A deep learning application to approximate the geometric orifice and coaptation areas of the polymeric heart valves under time – varying transvalvular pressure

Gülbulak, Utku; Gecgel, Ozhan; Ertaş, A.

doi:10.1016/j.jmbbm.2021.104371

Cited by 11 publications

(7 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Furthermore, in the presented sequence of "generationmodeling -optimizer", we incorporated an element of surrogate FEM based on ML, which aims to accelerate the execution of Frontiers in Bioengineering and Biotechnology frontiersin.org numerical calculations, which becomes a bottleneck in cases with thousands of geometries in FEM. Such an approach for predicting the stress-strain state of the leaflet and its coaptation area (Balu et al, 2019) or the geometric orifice area (Gulbulak et al, 2021) has already demonstrated its validity. In this study, FEM modeling of valve biomechanics has been adapted through the integration of ML techniques.…”

Section: Discussionmentioning

confidence: 99%

“…FEM is particularly advanced among numerical simulation tools and has been demonstrated to be highly efficient in modeling PHVs ( Borazjani, 2013 ; Gilmanov and Sotiropoulos, 2016 ; Castravete et al, 2020 ; Lee et al, 2020 ; Chen et al, 2022 ). Furthermore, existing research has demonstrated advanced optimization algorithms that enable the semi-automatic generation and FEM analysis of significant quantities of PHVs, facilitating the selection of optimal candidates among them ( Hsu et al, 2015 ; Li and Sun, 2017 ; Abbasi and Azadani, 2020 ; Gulbulak et al, 2021 ). Among the most sophisticated approaches in this context, we consider the work of Travaglino et al ( Travaglino et al, 2020 ).…”

Section: Discussionmentioning

confidence: 99%

“…ML approach. To overcome the limitations of FEM, some authors have proposed to replace it with a ML based surrogate method ( Nallagonda, 2018 ; Balu et al, 2019 ; Liang and Sun, 2019 ; Gulbulak et al, 2021 ). This involves imitating the numerical modeling process with ML algorithms to quickly obtain results.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Perfect prosthetic heart valve: generative design with machine learning, modeling, and optimization

Danilov,

Klyshnikov,

Onishenko

et al. 2023

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Majority of modern techniques for creating and optimizing the geometry of medical devices are based on a combination of computer-aided designs and the utility of the finite element method This approach, however, is limited by the number of geometries that can be investigated and by the time required for design optimization. To address this issue, we propose a generative design approach that combines machine learning (ML) methods and optimization algorithms. We evaluate eight different machine learning methods, including decision tree-based and boosting algorithms, neural networks, and ensembles. For optimal design, we investigate six state-of-the-art optimization algorithms, including Random Search, Tree-structured Parzen Estimator, CMA-ES-based algorithm, Nondominated Sorting Genetic Algorithm, Multiobjective Tree-structured Parzen Estimator, and Quasi-Monte Carlo Algorithm. In our study, we apply the proposed approach to study the generative design of a prosthetic heart valve (PHV). The design constraints of the prosthetic heart valve, including spatial requirements, materials, and manufacturing methods, are used as inputs, and the proposed approach produces a final design and a corresponding score to determine if the design is effective. Extensive testing leads to the conclusion that utilizing a combination of ensemble methods in conjunction with a Tree-structured Parzen Estimator or a Nondominated Sorting Genetic Algorithm is the most effective method in generating new designs with a relatively low error rate. Specifically, the Mean Absolute Percentage Error was found to be 11.8% and 10.2% for lumen and peak stress prediction respectively. Furthermore, it was observed that both optimization techniques result in design scores of approximately 95%. From both a scientific and applied perspective, this approach aims to select the most efficient geometry with given input parameters, which can then be prototyped and used for subsequent in vitro experiments. By proposing this approach, we believe it will replace or complement CAD-FEM-based modeling, thereby accelerating the design process and finding better designs within given constraints. The repository, which contains the essential components of the study, including curated source code, dataset, and trained models, is publicly available at https://github.com/ViacheslavDanilov/generative_design.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Perfect prosthetic heart valve: generative design with machine learning, modeling, and optimization

