Multivariate prediction of Hippocampal atrophy in Alzheimer's disease

Hilkka Liedes, Jyrki Lötjönen, Juha M. Kortelainen, Gerald Novak, Mark van Gils, Mark Forrest Gordon

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Background: Hippocampal atrophy (HA) is one of the biomarkers for Alzheimer's disease (AD). Objective: To identify the best biomarkers and develop models for prediction of HA over 24 months using baseline data. Methods: The study included healthy elderly controls, subjects with mild cognitive impairment, and subjects with AD, obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI 1) and the Australian Imaging Biomarkers and Lifestyle Flagship Study of Ageing (AIBL) databases. Predictor variables included cognitive and neuropsychological tests, amyloid-, tau, and p-Tau from cerebrospinal fluid samples, apolipoprotein E, and features extracted from magnetic resonance images (MRI). Least-mean-squares regression with elastic net regularization and least absolute deviation regression models were tested using cross-validation in ADNI 1. The generalizability of the models including only MRI features was evaluated by training the models with ADNI 1 and testing them with AIBL. The models including the full set of variables were not evaluated with AIBL because not all needed variables were available in it. Results: The models including the full set of variables performed better than the models including only MRI features (root-mean-square error (RMSE) 1.76-1.82 versus 1.93-2.08). The MRI-only models performed well when applied to the independent validation cohort (RMSE 1.66-1.71). In the prediction of dichotomized HA (fast versus slow), the models achieved a reasonable prediction accuracy (0.79-0.87). Conclusions: These models can potentially help identifying subjects predicted to have a faster HA rate. This can help in selection of suitable patients into clinical trials testing disease-modifying drugs for AD.

Original languageEnglish
Pages (from-to)1453-1468
JournalJournal of Alzheimer's Disease
Volume68
Issue number4
DOIs
Publication statusPublished - 23 Apr 2019
MoE publication typeA1 Journal article-refereed

Fingerprint

Atrophy
Alzheimer Disease
Magnetic Resonance Spectroscopy
Biomarkers
Neuropsychological Tests
Apolipoproteins E
Least-Squares Analysis
Amyloid
Neuroimaging
Patient Selection
Cerebrospinal Fluid
Life Style
Clinical Trials
Databases
Pharmaceutical Preparations

Keywords

  • Alzheimer's disease
  • Atrophy
  • Decision support techniques
  • Disease progression
  • Hippocampus
  • Magnetic resonance imaging
  • Regression analysis
  • Statistical models

Cite this

Liedes, Hilkka ; Lötjönen, Jyrki ; Kortelainen, Juha M. ; Novak, Gerald ; van Gils, Mark ; Gordon, Mark Forrest. / Multivariate prediction of Hippocampal atrophy in Alzheimer's disease. In: Journal of Alzheimer's Disease. 2019 ; Vol. 68, No. 4. pp. 1453-1468.
@article{041a1e9590904d7592f32e900d5047b2,
title = "Multivariate prediction of Hippocampal atrophy in Alzheimer's disease",
abstract = "Background: Hippocampal atrophy (HA) is one of the biomarkers for Alzheimer's disease (AD). Objective: To identify the best biomarkers and develop models for prediction of HA over 24 months using baseline data. Methods: The study included healthy elderly controls, subjects with mild cognitive impairment, and subjects with AD, obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI 1) and the Australian Imaging Biomarkers and Lifestyle Flagship Study of Ageing (AIBL) databases. Predictor variables included cognitive and neuropsychological tests, amyloid-, tau, and p-Tau from cerebrospinal fluid samples, apolipoprotein E, and features extracted from magnetic resonance images (MRI). Least-mean-squares regression with elastic net regularization and least absolute deviation regression models were tested using cross-validation in ADNI 1. The generalizability of the models including only MRI features was evaluated by training the models with ADNI 1 and testing them with AIBL. The models including the full set of variables were not evaluated with AIBL because not all needed variables were available in it. Results: The models including the full set of variables performed better than the models including only MRI features (root-mean-square error (RMSE) 1.76-1.82 versus 1.93-2.08). The MRI-only models performed well when applied to the independent validation cohort (RMSE 1.66-1.71). In the prediction of dichotomized HA (fast versus slow), the models achieved a reasonable prediction accuracy (0.79-0.87). Conclusions: These models can potentially help identifying subjects predicted to have a faster HA rate. This can help in selection of suitable patients into clinical trials testing disease-modifying drugs for AD.",
keywords = "Alzheimer's disease, Atrophy, Decision support techniques, Disease progression, Hippocampus, Magnetic resonance imaging, Regression analysis, Statistical models",
author = "Hilkka Liedes and Jyrki L{\"o}tj{\"o}nen and Kortelainen, {Juha M.} and Gerald Novak and {van Gils}, Mark and Gordon, {Mark Forrest}",
note = "project 121842",
year = "2019",
month = "4",
day = "23",
doi = "10.3233/JAD-180484",
language = "English",
volume = "68",
pages = "1453--1468",
journal = "Journal of Alzheimer's Disease",
issn = "1387-2877",
publisher = "IOS Press",
number = "4",

}

Multivariate prediction of Hippocampal atrophy in Alzheimer's disease. / Liedes, Hilkka; Lötjönen, Jyrki; Kortelainen, Juha M.; Novak, Gerald; van Gils, Mark; Gordon, Mark Forrest.

