Learning a Generative Motion Model from Image Sequences based on a Latent Motion Matrix

TL;DR

This paper introduces a probabilistic, generative motion model based on a latent motion matrix for image sequence registration, enabling realistic motion simulation, interpolation, and improved data augmentation in medical imaging.

Contribution

The paper presents a novel unsupervised generative model with a multivariate Gaussian process prior for spatio-temporal registration and motion analysis, trained with amortized variational inference.

Findings

Improved registration accuracy over state-of-the-art methods.

Smoother, more realistic motion deformations.

Enhanced motion reconstruction from incomplete sequences.

Abstract

We propose to learn a probabilistic motion model from a sequence of images for spatio-temporal registration. Our model encodes motion in a low-dimensional probabilistic space - the motion matrix - which enables various motion analysis tasks such as simulation and interpolation of realistic motion patterns allowing for faster data acquisition and data augmentation. More precisely, the motion matrix allows to transport the recovered motion from one subject to another simulating for example a pathological motion in a healthy subject without the need for inter-subject registration. The method is based on a conditional latent variable model that is trained using amortized variational inference. This unsupervised generative model follows a novel multivariate Gaussian process prior and is applied within a temporal convolutional network which leads to a diffeomorphic motion model. Temporal consistency and generalizability is further improved by applying a temporal dropout training scheme. Applied to cardiac cine-MRI sequences, we show improved registration accuracy and spatio-temporally smoother deformations compared to three state-of-the-art registration algorithms. Besides, we demonstrate the model's applicability for motion analysis, simulation and super-resolution by an improved motion reconstruction from sequences with missing frames compared to linear and cubic interpolation.