Zero Shot Learning for Multi-Modal Real Time Image Registration
This work addresses the need for fast and unsupervised image registration in medical imaging, though it appears incremental as it adapts existing pre-trained models to a new application.
The authors tackled the problem of medical image registration without task-specific training by using a pre-trained deep neural network as a feature extractor, achieving robust and real-time results for affine transformations on brain MRI datasets.
In this report we present an unsupervised image registration framework, using a pre-trained deep neural network as a feature extractor. We refer this to zero-shot learning, due to nonoverlap between training and testing dataset (none of the network modules in the processing pipeline were trained specifically for the task of medical image registration). Highlights of our technique are: (a) No requirement of a training dataset (b) Keypoints i.e.locations of important features are automatically estimated (c) The number of key points in this model is fixed and can possibly be tuned as a hyperparameter. (d) Uncertaintycalculation of the proposed, transformation estimates (e) Real-time registration of images. Our technique was evaluated on BraTS, ALBERT, and collaborative hospital Brain MRI data. Results suggest that the method proved to be robust for affine transformation models and the results are practically instantaneous, irrespective of the size of the input image