東華大學圖書館 |

Modeling Neuron Material Transport Using Isogeometric Analysis, Deep Learning and PDE-Constrained Optimization.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Modeling Neuron Material Transport Using Isogeometric Analysis, Deep Learning and PDE-Constrained Optimization./
作者:	Li, Angran.
面頁冊數:	1 online resource (143 pages)
附註:	Source: Dissertations Abstracts International, Volume: 84-04, Section: B.
Contained By:	Dissertations Abstracts International84-04B.
標題:	Mechanical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29320297click for full text (PQDT)
ISBN:	9798351459530

Modeling Neuron Material Transport Using Isogeometric Analysis, Deep Learning and PDE-Constrained Optimization.
Li, Angran.

Modeling Neuron Material Transport Using Isogeometric Analysis, Deep Learning and PDE-Constrained Optimization. - 1 online resource (143 pages)

Source: Dissertations Abstracts International, Volume: 84-04, Section: B.

Thesis (Ph.D.)--Carnegie Mellon University, 2022.

Includes bibliographical references

The neuron exhibits a highly polarized structure that typically consists of a single long axon and many dendrites which are both extended from its cell body. Since most of the materials necessary for the neuron are synthesized in the cell body, they need to experience long-distance transport in axons or dendrites to reach their effective location. The intracellular material transport is therefore especially crucial to ensure necessary materials are delivered to the right locations for the development, function, and survival of neuron cells. The disruption of intracellular transport can lead to the abnormal accumulations of certain cellular material and extreme swelling of the axon, which have been observed in many neurological and neurodegenerative diseases such as Huntington's, Parkinson's, and Alzheimer's diseases. Therefore, it is essential to study and understand mechanisms of the transport function and dysfunction.We first develop an isogeometric analysis (IGA) based platform for material transport simulation in neurite networks. The transport process is governed by a generalized 3D motor-assisted transport model that couples reaction-diffusion-transport equations with Navier-Stokes equations. Given any neuronal morphology, we applied the skeleton-based sweeping method to generate hexahedral control mesh of the neuron and built hierarchical tricubic B-splines (THB-spline3D) on the hexahedral mesh to accurately represent the geometry. We then adopt IGA to solve partial differential equations (PDEs) of the transport model in different neuron geometries. We discover several spatial patterns of the transport process from the result of our study, which provides key insights into how material transport in neurite networks is mediated by their complex geometry. The developed IGA solver has been tested with multiple complex and representative neurite networks that prove its capability to perform robust large-scale simulation in supercomputer. While the IGA solver can provide high-fidelity simulation results, the high computational cost to perform 3D simulations has limited its application in the biomedical field when fast feedback from the computer simulation is needed. To address this issue, we develop a graph neural network (GNN) based surrogate model to boost the efficiency of the material transport simulation by learning the underlying physics behind the simulation data. The use of GNN is motivated by the extensive morphologies of neurite networks and the IGA simulation data stored in mesh structure. Moreover, the GNN model can achieve better computational efficiency than IGA simulation without sacrificing too much geometry information of neuron. To ensure the model is applicable to any neurite geometry, we build a graph representation of neurons by decomposing the neuron geometry into two basic structures: pipe and bifurcation. Different GNN simulators are designed for these two basic structures to predict the spatiotemporal concentration distribution given input simulation parameters and boundary conditions. Specifically, we add the residual terms from PDEs to instruct the model to learn the physics behind simulation data. To reconstruct the neurite network, a GNN-based assembly model is used to combine all the pipes and bifurcations following the graph representation. The loss function of the assembly model is designed to impose consistent concentration results on the interface between pipe and bifurcation. The well-trained GNN model can predict the dynamical concentration change during the transport process with an average error less than 10% and 120~330 times faster compared to IGA simulations. Besides the endeavor to integrate IGA simulation with machine learning, we also improve the transport model to further study the traffic control mechanism and explain the traffic jam formation during the transport process. We have developed a novel PDE-constrained optimization model (PDE-CO) with a designed optimization function constrained by the motor-assisted transport model. The transport is controlled to avoid traffic jam of materials by minimizing a pre-defined objective function. The optimization subjects to a set of PDE constraints that describe the material transport process based on the molecular-motor-assisted transport model of intracellular particles. The proposed PDE-CO model is solved in complex 2D and 3D tree structures by using IGA. Different simulation parameters are used to introduce traffic jams and study how neurons handle the transport issue. Specifically, we successfully model and explain the traffic jam caused by reduced number of microtubules (MTs) and MT swirls. Our model effectively simulates the material transport process in healthy neurons and also explains the formation of a traffic jam in abnormal neurons. The results demonstrate that both geometry and MT structure play important roles in achieving an optimal transport process in neuron. Based on the PDE-CO model, we develop a novel IGA-based physics-informed graph neural network (PGNN) to quickly predict normal and abnormal transport phenomena such as traffic jam in different neuron geometries. In particular, the proposed method learns from the IGA simulation of the intracellular transport process and provides accurate material concentration prediction of normal transport and MT-induced traffic jam. The IGA-based PGNN model contains simulators to handle local prediction of both normal and two MT-induced traffic jams in pipes, as well as another simulator to predict normal transport in bifurcations. Bezier extraction is adopted to incorporate the geometry information into the simulators to accurately compute the physics informed loss function with PDE residuals. The well-trained model effectively predicts the distribution of transport velocity and material concentration during traffic jam and normal transport with an average error less than 10% compared to IGA simulations.In summary, we have successfully simulated and explained the intracellular material transport within complex neuron geometries using several methods including IGA, deep learning and PDE-CO. We have provided the velocity and concentration results for 2D and 3D neuron geometries, and made comparison between normal and traffic jam transport conditions. These results have shown the significant impact of geometry and MTs on the transport process. We have also proposed the deep learning model to provide fast and accurate transport prediction, and get a better insight into the complex pattern of material distribution in different local regions within neuron.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798351459530Subjects--Topical Terms:

