A novel ensemble-based machine learning algorithm to predict the conversion from mild cognitive impairment to Alzheimer's disease using socio-demographic characteristics, clinical information, and neuropsychological measures

Alzheimer's Disease Neuroimaging Initiative

Research output: Contribution to journalArticle

Abstract

Background: Despite the increasing availability in brain health related data, clinically translatable methods to predict the conversion from Mild Cognitive Impairment (MCI) to Alzheimer's disease (AD) are still lacking. Although MCI typically precedes AD, only a fraction of 20-40% of MCI individuals will progress to dementia within 3 years following the initial diagnosis. As currently available and emerging therapies likely have the greatest impact when provided at the earliest disease stage, the prompt identification of subjects at high risk for conversion to AD is of great importance in the fight against this disease. In this work, we propose a highly predictive machine learning algorithm, based only on non-invasively and easily in-the-clinic collectable predictors, to identify MCI subjects at risk for conversion to AD. Methods: The algorithm was developed using the open dataset from the Alzheimer's Disease Neuroimaging Initiative (ADNI), employing a sample of 550 MCI subjects whose diagnostic follow-up is available for at least 3 years after the baseline assessment. A restricted set of information regarding sociodemographic and clinical characteristics, neuropsychological test scores was used as predictors and several different supervised machine learning algorithms were developed and ensembled in final algorithm. A site-independent stratified train/test split protocol was used to provide an estimate of the generalized performance of the algorithm. Results: The final algorithm demonstrated an AUROC of 0.88, sensitivity of 77.7%, and a specificity of 79.9% on excluded test data. The specificity of the algorithm was 40.2% for 100% sensitivity. Conclusions: The algorithm we developed achieved sound and high prognostic performance to predict AD conversion using easily clinically derived information that makes the algorithm easy to be translated into practice. This indicates beneficial application to improve recruitment in clinical trials and to more selectively prescribe new and newly emerging early interventions to high AD risk patients.

Original languageEnglish (US)
Article number756
JournalFrontiers in Neurology
Volume10
Issue numberJUL
DOIs
StatePublished - Jan 1 2019

Fingerprint

Alzheimer Disease
Demography
Cognitive Dysfunction
Machine Learning
Neuropsychological Tests
Neuroimaging
Dementia
Clinical Trials
Health
Brain

Keywords

  • Alzheimer's disease
  • Clinical prediction rule
  • Machine learning
  • Mild cognitive impairment
  • Neuropsychological tests
  • Personalized medicine
  • Precision medicine

ASJC Scopus subject areas

  • Neurology
  • Clinical Neurology

Cite this

@article{4e61551d6d284083a3df11fde70b2e32,
title = "A novel ensemble-based machine learning algorithm to predict the conversion from mild cognitive impairment to Alzheimer's disease using socio-demographic characteristics, clinical information, and neuropsychological measures",
abstract = "Background: Despite the increasing availability in brain health related data, clinically translatable methods to predict the conversion from Mild Cognitive Impairment (MCI) to Alzheimer's disease (AD) are still lacking. Although MCI typically precedes AD, only a fraction of 20-40{\%} of MCI individuals will progress to dementia within 3 years following the initial diagnosis. As currently available and emerging therapies likely have the greatest impact when provided at the earliest disease stage, the prompt identification of subjects at high risk for conversion to AD is of great importance in the fight against this disease. In this work, we propose a highly predictive machine learning algorithm, based only on non-invasively and easily in-the-clinic collectable predictors, to identify MCI subjects at risk for conversion to AD. Methods: The algorithm was developed using the open dataset from the Alzheimer's Disease Neuroimaging Initiative (ADNI), employing a sample of 550 MCI subjects whose diagnostic follow-up is available for at least 3 years after the baseline assessment. A restricted set of information regarding sociodemographic and clinical characteristics, neuropsychological test scores was used as predictors and several different supervised machine learning algorithms were developed and ensembled in final algorithm. A site-independent stratified train/test split protocol was used to provide an estimate of the generalized performance of the algorithm. Results: The final algorithm demonstrated an AUROC of 0.88, sensitivity of 77.7{\%}, and a specificity of 79.9{\%} on excluded test data. The specificity of the algorithm was 40.2{\%} for 100{\%} sensitivity. Conclusions: The algorithm we developed achieved sound and high prognostic performance to predict AD conversion using easily clinically derived information that makes the algorithm easy to be translated into practice. This indicates beneficial application to improve recruitment in clinical trials and to more selectively prescribe new and newly emerging early interventions to high AD risk patients.",
keywords = "Alzheimer's disease, Clinical prediction rule, Machine learning, Mild cognitive impairment, Neuropsychological tests, Personalized medicine, Precision medicine",
author = "{Alzheimer's Disease Neuroimaging Initiative} and Massimiliano Grassi and Nadine Rouleaux and Daniela Caldirola and David Loewenstein and Koen Schruers and Giampaolo Perna and Michel Dumontier",
year = "2019",
month = "1",
day = "1",
doi = "10.3389/fneur.2019.00756",
language = "English (US)",
volume = "10",
journal = "Frontiers in Neurology",
issn = "1664-2295",
publisher = "Frontiers Research Foundation",
number = "JUL",

