Identification of progressive mild cognitive impairment patients using incomplete longitudinal MRI scans

Kim Han Thung, Chong Yaw Wee, Pew Thian Yap, Dinggang Shen

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Distinguishing progressive mild cognitive impairment (pMCI) from stable mild cognitive impairment (sMCI) is critical for identification of patients who are at risk for Alzheimer’s disease (AD), so that early treatment can be administered. In this paper, we propose a pMCI/sMCI classification framework that harnesses information available in longitudinal magnetic resonance imaging (MRI) data, which could be incomplete, to improve diagnostic accuracy. Volumetric features were first extracted from the baseline MRI scan and subsequent scans acquired after 6, 12, and 18 months. Dynamic features were then obtained using the 18th month scan as the reference and computing the ratios of feature differences for the earlier scans. Features that are linearly or non-linearly correlated with diagnostic labels are then selected using two elastic net sparse learning algorithms. Missing feature values due to the incomplete longitudinal data are imputed using a low-rank matrix completion method. Finally, based on the completed feature matrix, we build a multi-kernel support vector machine (mkSVM) to predict the diagnostic label of samples with unknown diagnostic statuses. Our evaluation indicates that a diagnosis accuracy as high as 78.2 % can be achieved when information from the longitudinal scans is used—6.6 % higher than the case using only the reference time point image. In other words, information provided by the longitudinal history of the disease improves diagnosis accuracy.

Original languageEnglish
Pages (from-to)1-17
Number of pages17
JournalBrain Structure and Function
DOIs
Publication statusAccepted/In press - 2015 Nov 24

Fingerprint

Magnetic Resonance Imaging
Alzheimer Disease
History
Learning
Cognitive Dysfunction
Therapeutics

Keywords

  • Elastic net
  • Longitudinal MRI
  • Low-rank matrix completion
  • Missing data
  • Multi-kernel learning
  • Progressive mild cognitive impairment (pMCI)

ASJC Scopus subject areas

  • Neuroscience(all)
  • Anatomy
  • Histology

Cite this

Identification of progressive mild cognitive impairment patients using incomplete longitudinal MRI scans. / Thung, Kim Han; Wee, Chong Yaw; Yap, Pew Thian; Shen, Dinggang.

In: Brain Structure and Function, 24.11.2015, p. 1-17.

Research output: Contribution to journalArticle

@article{2a4fd01844b846248e5057b8ec8864ee,
title = "Identification of progressive mild cognitive impairment patients using incomplete longitudinal MRI scans",
abstract = "Distinguishing progressive mild cognitive impairment (pMCI) from stable mild cognitive impairment (sMCI) is critical for identification of patients who are at risk for Alzheimer’s disease (AD), so that early treatment can be administered. In this paper, we propose a pMCI/sMCI classification framework that harnesses information available in longitudinal magnetic resonance imaging (MRI) data, which could be incomplete, to improve diagnostic accuracy. Volumetric features were first extracted from the baseline MRI scan and subsequent scans acquired after 6, 12, and 18 months. Dynamic features were then obtained using the 18th month scan as the reference and computing the ratios of feature differences for the earlier scans. Features that are linearly or non-linearly correlated with diagnostic labels are then selected using two elastic net sparse learning algorithms. Missing feature values due to the incomplete longitudinal data are imputed using a low-rank matrix completion method. Finally, based on the completed feature matrix, we build a multi-kernel support vector machine (mkSVM) to predict the diagnostic label of samples with unknown diagnostic statuses. Our evaluation indicates that a diagnosis accuracy as high as 78.2 {\%} can be achieved when information from the longitudinal scans is used—6.6 {\%} higher than the case using only the reference time point image. In other words, information provided by the longitudinal history of the disease improves diagnosis accuracy.",
keywords = "Elastic net, Longitudinal MRI, Low-rank matrix completion, Missing data, Multi-kernel learning, Progressive mild cognitive impairment (pMCI)",
author = "Thung, {Kim Han} and Wee, {Chong Yaw} and Yap, {Pew Thian} and Dinggang Shen",
year = "2015",
month = "11",
day = "24",
doi = "10.1007/s00429-015-1140-6",
language = "English",
pages = "1--17",
journal = "Anatomische Hefte",
issn = "0177-5154",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Identification of progressive mild cognitive impairment patients using incomplete longitudinal MRI scans

AU - Thung, Kim Han

AU - Wee, Chong Yaw

AU - Yap, Pew Thian

AU - Shen, Dinggang

PY - 2015/11/24

Y1 - 2015/11/24

N2 - Distinguishing progressive mild cognitive impairment (pMCI) from stable mild cognitive impairment (sMCI) is critical for identification of patients who are at risk for Alzheimer’s disease (AD), so that early treatment can be administered. In this paper, we propose a pMCI/sMCI classification framework that harnesses information available in longitudinal magnetic resonance imaging (MRI) data, which could be incomplete, to improve diagnostic accuracy. Volumetric features were first extracted from the baseline MRI scan and subsequent scans acquired after 6, 12, and 18 months. Dynamic features were then obtained using the 18th month scan as the reference and computing the ratios of feature differences for the earlier scans. Features that are linearly or non-linearly correlated with diagnostic labels are then selected using two elastic net sparse learning algorithms. Missing feature values due to the incomplete longitudinal data are imputed using a low-rank matrix completion method. Finally, based on the completed feature matrix, we build a multi-kernel support vector machine (mkSVM) to predict the diagnostic label of samples with unknown diagnostic statuses. Our evaluation indicates that a diagnosis accuracy as high as 78.2 % can be achieved when information from the longitudinal scans is used—6.6 % higher than the case using only the reference time point image. In other words, information provided by the longitudinal history of the disease improves diagnosis accuracy.

AB - Distinguishing progressive mild cognitive impairment (pMCI) from stable mild cognitive impairment (sMCI) is critical for identification of patients who are at risk for Alzheimer’s disease (AD), so that early treatment can be administered. In this paper, we propose a pMCI/sMCI classification framework that harnesses information available in longitudinal magnetic resonance imaging (MRI) data, which could be incomplete, to improve diagnostic accuracy. Volumetric features were first extracted from the baseline MRI scan and subsequent scans acquired after 6, 12, and 18 months. Dynamic features were then obtained using the 18th month scan as the reference and computing the ratios of feature differences for the earlier scans. Features that are linearly or non-linearly correlated with diagnostic labels are then selected using two elastic net sparse learning algorithms. Missing feature values due to the incomplete longitudinal data are imputed using a low-rank matrix completion method. Finally, based on the completed feature matrix, we build a multi-kernel support vector machine (mkSVM) to predict the diagnostic label of samples with unknown diagnostic statuses. Our evaluation indicates that a diagnosis accuracy as high as 78.2 % can be achieved when information from the longitudinal scans is used—6.6 % higher than the case using only the reference time point image. In other words, information provided by the longitudinal history of the disease improves diagnosis accuracy.

KW - Elastic net

KW - Longitudinal MRI

KW - Low-rank matrix completion

KW - Missing data

KW - Multi-kernel learning

KW - Progressive mild cognitive impairment (pMCI)

UR - http://www.scopus.com/inward/record.url?scp=84948152359&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948152359&partnerID=8YFLogxK

U2 - 10.1007/s00429-015-1140-6

DO - 10.1007/s00429-015-1140-6

M3 - Article

SP - 1

EP - 17

JO - Anatomische Hefte

JF - Anatomische Hefte

SN - 0177-5154

ER -