Skip to main content

MECANICA COMPUTACIONAL Y APRENDIZAJE AUTOMATICO PARA EL APOYO EN LA TOMA DE DECISIONES URGENTE EN EL TRATAMIENTO DEL ICTUS ISQUEMICO

PID2022-136668OA-I00

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Call year 2022
State Plan 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre CENTRE INTERNACIONAL DE METODES NUMERICS A L ENGINYERIA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 4
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Machine learning and statistical shape modelling for real-time prediction of stent deployment in realistic anatomies

  • Bisighini, Beatrice
  • Aguirre Font, Miquel|||0000-0003-1798-7908
  • Pierrat, Baptiste
  • Avril, Stephane
Background and Objective. The rise in minimally invasive procedures has created a demand for efficient and reliable planning software to predict intra- and post-operative outcomes. Surrogate modelling has shown promise, but challenges remain, particularly in cardiovascular applications, due to the complexity of parametrising anatomical structures and the need for large training datasets. This study aims to apply statistical shape modelling and machine learning for predicting stent deployment in real time using patient-specific models. Methods. We built a statistical shape model starting from an open-source clinical dataset, which we then used to generate new synthetic cases. Finite element simulations of stent deployment were performed on these cases using an in-house software. A surrogate model was then trained to map the statistical features of the synthetic models to the corresponding stent configurations, evaluating sensitivity to dataset size. Results. Even with the smallest dataset (400 samples), the average prediction error in position among the tested cases never exceeded 8.6%, with a median one within the testing dataset of 1.6%. As the number of training samples increased (4900), we achieved a median position error lower than 0.1 mm (0.97%) and a maximum position error of 0.5 mm (4.8%). Notably, the largest errors occur in the radial direction of the stent, while the deployed length is accurately predicted in all the cases. Conclusions. The consistent success in performance strongly suggests that surrogate modelling represents a clinically valuable tool for accurately computing stent deployment outcomes in real time, even within complex anatomical scenarios., This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement 859836 and the French region Auvergne-Rhône-Alpes under the “Ambition Internationale” project. Miquel Aguirre acknowledges the support of Grant PID2022-136668OA-I00 funded by MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe and a María Zambrano research fellowship at Universitat Politècnica de Catalunya granted by Ministerio de Universidades ., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A systematic comparison of membrane, shell, and 3D solid formulations for nonlinear vascular biomechanics

  • Kim, Taeouk|||0000-0003-2690-123X
  • Ortigosa Martínez, Rogelio|||0000-0002-4542-2237
  • Nama, Nitesh
  • Aguirre Font, Miquel|||0000-0003-1798-7908
  • Gil, Antonio J.
  • Humphrey, Jay D.
  • Figueroa, C. Alberto
Typical computational methods for vascular biosolid mechanics represent the blood vessel wall as a membrane, shell, or 3D solid. Each of these formulations has advantages and disadvantages concerning accuracy, ease of implementation, and computational costs. Despite the widespread use of these formulations, a systematic comparison of the performance and accuracy of these formulations for nonlinear vascular biomechanics has remained wanting. Therefore, the decision regarding the optimal choice often relies on intuition or previous experience, with unclear consequences of choosing one approach over the other. Here, we present a systematic comparison among three different formulations to represent the vessel wall as: (i) a nonlinear membrane, (ii) a nonlinear, rotation-free shell, and (iii) a nonlinear 3D solid. For the 3D solid model, we consider two different implementations employing linear and quadratic interpolation. Convergence analysis for displacement and stress are presented for all formulations. We compare results in both idealized and subject-specific mouse aortic geometries. For the idealized cylindrical geometry, we compare our results against the axisymmetric solution for five different wall thickness-to-radius ratios. Subsequently, a comparison of these approaches is presented for an idealized arterial bifurcation having regionally varying wall thickness. Lastly, we compare results for a subject-specific mouse geometry with regionally varying material properties and wall thickness. External tissue support boundary conditions model the effect of perivascular tissue. Based on our results, the rotation-free shell formulation represents the most advantageous compromise between computational cost and accuracy for large scale vascular biomechanics applications that include complex geometries., C.A. Figueroa acknowledges the support from National Institutes of Health, United States (R01-HL158723 and U01HL135842) and by the Edward B. Diethrich M.D. Professorship. Furthermore, Dr. Figueroa acknowledges (R01 HL105297) with J.D. Humphrey. N. Nama acknowledges the support from the American Heart Association, United States (23CDA1048343). R. Ortigosa acknowledges support of grant PID2022- 141957OA-C22 funded by MICIU/AEI/10.13039/501100011033 and by ERDF A way of making Europe, and also the support of grant 21996/PI/22 funded by Fundacion Seneca - Agencia de Ciencia y Tecnologia de la Region de Murcia. M. Aguirre acknowledges the support of Grant PID2022-136668OA-I00 funded by MCIN/AEI/10.13039/ 501100011033 and by ERDF A way of making Europe, and by grant RYC2023-042592-I funded by MICIU/AEI/10.13039/501100011033 and by ESF+. A.J. Gil acknowledges the support of The UK Leverhulme Trust through a Leverhulme Fellowship., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Machine learning and statistical shape modelling for real-time prediction of stent deployment in realistic anatomies