649730
Mechanical engineering.
Subjects--Index Terms:

Neuron material transportIndex Terms--Genre/Form:

542853
Electronic books.

Modeling Neuron Material Transport Using Isogeometric Analysis, Deep Learning and PDE-Constrained Optimization.
LDR:08086nmm a2200421K 4500 001 2358329
005 20230731112619.5
006 m o d
007 cr mn ---uuuuu
008 241011s2022 xx obm 000 0 eng d
020 $a 9798351459530
035 $a (MiAaPQ)AAI29320297
035 $a AAI29320297
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Li, Angran. $3 3698864
245 1 0 $a Modeling Neuron Material Transport Using Isogeometric Analysis, Deep Learning and PDE-Constrained Optimization.
264 0 $c 2022
300 $a 1 online resource (143 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 84-04, Section: B.
500 $a Advisor: Zhang, Yongjie Jessica.
502 $a Thesis (Ph.D.)--Carnegie Mellon University, 2022.
504 $a Includes bibliographical references
520 $a The neuron exhibits a highly polarized structure that typically consists of a single long axon and many dendrites which are both extended from its cell body. Since most of the materials necessary for the neuron are synthesized in the cell body, they need to experience long-distance transport in axons or dendrites to reach their effective location. The intracellular material transport is therefore especially crucial to ensure necessary materials are delivered to the right locations for the development, function, and survival of neuron cells. The disruption of intracellular transport can lead to the abnormal accumulations of certain cellular material and extreme swelling of the axon, which have been observed in many neurological and neurodegenerative diseases such as Huntington's, Parkinson's, and Alzheimer's diseases. Therefore, it is essential to study and understand mechanisms of the transport function and dysfunction.We first develop an isogeometric analysis (IGA) based platform for material transport simulation in neurite networks. The transport process is governed by a generalized 3D motor-assisted transport model that couples reaction-diffusion-transport equations with Navier-Stokes equations. Given any neuronal morphology, we applied the skeleton-based sweeping method to generate hexahedral control mesh of the neuron and built hierarchical tricubic B-splines (THB-spline3D) on the hexahedral mesh to accurately represent the geometry. We then adopt IGA to solve partial differential equations (PDEs) of the transport model in different neuron geometries. We discover several spatial patterns of the transport process from the result of our study, which provides key insights into how material transport in neurite networks is mediated by their complex geometry. The developed IGA solver has been tested with multiple complex and representative neurite networks that prove its capability to perform robust large-scale simulation in supercomputer. While the IGA solver can provide high-fidelity simulation results, the high computational cost to perform 3D simulations has limited its application in the biomedical field when fast feedback from the computer simulation is needed. To address this issue, we develop a graph neural network (GNN) based surrogate model to boost the efficiency of the material transport simulation by learning the underlying physics behind the simulation data. The use of GNN is motivated by the extensive morphologies of neurite networks and the IGA simulation data stored in mesh structure. Moreover, the GNN model can achieve better computational efficiency than IGA simulation without sacrificing too much geometry information of neuron. To ensure the model is applicable to any neurite geometry, we build a graph representation of neurons by decomposing the neuron geometry into two basic structures: pipe and bifurcation. Different GNN simulators are designed for these two basic structures to predict the spatiotemporal concentration distribution given input simulation parameters and boundary conditions. Specifically, we add the residual terms from PDEs to instruct the model to learn the physics behind simulation data. To reconstruct the neurite network, a