Machine Learning-based Individual Assessment of Cortical Atrophy Pattern in Alzheimer's Disease Spectrum: Development of the Classifier and Longitudinal Evaluation

Jin San Lee, Changsoo Kim, Jeong Hyeon Shin, Hanna Cho, Dae Seock Shin, Nakyoung Kim, Hee Jin Kim, Yeshin Kim, Samuel N. Lockhart, Duk L. Na, Sang Won Seo, Joon Kyung Seong

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

To develop a new method for measuring Alzheimer's disease (AD)-specific similarity of cortical atrophy patterns at the individual-level, we employed an individual-level machine learning algorithm. A total of 869 cognitively normal (CN) individuals and 473 patients with probable AD dementia who underwent high-resolution 3T brain MRI were included. We propose a machine learning-based method for measuring the similarity of an individual subject's cortical atrophy pattern with that of a representative AD patient cohort. In addition, we validated this similarity measure in two longitudinal cohorts consisting of 79 patients with amnestic-mild cognitive impairment (aMCI) and 27 patients with probable AD dementia. Surface-based morphometry classifier for discriminating AD from CN showed sensitivity and specificity values of 87.1% and 93.3%, respectively. In the longitudinal validation study, aMCI-converts had higher atrophy similarity at both baseline (p < 0.001) and first year visits (p < 0.001) relative to non-converters. Similarly, AD patients with faster decline had higher atrophy similarity than slower decliners at baseline (p = 0.042), first year (p = 0.028), and third year visits (p = 0.027). The AD-specific atrophy similarity measure is a novel approach for the prediction of dementia risk and for the evaluation of AD trajectories on an individual subject level.

Original languageEnglish
Article number4161
JournalScientific reports
Volume8
Issue number1
DOIs
Publication statusPublished - 2018 Dec 1

Fingerprint

Atrophy
Alzheimer Disease
Machine Learning
Validation Studies
Longitudinal Studies
Dementia
Sensitivity and Specificity
Brain

All Science Journal Classification (ASJC) codes

  • General

Cite this

Lee, Jin San ; Kim, Changsoo ; Shin, Jeong Hyeon ; Cho, Hanna ; Shin, Dae Seock ; Kim, Nakyoung ; Kim, Hee Jin ; Kim, Yeshin ; Lockhart, Samuel N. ; Na, Duk L. ; Seo, Sang Won ; Seong, Joon Kyung. / Machine Learning-based Individual Assessment of Cortical Atrophy Pattern in Alzheimer's Disease Spectrum : Development of the Classifier and Longitudinal Evaluation. In: Scientific reports. 2018 ; Vol. 8, No. 1.
@article{375813f025564903bdd3796842b5007d,
title = "Machine Learning-based Individual Assessment of Cortical Atrophy Pattern in Alzheimer's Disease Spectrum: Development of the Classifier and Longitudinal Evaluation",
abstract = "To develop a new method for measuring Alzheimer's disease (AD)-specific similarity of cortical atrophy patterns at the individual-level, we employed an individual-level machine learning algorithm. A total of 869 cognitively normal (CN) individuals and 473 patients with probable AD dementia who underwent high-resolution 3T brain MRI were included. We propose a machine learning-based method for measuring the similarity of an individual subject's cortical atrophy pattern with that of a representative AD patient cohort. In addition, we validated this similarity measure in two longitudinal cohorts consisting of 79 patients with amnestic-mild cognitive impairment (aMCI) and 27 patients with probable AD dementia. Surface-based morphometry classifier for discriminating AD from CN showed sensitivity and specificity values of 87.1{\%} and 93.3{\%}, respectively. In the longitudinal validation study, aMCI-converts had higher atrophy similarity at both baseline (p < 0.001) and first year visits (p < 0.001) relative to non-converters. Similarly, AD patients with faster decline had higher atrophy similarity than slower decliners at baseline (p = 0.042), first year (p = 0.028), and third year visits (p = 0.027). The AD-specific atrophy similarity measure is a novel approach for the prediction of dementia risk and for the evaluation of AD trajectories on an individual subject level.",
author = "Lee, {Jin San} and Changsoo Kim and Shin, {Jeong Hyeon} and Hanna Cho and Shin, {Dae Seock} and Nakyoung Kim and Kim, {Hee Jin} and Yeshin Kim and Lockhart, {Samuel N.} and Na, {Duk L.} and Seo, {Sang Won} and Seong, {Joon Kyung}",
year = "2018",
month = "12",
day = "1",
doi = "10.1038/s41598-018-22277-x",
language = "English",
volume = "8",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

