Multi-feature computational framework for combined signatures of dementia in underrepresented settings
Journal
Journal of Neural Engineering
ISSN
1741-2560
1741-2552
Date Issued
2022
Author(s)
Sebastian Moguilner
Agustina Birba
Sol Fittipaldi
Cecilia Gonzalez-Campo
Enzo Tagliazucchi
REYES, PABLO
Diana Matallana
Mario A Parra
Gonzalo Farías
Josefina Cruzat
Adolfo García
Harris A Eyre
Renaud La Joie
Gil Rabinovici
Robert Whelan
Agustín Ibáñez
Type
Resource Types::text::journal::journal article
URL Institutional Repository
Abstract
<jats:title>Abstract</jats:title>
<jats:p>
<jats:italic>Objective.</jats:italic> The differential diagnosis of behavioral variant frontotemporal dementia (bvFTD) and Alzheimer’s disease (AD) remains challenging in underrepresented, underdiagnosed groups, including Latinos, as advanced biomarkers are rarely available. Recent guidelines for the study of dementia highlight the critical role of biomarkers. Thus, novel cost-effective complementary approaches are required in clinical settings. <jats:italic>Approach</jats:italic>. We developed a novel framework based on a gradient boosting machine learning classifier, tuned by Bayesian optimization, on a multi-feature multimodal approach (combining demographic, neuropsychological, magnetic resonance imaging (MRI), and electroencephalography/functional MRI connectivity data) to characterize neurodegeneration using site harmonization and sequential feature selection. We assessed 54 bvFTD and 76 AD patients and 152 healthy controls (HCs) from a Latin American consortium (ReDLat). <jats:italic>Main results</jats:italic>. The multimodal model yielded high area under the curve classification values (bvFTD patients vs HCs: 0.93 (±0.01); AD patients vs HCs: 0.95 (±0.01); bvFTD vs AD patients: 0.92 (±0.01)). The feature selection approach successfully filtered non-informative multimodal markers (from thousands to dozens). <jats:italic>Results</jats:italic>. Proved robust against multimodal heterogeneity, sociodemographic variability, and missing data. <jats:italic>Significance</jats:italic>. The model accurately identified dementia subtypes using measures readily available in underrepresented settings, with a similar performance than advanced biomarkers. This approach, if confirmed and replicated, may potentially complement clinical assessments in developing countries.</jats:p>
<jats:p>
<jats:italic>Objective.</jats:italic> The differential diagnosis of behavioral variant frontotemporal dementia (bvFTD) and Alzheimer’s disease (AD) remains challenging in underrepresented, underdiagnosed groups, including Latinos, as advanced biomarkers are rarely available. Recent guidelines for the study of dementia highlight the critical role of biomarkers. Thus, novel cost-effective complementary approaches are required in clinical settings. <jats:italic>Approach</jats:italic>. We developed a novel framework based on a gradient boosting machine learning classifier, tuned by Bayesian optimization, on a multi-feature multimodal approach (combining demographic, neuropsychological, magnetic resonance imaging (MRI), and electroencephalography/functional MRI connectivity data) to characterize neurodegeneration using site harmonization and sequential feature selection. We assessed 54 bvFTD and 76 AD patients and 152 healthy controls (HCs) from a Latin American consortium (ReDLat). <jats:italic>Main results</jats:italic>. The multimodal model yielded high area under the curve classification values (bvFTD patients vs HCs: 0.93 (±0.01); AD patients vs HCs: 0.95 (±0.01); bvFTD vs AD patients: 0.92 (±0.01)). The feature selection approach successfully filtered non-informative multimodal markers (from thousands to dozens). <jats:italic>Results</jats:italic>. Proved robust against multimodal heterogeneity, sociodemographic variability, and missing data. <jats:italic>Significance</jats:italic>. The model accurately identified dementia subtypes using measures readily available in underrepresented settings, with a similar performance than advanced biomarkers. This approach, if confirmed and replicated, may potentially complement clinical assessments in developing countries.</jats:p>
Subjects
feature selection
;
harmonization
;
machine learning
;
multimodal neuroimaging
;
neurodegeneration
;
alzheimer disease
;
bayes theorem
;
biomarkers
;
brain
;
frontotemporal dementia
;
humans
;
magnetic resonance imaging
;
neuropsychological tests
;
adaptive boosting
;
cost effectiveness
;
electroencephalography
;
electrophysiology
;
feature selection
;
gradient methods
;
magnetic resonance imaging
;
neurodegenerative diseases
;
neuroimaging
;
site selection
;
biological marker
;
biological marker
;
alzheimers disease
;
features selection
;
frontotemporal dementias
;
harmonisation
;
healthy controls
;
machine-learning
;
multi-modal
;
multifeatures
;
multimodal neuroimaging
;
neurodegeneration
;
aged
;
article
;
bayesian network
;
clinical feature
;
comparative study
;
controlled study
;
cost effectiveness analysis
;
electroencephalography
;
feature selection
;
female
;
frontal variant frontotemporal dementia
;
functional magnetic resonance imaging
;
human
;
machine learning
;
major clinical study
;
male
;
nerve degeneration
;
neuropsychology
;
nuclear magnetic resonance imaging
;
patient coding
;
receiver operating characteristic
;
sociodemographics
;
alzheimer disease
;
bayes theorem
;
brain
;
frontotemporal dementia
;
neuropsychological test
;
pathology
;
procedures
;
psychology
;
biomarkers