東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Methods for Multidimensional Functional Data Analysis in Modern Neuroimaging.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Methods for Multidimensional Functional Data Analysis in Modern Neuroimaging./
作者:	Consagra, William.
面頁冊數:	1 online resource (206 pages)
附註:	Source: Dissertations Abstracts International, Volume: 84-03, Section: B.
Contained By:	Dissertations Abstracts International84-03B.
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29320156click for full text (PQDT)
ISBN:	9798845421319

Methods for Multidimensional Functional Data Analysis in Modern Neuroimaging.
Consagra, William.

Methods for Multidimensional Functional Data Analysis in Modern Neuroimaging. - 1 online resource (206 pages)

Source: Dissertations Abstracts International, Volume: 84-03, Section: B.

Thesis (Ph.D.)--University of Rochester, 2022.

Includes bibliographical references

Continued advances in brain imaging techniques hold great promise to bolster our understanding of the structure and function of the human brain. Diffusion magnetic resonance imaging (MRI) is one such popular technique which measures the micro-structure of neural tissue in the white matter. Diffusion MRI data can be analyzed in a variety of ways; from voxel level summaries to the reconstruction of large-scale neural pathways called white matter fiber tracts, the collection of which is referred to as the structural connectome. Many of these datasets can be considered as functional data defined over multidimensional and/or non-Euclidean domains. This dissertation addresses the development of new methodologies for multidimensional functional data analysis (FDA) of diffusion MRI datasets for applications in brain structural analysis. In Chapter 2, we propose a novel methodology for learning a set of data-driven basis functions for the representation of multidimensional functional data that is immune to several manifestations of the curse of dimensionality. We prove that estimating these basis functions is equivalent to the tensor decomposition of a carefully defined reduction transformation of the observed data. The advantages of the proposed method over competing methods are demonstrated in a simulation study, and we illustrate our method on a clinical diffusion MRI dataset from a traumatic brain injury study.Chapter 3 proposes a new approach to the optimal design of sparse high angular resolution diffusion imaging (HARDI) acquisition protocols and resulting diffusion signal estimation. Our method leverages relevant historical diffusion MRI data to build empirical priors which guide both the selection of the optimal design and facilitate the construction of an estimator of the local diffusion profile that performs well in sparse samples. Simulation studies and real data experiments using the Human Connectome Project data demonstrate significant advantages over the existing HARDI sampling and analysis frameworks in tasks related to human brain structural connectome reconstruction. In Chapter 4, we propose a continuous framework to characterize the population-level variability of brain structural connectivity. Our framework assumes the observed white matter fiber tract endpoints are driven by a latent random function defined over a product manifold domain. To overcome the computational challenges of analyzing such complex latent functions, we adapt techniques from Chapter 2 to develop an efficient algorithm to construct a data-driven reduced-rank function space to represent the latent continuous connectivity. Using real data from the Human Connectome Project, we show that our method outperforms several state-of-the-art approaches on connectivity analysis tasks of interest.

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

Mode of access: World Wide Web

ISBN: 9798845421319Subjects--Topical Terms:

517247
Statistics.
Subjects--Index Terms:

Computational neuroscienceIndex Terms--Genre/Form:

542853
Electronic books.

Methods for Multidimensional Functional Data Analysis in Modern Neuroimaging.
LDR:04226nmm a2200397K 4500 001 2357470
005 20230710070043.5
006 m o d
007 cr mn ---uuuuu
008 241011s2022 xx obm 000 0 eng d
020 $a 9798845421319
035 $a (MiAaPQ)AAI29320156
035 $a AAI29320156
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Consagra, William. $3 3698001
245 1 0 $a Methods for Multidimensional Functional Data Analysis in Modern Neuroimaging.
264 0 $c 2022
300 $a 1 online resource (206 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: 84-03, Section: B.
500 $a Advisor: Qiu, Xing.
502 $a Thesis (Ph.D.)--University of Rochester, 2022.
504 $a Includes bibliographical references
520 $a Continued advances in brain imaging techniques hold great promise to bolster our understanding of the structure and function of the human brain. Diffusion magnetic resonance imaging (MRI) is one such popular technique which measures the micro-structure of neural tissue in the white matter. Diffusion MRI data can be analyzed in a variety of ways; from voxel level summaries to the reconstruction of large-scale neural pathways called white matter fiber tracts, the collection of which is referred to as the structural connectome. Many of these datasets can be considered as functional data defined over multidimensional and/or non-Euclidean domains. This dissertation addresses the development of new methodologies for multidimensional functional data analysis (FDA) of diffusion MRI datasets for applications in brain structural analysis. In Chapter 2, we propose a novel methodology for learning a set of data-driven basis functions for the representation of multidimensional functional data that is immune to several manifestations of the curse of dimensionality. We prove that estimating these basis functions is equivalent to the tensor decomposition of a carefully defined reduction transformation of the observed data. The advantages of the proposed method over competing methods are demonstrated in a simulation study, and we illustrate our method on a clinical diffusion MRI dataset from a traumatic brain injury study.Chapter 3 proposes a new approach to the optimal design of sparse high angular resolution diffusion imaging (HARDI) acquisition protocols and resulting diffusion signal estimation. Our method leverages relevant historical diffusion MRI data to build empirical priors which guide both the selection of the optimal design and facilitate the construction of an estimator of the local diffusion profile that performs well in sparse samples. Simulation studies and real data experiments using the Human Connectome Project data demonstrate significant advantages over the existing HARDI sampling and analysis frameworks in tasks related to human brain structural connectome reconstruction. In Chapter 4, we propose a continuous framework to characterize the population-level variability of brain structural connectivity. Our framework assumes the observed white matter fiber tract endpoints are driven by a latent random function defined over a product manifold domain. To overcome the computational challenges of analyzing such complex latent functions, we adapt techniques from Chapter 2 to develop an efficient algorithm to construct a data-driven reduced-rank function space to represent the latent continuous connectivity. Using real data from the Human Connectome Project, we show that our method outperforms several state-of-the-art approaches on connectivity analysis tasks of interest.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Statistics. $3 517247
650 4 $a Biostatistics. $3 1002712
653 $a Computational neuroscience
653 $a Connectomics
653 $a Functional data analysis
653 $a High-dimensional statistics
653 $a Low-rank modeling
653 $a Machine learning
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0463
690 $a 0308
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a University of Rochester. $b School of Medicine and Dentistry. $3 3174447
773 0 $t Dissertations Abstracts International $g 84-03B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29320156 $z click for full text (PQDT)