Predicting Alzheimer’s disease progression using multi-modal deep learning approach

for Alzheimer’s Disease Neuroimaging Initiative

doi:10.1038/s41598-018-37769-z

Predicting Alzheimer’s disease progression using multi-modal deep learning approach

for Alzheimer’s Disease Neuroimaging Initiative

Neurology

Research output: Contribution to journal › Article › peer-review

331 Scopus citations

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative condition marked by a decline in cognitive functions with no validated disease modifying treatment. It is critical for timely treatment to detect AD in its earlier stage before clinical manifestation. Mild cognitive impairment (MCI) is an intermediate stage between cognitively normal older adults and AD. To predict conversion from MCI to probable AD, we applied a deep learning approach, multimodal recurrent neural network. We developed an integrative framework that combines not only cross-sectional neuroimaging biomarkers at baseline but also longitudinal cerebrospinal fluid (CSF) and cognitive performance biomarkers obtained from the Alzheimer’s Disease Neuroimaging Initiative cohort (ADNI). The proposed framework integrated longitudinal multi-domain data. Our results showed that 1) our prediction model for MCI conversion to AD yielded up to 75% accuracy (area under the curve (AUC) = 0.83) when using only single modality of data separately; and 2) our prediction model achieved the best performance with 81% accuracy (AUC = 0.86) when incorporating longitudinal multi-domain data. A multi-modal deep learning approach has potential to identify persons at risk of developing AD who might benefit most from a clinical trial or as a stratification approach within clinical trials.

Original language	English (US)
Article number	1952
Journal	Scientific reports
Volume	9
Issue number	1
DOIs	https://doi.org/10.1038/s41598-018-37769-z
State	Published - Dec 1 2019

Funding

Data used in the preparation of this article were obtained from the Alzheimer\u2019s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found http://adni.loni.usc.edu/wp-content/themes/freshnews-dev-v2/documents/policy/ADNI_Acknowledgement_List%205-29-18.pdf. Data collection and sharing for this project was funded by the Alzheimer\u2019s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: Alzheimer\u2019s Association; Alzheimer\u2019s Drug Discovery Foundation; BioClinica, Inc.; Biogen Idec Inc.; Bristol-Myers Squibb Company; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; GE Healthcare; Innogenetics, N.V.; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Medpace, Inc.; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Synarc Inc.; and Takeda Pharmaceutical Company. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health (www. fnih.org). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer\u2019s Disease Cooperative Study at the University of California, San Diego. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Samples from the National Cell Repository for AD (NCRAD), which receives government support under a cooperative agreement grant (U24 AG21886) awarded by the National Institute on Aging (AIG), were used in this study. Additional support for data analysis was provided by NLM R01 LM012535, NIA R03 AG054936, and the Pennsylvania Department of Health (#SAP 4100070267). The Department specifically disclaims responsibility for any analyses, interpretations or conclusions. This research was also supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education [NRF-2016R1D1A1B03933875].

ASJC Scopus subject areas

General

Access to Document

10.1038/s41598-018-37769-z

Cite this

@article{6f794df1b15e423f948427547bc71dbc,

title = "Predicting Alzheimer{\textquoteright}s disease progression using multi-modal deep learning approach",

abstract = "Alzheimer{\textquoteright}s disease (AD) is a progressive neurodegenerative condition marked by a decline in cognitive functions with no validated disease modifying treatment. It is critical for timely treatment to detect AD in its earlier stage before clinical manifestation. Mild cognitive impairment (MCI) is an intermediate stage between cognitively normal older adults and AD. To predict conversion from MCI to probable AD, we applied a deep learning approach, multimodal recurrent neural network. We developed an integrative framework that combines not only cross-sectional neuroimaging biomarkers at baseline but also longitudinal cerebrospinal fluid (CSF) and cognitive performance biomarkers obtained from the Alzheimer{\textquoteright}s Disease Neuroimaging Initiative cohort (ADNI). The proposed framework integrated longitudinal multi-domain data. Our results showed that 1) our prediction model for MCI conversion to AD yielded up to 75% accuracy (area under the curve (AUC) = 0.83) when using only single modality of data separately; and 2) our prediction model achieved the best performance with 81% accuracy (AUC = 0.86) when incorporating longitudinal multi-domain data. A multi-modal deep learning approach has potential to identify persons at risk of developing AD who might benefit most from a clinical trial or as a stratification approach within clinical trials.",

author = "{for Alzheimer{\textquoteright}s Disease Neuroimaging Initiative} and Garam Lee and Kwangsik Nho and Byungkon Kang and Sohn, {Kyung Ah} and Dokyoon Kim and Weiner, {Michael W.} and Paul Aisen and Ronald Petersen and Jack, {Clifford R.} and William Jagust and Trojanowki, {John Q.} and Toga, {Arthur W.} and Laurel Beckett and Green, {Robert C.} and Saykin, {Andrew J.} and John Morris and Shaw, {Leslie M.} and Zaven Khachaturian and Greg Sorensen and Maria Carrillo and Lew Kuller and Marc Raichle and Steven Paul and Peter Davies and Howard Fillit and Franz Hefti and Davie Holtzman and Mesulam, {M. Marcel} and William Potter and Peter Snyder and Tom Montine and Thomas, {Ronald G.} and Michael Donohue and Sarah Walter and Tamie Sather and Gus Jiminez and Balasubramanian, {Archana B.} and Jennifer Mason and Iris Sim and Danielle Harvey and Matthew Bernstein and Nick Fox and Paul Thompson and Norbert Schuff and Charles DeCArli and Bret Borowski and Jeff Gunter and Matt Senjem and Prashanthi Vemuri and David Jones",

note = "Publisher Copyright: {\textcopyright} 2019, The Author(s).",

year = "2019",

month = dec,

day = "1",

doi = "10.1038/s41598-018-37769-z",

language = "English (US)",

volume = "9",

journal = "Scientific reports",

issn = "2045-2322",

publisher = "Nature Research",

number = "1",

}

TY - JOUR

T1 - Predicting Alzheimer’s disease progression using multi-modal deep learning approach

