A Deep Metric for Multimodal Registration

TL;DR

This paper introduces a learning-based deep neural network metric for multimodal medical image registration, demonstrating superior performance over traditional measures like mutual information and good generalization across different datasets.

Contribution

The authors propose a convolutional neural network-based similarity metric that can be trained from scratch with limited data and generalizes well to different datasets for multimodal registration.

Findings

Outperforms mutual information significantly in experiments.

Can be trained from scratch with few aligned image pairs.

Generalizes effectively to different datasets.

Abstract

Multimodal registration is a challenging problem in medical imaging due the high variability of tissue appearance under different imaging modalities. The crucial component here is the choice of the right similarity measure. We make a step towards a general learning-based solution that can be adapted to specific situations and present a metric based on a convolutional neural network. Our network can be trained from scratch even from a few aligned image pairs. The metric is validated on intersubject deformable registration on a dataset different from the one used for training, demonstrating good generalization. In this task, we outperform mutual information by a significant margin.