  • Bisighini, Beatrice
  • Aguirre Font, Miquel
  • Pierrat, Baptiste
  • Avril, Stephane
Background and Objective. The rise in minimally invasive procedures has created a demand for efficient and reliable planning software to predict intra- and post-operative outcomes. Surrogate modelling has shown promise, but challenges remain, particularly in cardiovascular applications, due to the complexity of parametrising anatomical structures and the need for large training datasets. This study aims to apply statistical shape modelling and machine learning for predicting stent deployment in real time using patient-specific models. Methods. We built a statistical shape model starting from an open-source clinical dataset, which we then used to generate new synthetic cases. Finite element simulations of stent deployment were performed on these cases using an in-house software. A surrogate model was then trained to map the statistical features of the synthetic models to the corresponding stent configurations, evaluating sensitivity to dataset size. Results. Even with the smallest dataset (400 samples), the average prediction error in position among the tested cases never exceeded 8.6%, with a median one within the testing dataset of 1.6%. As the number of training samples increased (4900), we achieved a median position error lower than 0.1 mm (0.97%) and a maximum position error of 0.5 mm (4.8%). Notably, the largest errors occur in the radial direction of the stent, while the deployed length is accurately predicted in all the cases. Conclusions. The consistent success in performance strongly suggests that surrogate modelling represents a clinically valuable tool for accurately computing stent deployment outcomes in real time, even within complex anatomical scenarios., This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement 859836 and the French region Auvergne-Rhône-Alpes under the “Ambition Internationale” project. Miquel Aguirre acknowledges the support of Grant PID2022-136668OA-I00 funded by MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe and a María Zambrano research fellowship at Universitat Politècnica de Catalunya granted by Ministerio de Universidades ., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A systematic comparison of membrane, shell, and 3D solid formulations for nonlinear vascular biomechanics

  • Kim, Taeouk
  • Ortigosa Martínez, Rogelio
  • Nama, Nitesh
  • Aguirre Font, Miquel
  • Gil, Antonio J.
  • Humphrey, Jay D.
  • Figueroa, C. Alberto
Typical computational methods for vascular biosolid mechanics represent the blood vessel wall as a membrane, shell, or 3D solid. Each of these formulations has advantages and disadvantages concerning accuracy, ease of implementation, and computational costs. Despite the widespread use of these formulations, a systematic comparison of the performance and accuracy of these formulations for nonlinear vascular biomechanics has remained wanting. Therefore, the decision regarding the optimal choice often relies on intuition or previous experience, with unclear consequences of choosing one approach over the other. Here, we present a systematic comparison among three different formulations to represent the vessel wall as: (i) a nonlinear membrane, (ii) a nonlinear, rotation-free shell, and (iii) a nonlinear 3D solid. For the 3D solid model, we consider two different implementations employing linear and quadratic interpolation. Convergence analysis for displacement and stress are presented for all formulations. We compare results in both idealized and subject-specific mouse aortic geometries. For the idealized cylindrical geometry, we compare our results against the axisymmetric solution for five different wall thickness-to-radius ratios. Subsequently, a comparison of these approaches is presented for an idealized arterial bifurcation having regionally varying wall thickness. Lastly, we compare results for a subject-specific mouse geometry with regionally varying material properties and wall thickness. External tissue support boundary conditions model the effect of perivascular tissue. Based on our results, the rotation-free shell formulation represents the most advantageous compromise between computational cost and accuracy for large scale vascular biomechanics applications that include complex geometries., C.A. Figueroa acknowledges the support from National Institutes of Health, United States (R01-HL158723 and U01HL135842) and by the Edward B. Diethrich M.D. Professorship. Furthermore, Dr. Figueroa acknowledges (R01 HL105297) with J.D. Humphrey. N. Nama acknowledges the support from the American Heart Association, United States (23CDA1048343). R. Ortigosa acknowledges support of grant PID2022- 141957OA-C22 funded by MICIU/AEI/10.13039/501100011033 and by ERDF A way of making Europe, and also the support of grant 21996/PI/22 funded by Fundacion Seneca - Agencia de Ciencia y Tecnologia de la Region de Murcia. M. Aguirre acknowledges the support of Grant PID2022-136668OA-I00 funded by MCIN/AEI/10.13039/ 501100011033 and by ERDF A way of making Europe, and by grant RYC2023-042592-I funded by MICIU/AEI/10.13039/501100011033 and by ESF+. A.J. Gil acknowledges the support of The UK Leverhulme Trust through a Leverhulme Fellowship., Peer Reviewed, Postprint (published version)