Predicting Shape Development: a Riemannian Method

TL;DR

This paper introduces a Riemannian shape space-based prediction method for anatomical shape development, outperforming existing deep learning and state-of-the-art techniques in clinical and motion prediction tasks.

Contribution

It presents a novel Riemannian approach for shape prediction that effectively models nonlinear shape changes in curved spaces, improving accuracy over existing methods.

Findings

Outperforms deep learning-supported variants in shape prediction tasks.

Effective in predicting hippocampal shape changes in Alzheimer's disease.

Accurately predicts human body motion using the proposed method.

Abstract

Predicting the future development of an anatomical shape from a single baseline observation is a challenging task. But it can be essential for clinical decision-making. Research has shown that it should be tackled in curved shape spaces, as (e.g., disease-related) shape changes frequently expose nonlinear characteristics. We thus propose a novel prediction method that encodes the whole shape in a Riemannian shape space. It then learns a simple prediction technique founded on hierarchical statistical modeling of longitudinal training data. When applied to predict the future development of the shape of the right hippocampus under Alzheimer's disease and to human body motion, it outperforms deep learning-supported variants as well as state-of-the-art.