AuthorsJ. L. Bruse, K. S. Mcleod, G. Biglino, H. N. Ntsinjana, C. Capelli, T. Hsia, M. Sermesant, X. Pennec, A. Taylor and S. Schievano
TitleA Non-parametric Statistical Shape Model for Assessment of the Surgically Repaired Aortic Arch in Coarctation of the Aorta: How Normal is Abnormal?
AfilliationScientific Computing, Scientific Computing, Scientific Computing
StatusPublished
Publication TypeProceedings, refereed
Year of Publication2015
Conference NameStatistical Atlases and Computational Modeling of the Heart (STACOM 2015)
EditionMICCAI Workshop
Date Published10/2015
PublisherLecture Notes in Computer Science, Springer. Verlag
KeywordsAortic Arch, Coarctation of the Aorta, Mathematical Currents, Non-parametric Statistical Shape Model, Partial Least Square Regression
Abstract

Coarctation of the Aorta (CoA) is a cardiac defect that re- quires surgical intervention aiming to restore an unobstructed aortic arch shape. Many patients suffer from complications post-repair, which are commonly associated with arch shape abnormalities. Determining the degree of shape abnormality could improve risk stratification in recom- mended screening procedures. Yet, traditional morphometry struggles to capture the highly complex arch geometries. Therefore, we use a non- parametric Statistical Shape Model based on mathematical currents to fully account for 3D global and regional shape features. By comput- ing a template aorta of a population of healthy subjects and analysing its transformations towards CoA arch shape models using Partial Least Squares regression techniques, we derived a shape vector as a measure of subject-specific shape abnormality. Results were compared to a shape ranking by clinical experts. Our study suggests Statistical Shape Mod- elling to be a promising diagnostic tool for improved screening of complex cardiac defects. 

Notes

Oral Presentation - Jan L. Bruse

Citation Key23826