Exploring the Relationship Between Pulmonary Artery Shape and Pressure in Pulmonary Hypertension: A Statistical Shape Analysis Study

TY - CHAP

T1 - Exploring the Relationship Between Pulmonary Artery Shape and Pressure in Pulmonary Hypertension

T2 - 14th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2023 held in Conjunction with MICCAI 2023

AU - Sabry, Malak

AU - Hermida, Uxio

AU - Hassan, Ahmed

AU - Nagy, Michael

AU - Stojanovski, David

AU - Samuel, Irini

AU - Locas, John

AU - Yacoub, Magdi H.

AU - De Vecchi, Adelaide

AU - Lamata, Pablo

N1 - Publisher Copyright: © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.

PY - 2024

Y1 - 2024

N2 - Pulmonary Hypertension (PH) is a progressive condition affecting the right heart, defined by a mean pulmonary arterial pressure (mPAP) greater than 20 mmHg. Measuring mPAP with a pressure catheter is the gold standard for diagnosing PH despite its associated costs and risks. As an alternative, this work investigates the inference of mPAP from pulmonary vasculature anatomy. We thus studied the shape of the main pulmonary artery trunk along with its left and right branches (MPA, LPA, and RPA) across a population of group I PH patients and investigated the relationship between shape and mPAP. Computed Tomography images from 80 confirmed PH cases were used to create a statistical shape model: anatomy was manually segmented, represented by 3 centerlines and radii at evenly spaced control points, and reduced in dimensionality by Principal Component Analysis (PCA). The correlation between each PCA mode of variation and mPAP was then used to identify relevant shape features associated with elevated pressure, which were finally combined in a linear regression model. Results reveal that changes in MPA’s diameter and bulging, as well as in the angle between RPA and LPA, related to mPAP. A linear combination of the first 12 modes, with a cumulative variance of 95%, resulted in a linear regression model explaining 36% of mPAP variability in the population, while a combination of the 6 most relevant PCA modes was able to explain 34% of mPAP variability. These results provide initial evidence of the ability to infer mPAP from pulmonary arteries anatomy in PH patients.

AB - Pulmonary Hypertension (PH) is a progressive condition affecting the right heart, defined by a mean pulmonary arterial pressure (mPAP) greater than 20 mmHg. Measuring mPAP with a pressure catheter is the gold standard for diagnosing PH despite its associated costs and risks. As an alternative, this work investigates the inference of mPAP from pulmonary vasculature anatomy. We thus studied the shape of the main pulmonary artery trunk along with its left and right branches (MPA, LPA, and RPA) across a population of group I PH patients and investigated the relationship between shape and mPAP. Computed Tomography images from 80 confirmed PH cases were used to create a statistical shape model: anatomy was manually segmented, represented by 3 centerlines and radii at evenly spaced control points, and reduced in dimensionality by Principal Component Analysis (PCA). The correlation between each PCA mode of variation and mPAP was then used to identify relevant shape features associated with elevated pressure, which were finally combined in a linear regression model. Results reveal that changes in MPA’s diameter and bulging, as well as in the angle between RPA and LPA, related to mPAP. A linear combination of the first 12 modes, with a cumulative variance of 95%, resulted in a linear regression model explaining 36% of mPAP variability in the population, while a combination of the 6 most relevant PCA modes was able to explain 34% of mPAP variability. These results provide initial evidence of the ability to infer mPAP from pulmonary arteries anatomy in PH patients.

KW - Pulmonary Hypertension

KW - Remodelling

KW - Statistical Shape Analysis

UR - https://www.scopus.com/pages/publications/85186703271

U2 - 10.1007/978-3-031-52448-6_18

DO - 10.1007/978-3-031-52448-6_18

M3 - Conference paper

AN - SCOPUS:85186703271

SN - 9783031524479

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 186

EP - 195

BT - Statistical Atlases and Computational Models of the Heart. Regular and CMRxRecon Challenge Papers - 14th International Workshop, STACOM 2023, Held in Conjunction with MICCAI 2023, Revised Selected Papers

A2 - Camara, Oscar

A2 - Puyol-Antón, Esther

A2 - Suinesiaputra, Avan

A2 - Young, Alistair

A2 - Sermesant, Maxime

A2 - Tao, Qian

A2 - Wang, Chengyan

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 12 October 2023 through 12 October 2023

ER -