Geometric tree kernels: classification of COPD from airway tree geometry

TY - GEN

T1 - Geometric tree kernels

T2 - 23rd International Conference on Information Processing in Medical Imaging

AU - Feragen, Aasa

AU - Petersen, Jens

AU - Grimm, Dominik

AU - Dirksen, Asger

AU - Pedersen, Jesper Johannes Holst

AU - Borgwardt, Karsten

AU - de Bruijne, Marleen

N1 - Conference code: 23

PY - 2013

Y1 - 2013

N2 - Methodological contributions: This paper introduces a family of kernels for analyzing (anatomical) trees endowed with vector valued measurements made along the tree. While state-of-the-art graph and tree kernels use combinatorial tree/graph structure with discrete node and edge labels, the kernels presented in this paper can include geometric information such as branch shape, branch radius or other vector valued properties. In addition to being flexible in their ability to model different types of attributes, the presented kernels are computationally efficient and some of them can easily be computed for large datasets (N ~10.000) of trees with 30 - 600 branches. Combining the kernels with standard machine learning tools enables us to analyze the relation between disease and anatomical tree structure and geometry. Experimental results: The kernels are used to compare airway trees segmented from low-dose CT, endowed with branch shape descriptors and airway wall area percentage measurements made along the tree. Using kernelized hypothesis testing we show that the geometric airway trees are significantly differently distributed in patients with Chronic Obstructive Pulmonary Disease (COPD) than in healthy individuals. The geometric tree kernels also give a significant increase in the classification accuracy of COPD from geometric tree structure endowed with airway wall thickness measurements in comparison with state-of-the-art methods, giving further insight into the relationship between airway wall thickness and COPD. Software: Software for computing kernels and statistical tests is available at http://image.diku.dk/aasa/software.php.

AB - Methodological contributions: This paper introduces a family of kernels for analyzing (anatomical) trees endowed with vector valued measurements made along the tree. While state-of-the-art graph and tree kernels use combinatorial tree/graph structure with discrete node and edge labels, the kernels presented in this paper can include geometric information such as branch shape, branch radius or other vector valued properties. In addition to being flexible in their ability to model different types of attributes, the presented kernels are computationally efficient and some of them can easily be computed for large datasets (N ~10.000) of trees with 30 - 600 branches. Combining the kernels with standard machine learning tools enables us to analyze the relation between disease and anatomical tree structure and geometry. Experimental results: The kernels are used to compare airway trees segmented from low-dose CT, endowed with branch shape descriptors and airway wall area percentage measurements made along the tree. Using kernelized hypothesis testing we show that the geometric airway trees are significantly differently distributed in patients with Chronic Obstructive Pulmonary Disease (COPD) than in healthy individuals. The geometric tree kernels also give a significant increase in the classification accuracy of COPD from geometric tree structure endowed with airway wall thickness measurements in comparison with state-of-the-art methods, giving further insight into the relationship between airway wall thickness and COPD. Software: Software for computing kernels and statistical tests is available at http://image.diku.dk/aasa/software.php.

U2 - 10.1007/978-3-642-38868-2_15

DO - 10.1007/978-3-642-38868-2_15

M3 - Article in proceedings

SN - 978-3-642-38867-5

T3 - Lecture notes in computer science

SP - 171

EP - 183

BT - Information Processing in Medical Imaging

A2 - Gee, James C.

A2 - Joshi, Sarang

A2 - Pohl, Kilian M.

A2 - Wells, William M.

A2 - Zöllei, Lilla

PB - Springer

Y2 - 28 June 2013 through 3 July 2013

ER -