}

TY - JOUR

T1 - A novel ensemble-based machine learning algorithm to predict the conversion from mild cognitive impairment to Alzheimer's disease using socio-demographic characteristics, clinical information, and neuropsychological measures

AU - Alzheimer's Disease Neuroimaging Initiative

AU - Grassi, Massimiliano

AU - Rouleaux, Nadine

AU - Caldirola, Daniela

AU - Loewenstein, David

AU - Schruers, Koen

AU - Perna, Giampaolo

AU - Dumontier, Michel

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Background: Despite the increasing availability in brain health related data, clinically translatable methods to predict the conversion from Mild Cognitive Impairment (MCI) to Alzheimer's disease (AD) are still lacking. Although MCI typically precedes AD, only a fraction of 20-40% of MCI individuals will progress to dementia within 3 years following the initial diagnosis. As currently available and emerging therapies likely have the greatest impact when provided at the earliest disease stage, the prompt identification of subjects at high risk for conversion to AD is of great importance in the fight against this disease. In this work, we propose a highly predictive machine learning algorithm, based only on non-invasively and easily in-the-clinic collectable predictors, to identify MCI subjects at risk for conversion to AD. Methods: The algorithm was developed using the open dataset from the Alzheimer's Disease Neuroimaging Initiative (ADNI), employing a sample of 550 MCI subjects whose diagnostic follow-up is available for at least 3 years after the baseline assessment. A restricted set of information regarding sociodemographic and clinical characteristics, neuropsychological test scores was used as predictors and several different supervised machine learning algorithms were developed and ensembled in final algorithm. A site-independent stratified train/test split protocol was used to provide an estimate of the generalized performance of the algorithm. Results: The final algorithm demonstrated an AUROC of 0.88, sensitivity of 77.7%, and a specificity of 79.9% on excluded test data. The specificity of the algorithm was 40.2% for 100% sensitivity. Conclusions: The algorithm we developed achieved sound and high prognostic performance to predict AD conversion using easily clinically derived information that makes the algorithm easy to be translated into practice. This indicates beneficial application to improve recruitment in clinical trials and to more selectively prescribe new and newly emerging early interventions to high AD risk patients.

AB - Background: Despite the increasing availability in brain health related data, clinically translatable methods to predict the conversion from Mild Cognitive Impairment (MCI) to Alzheimer's disease (AD) are still lacking. Although MCI typically precedes AD, only a fraction of 20-40% of MCI individuals will progress to dementia within 3 years following the initial diagnosis. As currently available and emerging therapies likely have the greatest impact when provided at the earliest disease stage, the prompt identification of subjects at high risk for conversion to AD is of great importance in the fight against this disease. In this work, we propose a highly predictive machine learning algorithm, based only on non-invasively and easily in-the-clinic collectable predictors, to identify MCI subjects at risk for conversion to AD. Methods: The algorithm was developed using the open dataset from the Alzheimer's Disease Neuroimaging Initiative (ADNI), employing a sample of 550 MCI subjects whose diagnostic follow-up is available for at least 3 years after the baseline assessment. A restricted set of information regarding sociodemographic and clinical characteristics, neuropsychological test scores was used as predictors and several different supervised machine learning algorithms were developed and ensembled in final algorithm. A site-independent stratified train/test split protocol was used to provide an estimate of the generalized performance of the algorithm. Results: The final algorithm demonstrated an AUROC of 0.88, sensitivity of 77.7%, and a specificity of 79.9% on excluded test data. The specificity of the algorithm was 40.2% for 100% sensitivity. Conclusions: The algorithm we developed achieved sound and high prognostic performance to predict AD conversion using easily clinically derived information that makes the algorithm easy to be translated into practice. This indicates beneficial application to improve recruitment in clinical trials and to more selectively prescribe new and newly emerging early interventions to high AD risk patients.

KW - Alzheimer's disease

KW - Clinical prediction rule

KW - Machine learning

KW - Mild cognitive impairment

KW - Neuropsychological tests

KW - Personalized medicine

KW - Precision medicine

UR - http://www.scopus.com/inward/record.url?scp=85069758724&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069758724&partnerID=8YFLogxK

U2 - 10.3389/fneur.2019.00756

DO - 10.3389/fneur.2019.00756

M3 - Article

AN - SCOPUS:85069758724

VL - 10

JO - Frontiers in Neurology

JF - Frontiers in Neurology

SN - 1664-2295

IS - JUL

M1 - 756

ER -