AU - for Alzheimer’s Disease Neuroimaging Initiative

AU - Lee, Garam

AU - Nho, Kwangsik

AU - Kang, Byungkon

AU - Sohn, Kyung Ah

AU - Kim, Dokyoon

AU - Weiner, Michael W.

AU - Aisen, Paul

AU - Petersen, Ronald

AU - Jack, Clifford R.

AU - Jagust, William

AU - Trojanowki, John Q.

AU - Toga, Arthur W.

AU - Beckett, Laurel

AU - Green, Robert C.

AU - Saykin, Andrew J.

AU - Morris, John

AU - Shaw, Leslie M.

AU - Khachaturian, Zaven

AU - Sorensen, Greg

AU - Carrillo, Maria

AU - Kuller, Lew

AU - Raichle, Marc

AU - Paul, Steven

AU - Davies, Peter

AU - Fillit, Howard

AU - Hefti, Franz

AU - Holtzman, Davie

AU - Mesulam, M. Marcel

AU - Potter, William

AU - Snyder, Peter

AU - Montine, Tom

AU - Thomas, Ronald G.

AU - Donohue, Michael

AU - Walter, Sarah

AU - Sather, Tamie

AU - Jiminez, Gus

AU - Balasubramanian, Archana B.

AU - Mason, Jennifer

AU - Sim, Iris

AU - Harvey, Danielle

AU - Bernstein, Matthew

AU - Fox, Nick

AU - Thompson, Paul

AU - Schuff, Norbert

AU - DeCArli, Charles

AU - Borowski, Bret

AU - Gunter, Jeff

AU - Senjem, Matt

AU - Vemuri, Prashanthi

AU - Jones, David

PY - 2019/12/1

Y1 - 2019/12/1

N2 - Alzheimer’s disease (AD) is a progressive neurodegenerative condition marked by a decline in cognitive functions with no validated disease modifying treatment. It is critical for timely treatment to detect AD in its earlier stage before clinical manifestation. Mild cognitive impairment (MCI) is an intermediate stage between cognitively normal older adults and AD. To predict conversion from MCI to probable AD, we applied a deep learning approach, multimodal recurrent neural network. We developed an integrative framework that combines not only cross-sectional neuroimaging biomarkers at baseline but also longitudinal cerebrospinal fluid (CSF) and cognitive performance biomarkers obtained from the Alzheimer’s Disease Neuroimaging Initiative cohort (ADNI). The proposed framework integrated longitudinal multi-domain data. Our results showed that 1) our prediction model for MCI conversion to AD yielded up to 75% accuracy (area under the curve (AUC) = 0.83) when using only single modality of data separately; and 2) our prediction model achieved the best performance with 81% accuracy (AUC = 0.86) when incorporating longitudinal multi-domain data. A multi-modal deep learning approach has potential to identify persons at risk of developing AD who might benefit most from a clinical trial or as a stratification approach within clinical trials.

AB - Alzheimer’s disease (AD) is a progressive neurodegenerative condition marked by a decline in cognitive functions with no validated disease modifying treatment. It is critical for timely treatment to detect AD in its earlier stage before clinical manifestation. Mild cognitive impairment (MCI) is an intermediate stage between cognitively normal older adults and AD. To predict conversion from MCI to probable AD, we applied a deep learning approach, multimodal recurrent neural network. We developed an integrative framework that combines not only cross-sectional neuroimaging biomarkers at baseline but also longitudinal cerebrospinal fluid (CSF) and cognitive performance biomarkers obtained from the Alzheimer’s Disease Neuroimaging Initiative cohort (ADNI). The proposed framework integrated longitudinal multi-domain data. Our results showed that 1) our prediction model for MCI conversion to AD yielded up to 75% accuracy (area under the curve (AUC) = 0.83) when using only single modality of data separately; and 2) our prediction model achieved the best performance with 81% accuracy (AUC = 0.86) when incorporating longitudinal multi-domain data. A multi-modal deep learning approach has potential to identify persons at risk of developing AD who might benefit most from a clinical trial or as a stratification approach within clinical trials.

UR - http://www.scopus.com/inward/record.url?scp=85061494936&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061494936&partnerID=8YFLogxK

U2 - 10.1038/s41598-018-37769-z

DO - 10.1038/s41598-018-37769-z

M3 - Article

C2 - 30760848

AN - SCOPUS:85061494936

SN - 2045-2322

VL - 9

JO - Scientific reports

JF - Scientific reports

IS - 1

M1 - 1952

ER -

Predicting Alzheimer’s disease progression using multi-modal deep learning approach

Abstract

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this