Comparison of nine tractography algorithms for detecting abnormal structural brain networks in Alzheimer's disease

for the Alzheimer's Disease Neuroimaging Initiative

Research output: Contribution to journalArticle

47 Citations (Scopus)

Abstract

Alzheimer's disease (AD) involves a gradual breakdown of brain connectivity, and network analyses offer a promising new approach to track and understand disease progression. Even so, our ability to detect degenerative changes in brain networks depends on the methods used. Here we compared several tractography and feature extraction methods to see which ones gave best diagnostic classification for 202 people with AD, mild cognitive impairment or normal cognition, scanned with 41-gradient diffusion-weighted magnetic resonance imaging as part of the Alzheimer's Disease Neuroimaging Initiative (ADNI) project. We computed brain networks based on whole brain tractography with nine different methods - four of them tensor-based deterministic (FACT, RK2, SL, and TL), two orientation distribution function (ODF)-based deterministic (FACT, RK2), two ODF-based probabilistic approaches (Hough and PICo), and one "ball-and-stick" approach (Probtrackx). Brain networks derived from different tractography algorithms did not differ in terms of classification performance on ADNI, but performing principal components analysis on networks helped classification in some cases. Small differences may still be detectable in a truly vast cohort, but these experiments help assess the relative advantages of different tractography algorithms, and different post-processing choices, when used for classification.

Original languageEnglish (US)
Article number48
JournalFrontiers in Aging Neuroscience
Volume7
Issue numberAPR
DOIs
StatePublished - 2015

Fingerprint

Alzheimer Disease
Brain
Neuroimaging
Diffusion Magnetic Resonance Imaging
Aptitude
Principal Component Analysis
Cognition
Disease Progression

Keywords

  • Alzheimer's disease
  • Brain network
  • Classification
  • Diffusion MRI
  • GLRAM
  • PCA
  • Tractography

ASJC Scopus subject areas

  • Aging
  • Cognitive Neuroscience

Cite this

Comparison of nine tractography algorithms for detecting abnormal structural brain networks in Alzheimer's disease. / for the Alzheimer's Disease Neuroimaging Initiative.

In: Frontiers in Aging Neuroscience, Vol. 7, No. APR, 48, 2015.

Research output: Contribution to journalArticle

@article{84fdb6155a5a428d8221f91ecc96c56b,
title = "Comparison of nine tractography algorithms for detecting abnormal structural brain networks in Alzheimer's disease",
abstract = "Alzheimer's disease (AD) involves a gradual breakdown of brain connectivity, and network analyses offer a promising new approach to track and understand disease progression. Even so, our ability to detect degenerative changes in brain networks depends on the methods used. Here we compared several tractography and feature extraction methods to see which ones gave best diagnostic classification for 202 people with AD, mild cognitive impairment or normal cognition, scanned with 41-gradient diffusion-weighted magnetic resonance imaging as part of the Alzheimer's Disease Neuroimaging Initiative (ADNI) project. We computed brain networks based on whole brain tractography with nine different methods - four of them tensor-based deterministic (FACT, RK2, SL, and TL), two orientation distribution function (ODF)-based deterministic (FACT, RK2), two ODF-based probabilistic approaches (Hough and PICo), and one {"}ball-and-stick{"} approach (Probtrackx). Brain networks derived from different tractography algorithms did not differ in terms of classification performance on ADNI, but performing principal components analysis on networks helped classification in some cases. Small differences may still be detectable in a truly vast cohort, but these experiments help assess the relative advantages of different tractography algorithms, and different post-processing choices, when used for classification.",
keywords = "Alzheimer's disease, Brain network, Classification, Diffusion MRI, GLRAM, PCA, Tractography",
author = "{for the Alzheimer's Disease Neuroimaging Initiative} and Liang Zhan and Jiayu Zhou and Yalin Wang and Yan Jin and Neda Jahanshad and Gautam Prasad and Nir, {Talia M.} and Leonardo, {Cassandra D.} and Jieping Ye and Thompson, {Paul M.}",
year = "2015",
doi = "10.3389/fnagi.2015.00048",
language = "English (US)",
volume = "7",
journal = "Frontiers in Aging Neuroscience",
issn = "1663-4365",
publisher = "Frontiers Research Foundation",
number = "APR",

}

TY - JOUR

T1 - Comparison of nine tractography algorithms for detecting abnormal structural brain networks in Alzheimer's disease

AU - for the Alzheimer's Disease Neuroimaging Initiative

AU - Zhan, Liang

AU - Zhou, Jiayu

AU - Wang, Yalin

AU - Jin, Yan

AU - Jahanshad, Neda

AU - Prasad, Gautam

AU - Nir, Talia M.

AU - Leonardo, Cassandra D.

AU - Ye, Jieping

AU - Thompson, Paul M.

PY - 2015

Y1 - 2015

N2 - Alzheimer's disease (AD) involves a gradual breakdown of brain connectivity, and network analyses offer a promising new approach to track and understand disease progression. Even so, our ability to detect degenerative changes in brain networks depends on the methods used. Here we compared several tractography and feature extraction methods to see which ones gave best diagnostic classification for 202 people with AD, mild cognitive impairment or normal cognition, scanned with 41-gradient diffusion-weighted magnetic resonance imaging as part of the Alzheimer's Disease Neuroimaging Initiative (ADNI) project. We computed brain networks based on whole brain tractography with nine different methods - four of them tensor-based deterministic (FACT, RK2, SL, and TL), two orientation distribution function (ODF)-based deterministic (FACT, RK2), two ODF-based probabilistic approaches (Hough and PICo), and one "ball-and-stick" approach (Probtrackx). Brain networks derived from different tractography algorithms did not differ in terms of classification performance on ADNI, but performing principal components analysis on networks helped classification in some cases. Small differences may still be detectable in a truly vast cohort, but these experiments help assess the relative advantages of different tractography algorithms, and different post-processing choices, when used for classification.

AB - Alzheimer's disease (AD) involves a gradual breakdown of brain connectivity, and network analyses offer a promising new approach to track and understand disease progression. Even so, our ability to detect degenerative changes in brain networks depends on the methods used. Here we compared several tractography and feature extraction methods to see which ones gave best diagnostic classification for 202 people with AD, mild cognitive impairment or normal cognition, scanned with 41-gradient diffusion-weighted magnetic resonance imaging as part of the Alzheimer's Disease Neuroimaging Initiative (ADNI) project. We computed brain networks based on whole brain tractography with nine different methods - four of them tensor-based deterministic (FACT, RK2, SL, and TL), two orientation distribution function (ODF)-based deterministic (FACT, RK2), two ODF-based probabilistic approaches (Hough and PICo), and one "ball-and-stick" approach (Probtrackx). Brain networks derived from different tractography algorithms did not differ in terms of classification performance on ADNI, but performing principal components analysis on networks helped classification in some cases. Small differences may still be detectable in a truly vast cohort, but these experiments help assess the relative advantages of different tractography algorithms, and different post-processing choices, when used for classification.

KW - Alzheimer's disease

KW - Brain network

KW - Classification

KW - Diffusion MRI

KW - GLRAM

KW - PCA

KW - Tractography

UR - http://www.scopus.com/inward/record.url?scp=84992236271&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992236271&partnerID=8YFLogxK

U2 - 10.3389/fnagi.2015.00048

DO - 10.3389/fnagi.2015.00048

M3 - Article

AN - SCOPUS:84992236271

VL - 7

JO - Frontiers in Aging Neuroscience

JF - Frontiers in Aging Neuroscience

SN - 1663-4365

IS - APR

M1 - 48

ER -