Development of a statistical model for discrimination of rupture status in posterior communicating artery aneurysms

Felicitas J. Detmer, Bong Jae Chung, Fernando Mut, Michael Pritz, Martin Slawski, Farid Hamzei-Sichani, David Kallmes, Christopher Putman, Carlos Jimenez, Juan R. Cebral

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Background: Intracranial aneurysms at the posterior communicating artery (PCOM) are known to have high rupture rates compared to other locations. We developed and internally validated a statistical model discriminating between ruptured and unruptured PCOM aneurysms based on hemodynamic and geometric parameters, angio-architectures, and patient age with the objective of its future use for aneurysm risk assessment. Methods: A total of 289 PCOM aneurysms in 272 patients modeled with image-based computational fluid dynamics (CFD) were used to construct statistical models using logistic group lasso regression. These models were evaluated with respect to discrimination power and goodness of fit using tenfold nested cross-validation and a split-sample approach to mimic external validation. Results: The final model retained maximum and minimum wall shear stress (WSS), mean parent artery WSS, maximum and minimum oscillatory shear index, shear concentration index, and aneurysm peak flow velocity, along with aneurysm height and width, bulge location, non-sphericity index, mean Gaussian curvature, angio-architecture type, and patient age. The corresponding area under the curve (AUC) was 0.8359. When omitting data from each of the three largest contributing hospitals in turn, and applying the corresponding model on the left-out data, the AUCs were 0.7507, 0.7081, and 0.5842, respectively. Conclusions: Statistical models based on a combination of patient age, angio-architecture, hemodynamics, and geometric characteristics can discriminate between ruptured and unruptured PCOM aneurysms with an AUC of 84%. It is important to include data from different hospitals to create models of aneurysm rupture that are valid across hospital populations.

Original languageEnglish
Pages (from-to)1643-1652
Number of pages10
JournalActa Neurochirurgica
Volume160
Issue number8
DOIs
StatePublished - 1 Aug 2018

Fingerprint

Intracranial Aneurysm
Statistical Models
Aneurysm
Rupture
Area Under Curve
Arteries
Hemodynamics
Hydrodynamics
Population

Keywords

  • Cerebral aneurysm
  • Hemodynamics
  • Morphology
  • Posterior communicating artery
  • Prediction
  • Rupture

Cite this

Detmer, Felicitas J. ; Chung, Bong Jae ; Mut, Fernando ; Pritz, Michael ; Slawski, Martin ; Hamzei-Sichani, Farid ; Kallmes, David ; Putman, Christopher ; Jimenez, Carlos ; Cebral, Juan R. / Development of a statistical model for discrimination of rupture status in posterior communicating artery aneurysms. In: Acta Neurochirurgica. 2018 ; Vol. 160, No. 8. pp. 1643-1652.
@article{a4f04fede93a42a1964d963e075b24f2,
title = "Development of a statistical model for discrimination of rupture status in posterior communicating artery aneurysms",
abstract = "Background: Intracranial aneurysms at the posterior communicating artery (PCOM) are known to have high rupture rates compared to other locations. We developed and internally validated a statistical model discriminating between ruptured and unruptured PCOM aneurysms based on hemodynamic and geometric parameters, angio-architectures, and patient age with the objective of its future use for aneurysm risk assessment. Methods: A total of 289 PCOM aneurysms in 272 patients modeled with image-based computational fluid dynamics (CFD) were used to construct statistical models using logistic group lasso regression. These models were evaluated with respect to discrimination power and goodness of fit using tenfold nested cross-validation and a split-sample approach to mimic external validation. Results: The final model retained maximum and minimum wall shear stress (WSS), mean parent artery WSS, maximum and minimum oscillatory shear index, shear concentration index, and aneurysm peak flow velocity, along with aneurysm height and width, bulge location, non-sphericity index, mean Gaussian curvature, angio-architecture type, and patient age. The corresponding area under the curve (AUC) was 0.8359. When omitting data from each of the three largest contributing hospitals in turn, and applying the corresponding model on the left-out data, the AUCs were 0.7507, 0.7081, and 0.5842, respectively. Conclusions: Statistical models based on a combination of patient age, angio-architecture, hemodynamics, and geometric characteristics can discriminate between ruptured and unruptured PCOM aneurysms with an AUC of 84{\%}. It is important to include data from different hospitals to create models of aneurysm rupture that are valid across hospital populations.",
keywords = "Cerebral aneurysm, Hemodynamics, Morphology, Posterior communicating artery, Prediction, Rupture",
author = "Detmer, {Felicitas J.} and Chung, {Bong Jae} and Fernando Mut and Michael Pritz and Martin Slawski and Farid Hamzei-Sichani and David Kallmes and Christopher Putman and Carlos Jimenez and Cebral, {Juan R.}",
year = "2018",
month = "8",
day = "1",
doi = "10.1007/s00701-018-3595-8",
language = "English",
volume = "160",
pages = "1643--1652",
journal = "Acta Neurochirurgica",
issn = "0001-6268",
number = "8",

