Meta-Learning Initializations for Interactive Medical Image Registration
This work addresses the problem of real-time, data-efficient registration for medical imaging, specifically for intraoperative MR-TRUS alignment, though it is incremental as it builds on existing learning-based methods.
The paper tackles interactive medical image registration by proposing a meta-learning framework that learns adaptable network initializations, achieving a registration error of 4.26 mm, comparable to non-interactive methods (3.97 mm) while using less data and operating in real-time.
We present a meta-learning framework for interactive medical image registration. Our proposed framework comprises three components: a learning-based medical image registration algorithm, a form of user interaction that refines registration at inference, and a meta-learning protocol that learns a rapidly adaptable network initialization. This paper describes a specific algorithm that implements the registration, interaction and meta-learning protocol for our exemplar clinical application: registration of magnetic resonance (MR) imaging to interactively acquired, sparsely-sampled transrectal ultrasound (TRUS) images. Our approach obtains comparable registration error (4.26 mm) to the best-performing non-interactive learning-based 3D-to-3D method (3.97 mm) while requiring only a fraction of the data, and occurring in real-time during acquisition. Applying sparsely sampled data to non-interactive methods yields higher registration errors (6.26 mm), demonstrating the effectiveness of interactive MR-TRUS registration, which may be applied intraoperatively given the real-time nature of the adaptation process.