Modality-agnostic self-supervised deep feature learning and fast instance optimisation for multimodal fusion in ultrasound-guided interventions

•We propose a fast self-supervised deep learning based registration framework for multi-modal medical 3D registration in ultrasound-guided interventions.•First deep learning model that achieves state-of-the-art accuracy of only 2.50mmtarget registration error on challenging alignment task in only 3 seconds.•Instance optimisation that couples a large discretized capture range with global transform conformity and enables significant gains in registration accuracy and robustness. Background and Objective: Fast and robust alignment of pre-operative MRI planning scans to intra-operative ultrasound is an important aspect for automatically supporting image-guided interventions. Thus far, learning-based approaches have failed to tackle the intertwined objectives of fast inference computation time and robustness to unexpectedly large motion and misalignment. In this work, we propose a novel method that decouples deep feature learning and the computation of long ranging local displacement probability maps from fast and robust global transformation prediction. Methods: In our approach, we firstly train a convolutional neural network (CNN) to extract modality-agnostic features with sub-second computation times for both 3D volumes during inference. Using sparsity-based network weight pruning, the model complexity and computation times can be substantially reduced. Based on these features, a large discretized search range of 3D motion vectors is explored to compute a probabilistic displacement map for each control point. These 3D probability maps are employed in our newly proposed, computationally efficient, instance optimisation that robustly estimates the most likely globally linear transformation that best reflects the local displacement beliefs subject to outlier rejection. Results: Our experimental validation demonstrates state-of-the-art accuracy on the challenging CuRIOUS dataset with average target registration errors of 2.50 mm, model size of only 1.2 MByte and run times of approx. 3 seconds for a full 3D multimodal registration. Conclusion: We show that a significant improvement in accuracy and robustness can be gained with instance optimisation and our fast self-supervised deep learning model can achieve state-of-the-art accuracy on challenging registration task in only 3 seconds.