A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia

Alzheimer's Disease Neuroimaging Initiative

Research output: Contribution to journalArticle

Abstract

BACKGROUND: Clinical trials in Alzheimer's disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging by training machine learning tools in a high-specificity regime. RESULTS: A multimodal signature of Alzheimer's dementia was first extracted from the ADNI1 dataset. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N = 235). The signature was optimized to predict progression to dementia over 3 years with low sensitivity (55.1%) but high specificity (95.6%), resulting in only moderate accuracy (69.3%) but high positive predictive value (80.4%, adjusted for a "typical" 33% prevalence rate of true progressors). These results were replicated in ADNI2 (N = 235), with 87.8% adjusted positive predictive value (96.7% specificity, 47.3% sensitivity, 85.1% accuracy). CONCLUSIONS: We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high-specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.

Original languageEnglish (US)
JournalGigaScience
Volume8
Issue number5
DOIs
StatePublished - May 1 2019

Fingerprint

Cognition
Atrophy
Learning systems
Brain
Alzheimer Disease
Neuroimaging
Dementia
Clinical Trials
ROC Curve
Sensitivity and Specificity
Machine Learning

Keywords

  • Alzheimer's disease
  • cognition
  • machine learning
  • mild cognitive impairment
  • neuroimaging

ASJC Scopus subject areas

  • Health Informatics
  • Computer Science Applications

Cite this

A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia. / Alzheimer's Disease Neuroimaging Initiative.

In: GigaScience, Vol. 8, No. 5, 01.05.2019.

Research output: Contribution to journalArticle

Alzheimer's Disease Neuroimaging Initiative. / A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia. In: GigaScience. 2019 ; Vol. 8, No. 5.
@article{6b09294ef8894e97b966de99fbe0465a,
title = "A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia",
abstract = "BACKGROUND: Clinical trials in Alzheimer's disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging by training machine learning tools in a high-specificity regime. RESULTS: A multimodal signature of Alzheimer's dementia was first extracted from the ADNI1 dataset. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N = 235). The signature was optimized to predict progression to dementia over 3 years with low sensitivity (55.1{\%}) but high specificity (95.6{\%}), resulting in only moderate accuracy (69.3{\%}) but high positive predictive value (80.4{\%}, adjusted for a {"}typical{"} 33{\%} prevalence rate of true progressors). These results were replicated in ADNI2 (N = 235), with 87.8{\%} adjusted positive predictive value (96.7{\%} specificity, 47.3{\%} sensitivity, 85.1{\%} accuracy). CONCLUSIONS: We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high-specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.",
keywords = "Alzheimer's disease, cognition, machine learning, mild cognitive impairment, neuroimaging",
author = "{Alzheimer's Disease Neuroimaging Initiative} and Angela Tam and Christian Dansereau and Yasser Iturria-Medina and Sebastian Urchs and Pierre Orban and Hanad Sharmarke and Breitner, {John C.S.} and Pierre Bellec",
year = "2019",
month = "5",
day = "1",
doi = "10.1093/gigascience/giz055",
language = "English (US)",
volume = "8",
journal = "GigaScience",
issn = "2047-217X",
publisher = "BioMed Central",
number = "5",

}

TY - JOUR

T1 - A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia

AU - Alzheimer's Disease Neuroimaging Initiative

AU - Tam, Angela

AU - Dansereau, Christian

AU - Iturria-Medina, Yasser

AU - Urchs, Sebastian

AU - Orban, Pierre

AU - Sharmarke, Hanad

AU - Breitner, John C.S.

AU - Bellec, Pierre

PY - 2019/5/1

Y1 - 2019/5/1

N2 - BACKGROUND: Clinical trials in Alzheimer's disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging by training machine learning tools in a high-specificity regime. RESULTS: A multimodal signature of Alzheimer's dementia was first extracted from the ADNI1 dataset. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N = 235). The signature was optimized to predict progression to dementia over 3 years with low sensitivity (55.1%) but high specificity (95.6%), resulting in only moderate accuracy (69.3%) but high positive predictive value (80.4%, adjusted for a "typical" 33% prevalence rate of true progressors). These results were replicated in ADNI2 (N = 235), with 87.8% adjusted positive predictive value (96.7% specificity, 47.3% sensitivity, 85.1% accuracy). CONCLUSIONS: We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high-specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.

AB - BACKGROUND: Clinical trials in Alzheimer's disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging by training machine learning tools in a high-specificity regime. RESULTS: A multimodal signature of Alzheimer's dementia was first extracted from the ADNI1 dataset. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N = 235). The signature was optimized to predict progression to dementia over 3 years with low sensitivity (55.1%) but high specificity (95.6%), resulting in only moderate accuracy (69.3%) but high positive predictive value (80.4%, adjusted for a "typical" 33% prevalence rate of true progressors). These results were replicated in ADNI2 (N = 235), with 87.8% adjusted positive predictive value (96.7% specificity, 47.3% sensitivity, 85.1% accuracy). CONCLUSIONS: We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high-specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.

KW - Alzheimer's disease

KW - cognition

KW - machine learning

KW - mild cognitive impairment

KW - neuroimaging

UR - http://www.scopus.com/inward/record.url?scp=85065946436&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065946436&partnerID=8YFLogxK

U2 - 10.1093/gigascience/giz055

DO - 10.1093/gigascience/giz055

M3 - Article

C2 - 31077314

AN - SCOPUS:85065946436

VL - 8

JO - GigaScience

JF - GigaScience

SN - 2047-217X

IS - 5

ER -