Multi-Structure Deep Segmentation with Shape Priors and Latent Adversarial Regularization

TL;DR

This paper introduces a novel deep learning segmentation method that uses shape priors and adversarial regularization to improve multi-structure bone delineation in pediatric MR images, addressing data scarcity and heterogeneity.

Contribution

It proposes a new shape code discriminator and a shape priors based adversarial regularization (SPAR) scheme for more accurate and consistent segmentation.

Findings

Outperforms state-of-the-art regularization methods on pediatric datasets

Effectively enforces anatomical shape constraints in segmentation

Improves robustness to data heterogeneity

Abstract

Automatic segmentation of the musculoskeletal system in pediatric magnetic resonance (MR) images is a challenging but crucial task for morphological evaluation in clinical practice. We propose a deep learning-based regularized segmentation method for multi-structure bone delineation in MR images, designed to overcome the inherent scarcity and heterogeneity of pediatric data. Based on a newly devised shape code discriminator, our adversarial regularization scheme enforces the deep network to follow a learnt shape representation of the anatomy. The novel shape priors based adversarial regularization (SPAR) exploits latent shape codes arising from ground truth and predicted masks to guide the segmentation network towards more consistent and plausible predictions. Our contribution is compared to state-of-the-art regularization methods on two pediatric musculoskeletal imaging datasets from ankle and shoulder joints.