Comparing a pre-defined versus deep learning approach for extracting brain atrophy patterns to predict cognitive decline due to Alzheimer's disease in patients with mild cognitive symptoms.
Alzheimer’s disease
cognitive decline
deep learning
Journal
Research square
Titre abrégé: Res Sq
Pays: United States
ID NLM: 101768035
Informations de publication
Date de publication:
08 Nov 2023
08 Nov 2023
Historique:
pubmed:
21
11
2023
medline:
21
11
2023
entrez:
21
11
2023
Statut:
epublish
Résumé
Predicting future Alzheimer's disease (AD)-related cognitive decline among individuals with subjective cognitive decline (SCD) or mild cognitive impairment (MCI) is an important task for healthcare. Structural brain imaging as measured by magnetic resonance imaging (MRI) could potentially contribute when making such predictions. It is unclear if the predictive performance of MRI can be improved using entire brain images in deep learning (DL) models compared to using pre-defined brain regions. A cohort of 332 individuals with SCD/MCI were included from the Swedish BioFINDER-1 study. The goal was to predict longitudinal SCD/MCI-to-AD dementia progression and change in Mini-Mental State Examination (MMSE) over four years. Four models were evaluated using different predictors: 1) clinical data only, including demographics, cognitive tests and In the test set, 21 patients (32.3%) progressed to AD dementia. The performance of the clinical data model for prediction of progression to AD dementia was area under the curve (AUC)=0.87 and four-year cognitive decline was R The DL model did not significantly improve the prediction of clinical disease progression in AD, compared to regression models with a single pre-defined brain region.
Sections du résumé
Background
UNASSIGNED
Predicting future Alzheimer's disease (AD)-related cognitive decline among individuals with subjective cognitive decline (SCD) or mild cognitive impairment (MCI) is an important task for healthcare. Structural brain imaging as measured by magnetic resonance imaging (MRI) could potentially contribute when making such predictions. It is unclear if the predictive performance of MRI can be improved using entire brain images in deep learning (DL) models compared to using pre-defined brain regions.
Methods
UNASSIGNED
A cohort of 332 individuals with SCD/MCI were included from the Swedish BioFINDER-1 study. The goal was to predict longitudinal SCD/MCI-to-AD dementia progression and change in Mini-Mental State Examination (MMSE) over four years. Four models were evaluated using different predictors: 1) clinical data only, including demographics, cognitive tests and
Results
UNASSIGNED
In the test set, 21 patients (32.3%) progressed to AD dementia. The performance of the clinical data model for prediction of progression to AD dementia was area under the curve (AUC)=0.87 and four-year cognitive decline was R
Conclusions
UNASSIGNED
The DL model did not significantly improve the prediction of clinical disease progression in AD, compared to regression models with a single pre-defined brain region.
Identifiants
pubmed: 37986841
doi: 10.21203/rs.3.rs-3569391/v1
pmc: PMC10659533
pii:
doi:
Types de publication
Preprint
Langues
eng
Subventions
Organisme : NIA NIH HHS
ID : R01 AG083740
Pays : United States
Déclaration de conflit d'intérêts
OH has acquired research support (for the institution) from ADx, AVID Radiopharmaceuticals, Biogen, Eli Lilly, Eisai, Fujirebio, GE Healthcare, Pfizer, and Roche. In the past 2 years, he has received consultancy/speaker fees from AC Immune, Amylyx, Alzpath, BioArctic, Biogen, Cerveau, Eisai, Eli Lilly, Fujirebio, Merck, Novartis, Novo Nordisk, Roche, Sanofi and Siemens. SP has acquired research support (for the institution) from ki elements / ADDF. In the past 2 years, he has received consultancy/speaker fees from BioArtic, Biogen, Lilly, and Roche.
Références
J Alzheimers Dis. 2020;75(4):1191-1201
pubmed: 32417771
Nat Commun. 2016 Mar 07;7:10918
pubmed: 26948379
Alzheimers Dement. 2019 Apr;15(4):570-580
pubmed: 30639421
Nat Rev Neurol. 2010 Feb;6(2):67-77
pubmed: 20139996
Sci Rep. 2020 Dec 17;10(1):22252
pubmed: 33335244
Am J Psychiatry. 1984 Nov;141(11):1356-64
pubmed: 6496779
Alzheimers Res Ther. 2021 Sep 28;13(1):162
pubmed: 34583745
Lancet. 1985 Jan 5;1(8419):14-6
pubmed: 2856948
Med Image Anal. 2020 Jul;63:101694
pubmed: 32417716
Sci Rep. 2021 Jan 22;11(1):2122
pubmed: 33483535
Sci Rep. 2021 Oct 6;11(1):19853
pubmed: 34615922
Med Image Anal. 2021 Jan;67:101850
pubmed: 33075641
Nat Med. 2021 Jun;27(6):954-963
pubmed: 34083813
Front Neurosci. 2020 May 13;14:259
pubmed: 32477040
Neurology. 2010 Jan 19;74(3):201-9
pubmed: 20042704
Nat Commun. 2022 Jun 20;13(1):3404
pubmed: 35725739
Neuroimage. 2012 Aug 15;62(2):782-90
pubmed: 21979382
JAMA. 2023 Aug 8;330(6):512-527
pubmed: 37459141
N Engl J Med. 2023 Jan 5;388(1):9-21
pubmed: 36449413
Neuroimage. 2019 Apr 1;189:276-287
pubmed: 30654174
Front Neuroinform. 2014 Apr 28;8:44
pubmed: 24817849
Magn Reson Imaging. 2021 May;78:119-126
pubmed: 33588019
Lancet Neurol. 2019 Nov;18(11):985-987
pubmed: 31526626
Nat Med. 2021 Jun;27(6):1034-1042
pubmed: 34031605
Comput Methods Programs Biomed. 2020 Apr;187:105242
pubmed: 31837630
Alzheimers Dement. 2022 Apr;18(4):700-789
pubmed: 35289055
Underst Interpret Mach Learn Med Image Comput Appl (2018). 2018 Sep;11038:79-86
pubmed: 33094293
Acta Psychiatr Scand. 2009 Apr;119(4):252-65
pubmed: 19236314