GNN-based assembly model is used to combine all the pipes and bifurcations following the graph representation. The loss function of the assembly model is designed to impose consistent concentration results on the interface between pipe and bifurcation. The well-trained GNN model can predict the dynamical concentration change during the transport process with an average error less than 10% and 120~330 times faster compared to IGA simulations. Besides the endeavor to integrate IGA simulation with machine learning, we also improve the transport model to further study the traffic control mechanism and explain the traffic jam formation during the transport process. We have developed a novel PDE-constrained optimization model (PDE-CO) with a designed optimization function constrained by the motor-assisted transport model. The transport is controlled to avoid traffic jam of materials by minimizing a pre-defined objective function. The optimization subjects to a set of PDE constraints that describe the material transport process based on the molecular-motor-assisted transport model of intracellular particles. The proposed PDE-CO model is solved in complex 2D and 3D tree structures by using IGA. Different simulation parameters are used to introduce traffic jams and study how neurons handle the transport issue. Specifically, we successfully model and explain the traffic jam caused by reduced number of microtubules (MTs) and MT swirls. Our model effectively simulates the material transport process in healthy neurons and also explains the formation of a traffic jam in abnormal neurons. The results demonstrate that both geometry and MT structure play important roles in achieving an optimal transport process in neuron. Based on the PDE-CO model, we develop a novel IGA-based physics-informed graph neural network (PGNN) to quickly predict normal and abnormal transport phenomena such as traffic jam in different neuron geometries. In particular, the proposed method learns from the IGA simulation of the intracellular transport process and provides accurate material concentration prediction of normal transport and MT-induced traffic jam. The IGA-based PGNN model contains simulators to handle local prediction of both normal and two MT-induced traffic jams in pipes, as well as another simulator to predict normal transport in bifurcations. Bezier extraction is adopted to incorporate the geometry information into the simulators to accurately compute the physics informed loss function with PDE residuals. The well-trained model effectively predicts the distribution of transport velocity and material concentration during traffic jam and normal transport with an average error less than 10% compared to IGA simulations.In summary, we have successfully simulated and explained the intracellular material transport within complex neuron geometries using several methods including IGA, deep learning and PDE-CO. We have provided the velocity and concentration results for 2D and 3D neuron geometries, and made comparison between normal and traffic jam transport conditions. These results have shown the significant impact of geometry and MTs on the transport process. We have also proposed the deep learning model to provide fast and accurate transport prediction, and get a better insight into the complex pattern of material distribution in different local regions within neuron.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Mechanical engineering. $3 649730
650 4 $a Neurosciences. $3 588700
650 4 $a Biomedical engineering. $3 535387
650 4 $a Biomechanics. $3 548685
650 4 $a Cellular biology. $3 3172791
653 $a Neuron material transport
653 $a Isogeometric
653 $a Deep learning
653 $a Partial differential equation
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0548
690 $a 0541
690 $a 0800
690 $a 0317
690 $a 0648
690 $a 0379
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a Carnegie Mellon University. $b Mechanical Engineering. $3 2096240
773 0 $t Dissertations Abstracts International $g 84-04B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29320297 $z click for full text (PQDT)