Appearance Learning for Image-based Motion Estimation in Tomography

TL;DR

This paper introduces an appearance learning method using a siamese triplet network to detect and quantify patient motion artifacts in tomographic imaging, improving accuracy over previous methods.

Contribution

It presents a novel multi-task learning approach that predicts reprojection error and its distribution to recognize rigid motion structures independently of the scanned object.

Findings

Achieved residual mean RPE of 0.013mm, twice as accurate as previous results.

Outperformed two state-of-the-art motion measures in 9 of 12 benchmark experiments.

Demonstrated clinical applicability on motion-affected patient data.

Abstract

In tomographic imaging, anatomical structures are reconstructed by applying a pseudo-inverse forward model to acquired signals. Geometric information within this process is usually depending on the system setting only, i. e., the scanner position or readout direction. Patient motion therefore corrupts the geometry alignment in the reconstruction process resulting in motion artifacts. We propose an appearance learning approach recognizing the structures of rigid motion independently from the scanned object. To this end, we train a siamese triplet network to predict the reprojection error (RPE) for the complete acquisition as well as an approximate distribution of the RPE along the single views from the reconstructed volume in a multi-task learning approach. The RPE measures the motioninduced geometric deviations independent of the object based on virtual marker positions, which are available during training. We train our network using 27 patients and deploy a 21-4-2 split for training, validation and testing. In average, we achieve a residual mean RPE of 0.013mm with an inter-patient standard deviation of 0.022 mm. This is twice the accuracy compared to previously published results. In a motion estimation benchmark the proposed approach achieves superior results in comparison with two state-of-the-art measures in nine out of twelve experiments. The clinical applicability of the proposed method is demonstrated on a motion-affected clinical dataset.