Danilov,

Klyshnikov,

Onishenko

et al. 2023

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

show abstract

“…Machine learning (ML), especially deep (machine) learning using deep neural networks (DNNs) [1], has the potential to transform the biomedical and healthcare fields by enabling more personalized and accurate diagnoses and prognoses, treatment recommendations, and disease prevention strategies [2]. For example, in the cardiovascular domain, ML techniques have been developed for various applications, such as automated ECG signal analysis [3], automated geometry reconstruction from medical images [4], tissue modeling [5][6][7][8], heart valve analysis [9], disease modeling and diagnosis [10][11][12].…”

Section: Introductionmentioning

confidence: 99%

Synergistic Integration of Deep Neural Networks and Finite Element Method with Applications for Biomechanical Analysis of Human Aorta

Liang

Liu

Elefteriades

et al. 2023

Preprint

View full text Add to dashboard Cite

Motivation: Patient-specific finite element analysis (FEA) has the potential to aid in the prognosis of cardiovascular diseases by providing accurate stress and deformation analysis in various scenarios. It is known that patient-specific FEA is time-consuming and unsuitable for time-sensitive clinical applications. To mitigate this challenge, machine learning (ML) techniques, including deep neural networks (DNNs), have been developed to construct fast FEA surrogates. However, due to the data-driven nature of these ML models, they may not generalize well on new data, leading to unacceptable errors. Methods: We propose a synergistic integration of DNNs and finite element method (FEM) to overcome each other's limitations. We demonstrated this novel integrative strategy in forward and inverse problems. For the forward problem, we developed DNNs using state-of-the-art architectures, and DNN outputs were then refined by FEM to ensure accuracy. For the inverse problem of heterogeneous material parameter identification, our method employs a DNN as regularization for the inverse analysis process to avoid erroneous material parameter distribution. Results: We tested our methods on biomechanical analysis of the human aorta. For the forward problem, the DNN-only models yielded acceptable stress errors in majority of test cases; yet, for some test cases that could be out of the training distribution (OOD), the peak stress errors were larger than 50%. The DNN-FEM integration eliminated the large errors for these OOD cases. Moreover, the DNN-FEM integration was magnitudes faster than the FEM-only approach. For the inverse problem, the FEM-only inverse method led to errors larger than 50%, and our DNN-FEM integration significantly improved performance on the inverse problem with errors less than 1%.

show abstract

“…Данный способ основан на анализе каждой модели экспертом. Второй подход -применение методов машинного обучения как для разработки створчатого аппарата [4], так и оценки возникающих напряжений при его деформации [5]. Преимущества второго варианта в том, что надлежащим образом настроенные искусственные нейронные сети могут находить взаимосвязи там, где это не всегда очевидно, для чего необходима обучающая выборка большого размера.…”

Section: Introductionunclassified

The concept of automated functional design of heart valve prostheses

Onishchenko

Klyshnikov

Резвова

et al. 2021

Kompleks. probl. serdečno-sosud. zabol.

View full text Add to dashboard Cite

Aim. To develop an algorithm for the automated functional design of the heart valve leaflet apparatus.Methods. The geometry of the aortic valve leaflet was designed in the Matlab programming environment (MathWorks, Massachusetts, USA). Numerical modeling of the opening process was performed using Abaqus/CAE (Dassault Systemes, France).Results. We developed an algorithm, with the help of which a set of models of the leaflet apparatus was designed. 8 models were subjected to numerical modeling of the stress-strain state. The locking pressure simulation has shown that the smallest von Mises stress value was recorded for a sample with a larger surface area of the leaflet belly and it equals 0.422 MPa. The results obtained show that the value of the radius of curvature significantly affects the behavior of the entire valve, which leads to the conclusion that it is necessary to carefully select the design of the valve apparatus for its correct functioning.Conclusion. The study provides the primary confirmation that the concept of the algorithm is efficient for the automated functional design of the aortic heart valve leaflet apparatus.

show abstract

A deep learning application to approximate the geometric orifice and coaptation areas of the polymeric heart valves under time – varying transvalvular pressure

Cited by 11 publications

References 43 publications

Perfect prosthetic heart valve: generative design with machine learning, modeling, and optimization

Perfect prosthetic heart valve: generative design with machine learning, modeling, and optimization

Synergistic Integration of Deep Neural Networks and Finite Element Method with Applications for Biomechanical Analysis of Human Aorta

The concept of automated functional design of heart valve prostheses

Contact Info

Product

Resources

About