasks via a short fine-tuning process and perform accurate registration. To verify the superiority of our model, we train the model using various types of medical data sets such as retinal Optical Coherence Tomography Angiography (OCTA) for choroidal vasculature, body CT scans, and brain MRI scans and then test it on registration of unseen retinal OCTA for Superficial Capillary Plexus (SCP). In our experiments, the proposed model obtained significantly improved performance in terms of accuracy and training time compared to other registration models.