}

Detmer, FJ, Chung, BJ, Mut, F, Pritz, M, Slawski, M, Hamzei-Sichani, F, Kallmes, D, Putman, C, Jimenez, C & Cebral, JR 2018, 'Development of a statistical model for discrimination of rupture status in posterior communicating artery aneurysms', Acta Neurochirurgica, vol. 160, no. 8, pp. 1643-1652. https://doi.org/10.1007/s00701-018-3595-8

Development of a statistical model for discrimination of rupture status in posterior communicating artery aneurysms. / Detmer, Felicitas J.; Chung, Bong Jae; Mut, Fernando; Pritz, Michael; Slawski, Martin; Hamzei-Sichani, Farid; Kallmes, David; Putman, Christopher; Jimenez, Carlos; Cebral, Juan R.

In: Acta Neurochirurgica, Vol. 160, No. 8, 01.08.2018, p. 1643-1652.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development of a statistical model for discrimination of rupture status in posterior communicating artery aneurysms

AU - Detmer, Felicitas J.

AU - Chung, Bong Jae

AU - Mut, Fernando

AU - Pritz, Michael

AU - Slawski, Martin

AU - Hamzei-Sichani, Farid

AU - Kallmes, David

AU - Putman, Christopher

AU - Jimenez, Carlos

AU - Cebral, Juan R.

PY - 2018/8/1

Y1 - 2018/8/1

N2 - Background: Intracranial aneurysms at the posterior communicating artery (PCOM) are known to have high rupture rates compared to other locations. We developed and internally validated a statistical model discriminating between ruptured and unruptured PCOM aneurysms based on hemodynamic and geometric parameters, angio-architectures, and patient age with the objective of its future use for aneurysm risk assessment. Methods: A total of 289 PCOM aneurysms in 272 patients modeled with image-based computational fluid dynamics (CFD) were used to construct statistical models using logistic group lasso regression. These models were evaluated with respect to discrimination power and goodness of fit using tenfold nested cross-validation and a split-sample approach to mimic external validation. Results: The final model retained maximum and minimum wall shear stress (WSS), mean parent artery WSS, maximum and minimum oscillatory shear index, shear concentration index, and aneurysm peak flow velocity, along with aneurysm height and width, bulge location, non-sphericity index, mean Gaussian curvature, angio-architecture type, and patient age. The corresponding area under the curve (AUC) was 0.8359. When omitting data from each of the three largest contributing hospitals in turn, and applying the corresponding model on the left-out data, the AUCs were 0.7507, 0.7081, and 0.5842, respectively. Conclusions: Statistical models based on a combination of patient age, angio-architecture, hemodynamics, and geometric characteristics can discriminate between ruptured and unruptured PCOM aneurysms with an AUC of 84%. It is important to include data from different hospitals to create models of aneurysm rupture that are valid across hospital populations.

AB - Background: Intracranial aneurysms at the posterior communicating artery (PCOM) are known to have high rupture rates compared to other locations. We developed and internally validated a statistical model discriminating between ruptured and unruptured PCOM aneurysms based on hemodynamic and geometric parameters, angio-architectures, and patient age with the objective of its future use for aneurysm risk assessment. Methods: A total of 289 PCOM aneurysms in 272 patients modeled with image-based computational fluid dynamics (CFD) were used to construct statistical models using logistic group lasso regression. These models were evaluated with respect to discrimination power and goodness of fit using tenfold nested cross-validation and a split-sample approach to mimic external validation. Results: The final model retained maximum and minimum wall shear stress (WSS), mean parent artery WSS, maximum and minimum oscillatory shear index, shear concentration index, and aneurysm peak flow velocity, along with aneurysm height and width, bulge location, non-sphericity index, mean Gaussian curvature, angio-architecture type, and patient age. The corresponding area under the curve (AUC) was 0.8359. When omitting data from each of the three largest contributing hospitals in turn, and applying the corresponding model on the left-out data, the AUCs were 0.7507, 0.7081, and 0.5842, respectively. Conclusions: Statistical models based on a combination of patient age, angio-architecture, hemodynamics, and geometric characteristics can discriminate between ruptured and unruptured PCOM aneurysms with an AUC of 84%. It is important to include data from different hospitals to create models of aneurysm rupture that are valid across hospital populations.

KW - Cerebral aneurysm

KW - Hemodynamics

KW - Morphology

KW - Posterior communicating artery

KW - Prediction

KW - Rupture

UR - http://www.scopus.com/inward/record.url?scp=85048746840&partnerID=8YFLogxK

U2 - 10.1007/s00701-018-3595-8

DO - 10.1007/s00701-018-3595-8

M3 - Article

C2 - 29922867

AN - SCOPUS:85048746840

VL - 160

SP - 1643

EP - 1652

JO - Acta Neurochirurgica

JF - Acta Neurochirurgica

SN - 0001-6268

IS - 8

ER -