Quantification and visualization of variation in anatomical trees

TL;DR

This paper introduces methods for quantifying and visualizing variations in anatomical tree datasets, using hypothesis testing, sparse classifiers, and hyperbolic embeddings, demonstrated on airway trees related to COPD.

Contribution

It proposes two novel approaches for analyzing anatomical tree variation: subtree localization with hypothesis testing and visualization via hyperbolic embeddings.

Findings

Identified significant subtree differences related to COPD

Demonstrated effective visualization of dataset structure

Validated methods on airway tree dataset

Abstract

This paper presents two approaches to quantifying and visualizing variation in datasets of trees. The first approach localizes subtrees in which significant population differences are found through hypothesis testing and sparse classifiers on subtree features. The second approach visualizes the global metric structure of datasets through low-distortion embedding into hyperbolic planes in the style of multidimensional scaling. A case study is made on a dataset of airway trees in relation to Chronic Obstructive Pulmonary Disease.