Registration of Longitudinal Spine CTs for Monitoring Lesion Growth

TL;DR

This paper introduces a fully automatic method for longitudinal spine CT registration that combines deep learning for vertebrae localization with Gaussian mixture model surface registration, enabling accurate lesion progression assessment.

Contribution

The novel two-step pipeline integrates deep learning and surface registration to improve automatic alignment of longitudinal spine CTs for monitoring lesions.

Findings

Achieved average Hausdorff distance of 0.65 mm

Attained average Dice score of 0.92

Successfully registered 111 vertebrae across multiple time points

Abstract

Accurate and reliable registration of longitudinal spine images is essential for assessment of disease progression and surgical outcome. Implementing a fully automatic and robust registration is crucial for clinical use, however, it is challenging due to substantial change in shape and appearance due to lesions. In this paper we present a novel method to automatically align longitudinal spine CTs and accurately assess lesion progression. Our method follows a two-step pipeline where vertebrae are first automatically localized, labeled and 3D surfaces are generated using a deep learning model, then longitudinally aligned using a Gaussian mixture model surface registration. We tested our approach on 37 vertebrae, from 5 patients, with baseline CTs and 3, 6, and 12 months follow-ups leading to 111 registrations. Our experiment showed accurate registration with an average Hausdorff distance of 0.65 mm and average Dice score of 0.92.