In: Journal of Alzheimer's Disease, Vol. 68, No. 4, 23.04.2019, p. 1453-1468.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Multivariate prediction of Hippocampal atrophy in Alzheimer's disease

AU - Liedes, Hilkka

AU - Lötjönen, Jyrki

AU - Kortelainen, Juha M.

AU - Novak, Gerald

AU - van Gils, Mark

AU - Gordon, Mark Forrest

N1 - project 121842

PY - 2019/4/23

Y1 - 2019/4/23

N2 - Background: Hippocampal atrophy (HA) is one of the biomarkers for Alzheimer's disease (AD). Objective: To identify the best biomarkers and develop models for prediction of HA over 24 months using baseline data. Methods: The study included healthy elderly controls, subjects with mild cognitive impairment, and subjects with AD, obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI 1) and the Australian Imaging Biomarkers and Lifestyle Flagship Study of Ageing (AIBL) databases. Predictor variables included cognitive and neuropsychological tests, amyloid-, tau, and p-Tau from cerebrospinal fluid samples, apolipoprotein E, and features extracted from magnetic resonance images (MRI). Least-mean-squares regression with elastic net regularization and least absolute deviation regression models were tested using cross-validation in ADNI 1. The generalizability of the models including only MRI features was evaluated by training the models with ADNI 1 and testing them with AIBL. The models including the full set of variables were not evaluated with AIBL because not all needed variables were available in it. Results: The models including the full set of variables performed better than the models including only MRI features (root-mean-square error (RMSE) 1.76-1.82 versus 1.93-2.08). The MRI-only models performed well when applied to the independent validation cohort (RMSE 1.66-1.71). In the prediction of dichotomized HA (fast versus slow), the models achieved a reasonable prediction accuracy (0.79-0.87). Conclusions: These models can potentially help identifying subjects predicted to have a faster HA rate. This can help in selection of suitable patients into clinical trials testing disease-modifying drugs for AD.

AB - Background: Hippocampal atrophy (HA) is one of the biomarkers for Alzheimer's disease (AD). Objective: To identify the best biomarkers and develop models for prediction of HA over 24 months using baseline data. Methods: The study included healthy elderly controls, subjects with mild cognitive impairment, and subjects with AD, obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI 1) and the Australian Imaging Biomarkers and Lifestyle Flagship Study of Ageing (AIBL) databases. Predictor variables included cognitive and neuropsychological tests, amyloid-, tau, and p-Tau from cerebrospinal fluid samples, apolipoprotein E, and features extracted from magnetic resonance images (MRI). Least-mean-squares regression with elastic net regularization and least absolute deviation regression models were tested using cross-validation in ADNI 1. The generalizability of the models including only MRI features was evaluated by training the models with ADNI 1 and testing them with AIBL. The models including the full set of variables were not evaluated with AIBL because not all needed variables were available in it. Results: The models including the full set of variables performed better than the models including only MRI features (root-mean-square error (RMSE) 1.76-1.82 versus 1.93-2.08). The MRI-only models performed well when applied to the independent validation cohort (RMSE 1.66-1.71). In the prediction of dichotomized HA (fast versus slow), the models achieved a reasonable prediction accuracy (0.79-0.87). Conclusions: These models can potentially help identifying subjects predicted to have a faster HA rate. This can help in selection of suitable patients into clinical trials testing disease-modifying drugs for AD.

KW - Alzheimer's disease

KW - Atrophy

KW - Decision support techniques

KW - Disease progression

KW - Hippocampus

KW - Magnetic resonance imaging

KW - Regression analysis

KW - Statistical models

UR - http://www.scopus.com/inward/record.url?scp=85064859223&partnerID=8YFLogxK

U2 - 10.3233/JAD-180484

DO - 10.3233/JAD-180484

M3 - Article

C2 - 30909211

AN - SCOPUS:85064859223

VL - 68

SP - 1453

EP - 1468

JO - Journal of Alzheimer's Disease

JF - Journal of Alzheimer's Disease

SN - 1387-2877

IS - 4

ER -