Machine Learning-based Individual Assessment of Cortical Atrophy Pattern in Alzheimer's Disease Spectrum : Development of the Classifier and Longitudinal Evaluation. / Lee, Jin San; Kim, Changsoo; Shin, Jeong Hyeon; Cho, Hanna; Shin, Dae Seock; Kim, Nakyoung; Kim, Hee Jin; Kim, Yeshin; Lockhart, Samuel N.; Na, Duk L.; Seo, Sang Won; Seong, Joon Kyung.

In: Scientific reports, Vol. 8, No. 1, 4161, 01.12.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Machine Learning-based Individual Assessment of Cortical Atrophy Pattern in Alzheimer's Disease Spectrum

T2 - Development of the Classifier and Longitudinal Evaluation

AU - Lee, Jin San

AU - Kim, Changsoo

AU - Shin, Jeong Hyeon

AU - Cho, Hanna

AU - Shin, Dae Seock

AU - Kim, Nakyoung

AU - Kim, Hee Jin

AU - Kim, Yeshin

AU - Lockhart, Samuel N.

AU - Na, Duk L.

AU - Seo, Sang Won

AU - Seong, Joon Kyung

PY - 2018/12/1

Y1 - 2018/12/1

N2 - To develop a new method for measuring Alzheimer's disease (AD)-specific similarity of cortical atrophy patterns at the individual-level, we employed an individual-level machine learning algorithm. A total of 869 cognitively normal (CN) individuals and 473 patients with probable AD dementia who underwent high-resolution 3T brain MRI were included. We propose a machine learning-based method for measuring the similarity of an individual subject's cortical atrophy pattern with that of a representative AD patient cohort. In addition, we validated this similarity measure in two longitudinal cohorts consisting of 79 patients with amnestic-mild cognitive impairment (aMCI) and 27 patients with probable AD dementia. Surface-based morphometry classifier for discriminating AD from CN showed sensitivity and specificity values of 87.1% and 93.3%, respectively. In the longitudinal validation study, aMCI-converts had higher atrophy similarity at both baseline (p < 0.001) and first year visits (p < 0.001) relative to non-converters. Similarly, AD patients with faster decline had higher atrophy similarity than slower decliners at baseline (p = 0.042), first year (p = 0.028), and third year visits (p = 0.027). The AD-specific atrophy similarity measure is a novel approach for the prediction of dementia risk and for the evaluation of AD trajectories on an individual subject level.

AB - To develop a new method for measuring Alzheimer's disease (AD)-specific similarity of cortical atrophy patterns at the individual-level, we employed an individual-level machine learning algorithm. A total of 869 cognitively normal (CN) individuals and 473 patients with probable AD dementia who underwent high-resolution 3T brain MRI were included. We propose a machine learning-based method for measuring the similarity of an individual subject's cortical atrophy pattern with that of a representative AD patient cohort. In addition, we validated this similarity measure in two longitudinal cohorts consisting of 79 patients with amnestic-mild cognitive impairment (aMCI) and 27 patients with probable AD dementia. Surface-based morphometry classifier for discriminating AD from CN showed sensitivity and specificity values of 87.1% and 93.3%, respectively. In the longitudinal validation study, aMCI-converts had higher atrophy similarity at both baseline (p < 0.001) and first year visits (p < 0.001) relative to non-converters. Similarly, AD patients with faster decline had higher atrophy similarity than slower decliners at baseline (p = 0.042), first year (p = 0.028), and third year visits (p = 0.027). The AD-specific atrophy similarity measure is a novel approach for the prediction of dementia risk and for the evaluation of AD trajectories on an individual subject level.

UR - http://www.scopus.com/inward/record.url?scp=85048269237&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048269237&partnerID=8YFLogxK

U2 - 10.1038/s41598-018-22277-x

DO - 10.1038/s41598-018-22277-x

M3 - Article

C2 - 29515131

AN - SCOPUS:85048269237

VL - 8

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 4161

ER -