AUTHOR=Besga Ariadna , Gonzalez Itxaso , Echeburua Enrique , Savio Alexandre , Ayerdi Borja , Chyzhyk Darya , Madrigal Jose L. M. , Leza Juan C. , Graña Manuel , Gonzalez-Pinto Ana Maria 

TITLE=Discrimination between Alzheimer’s Disease and Late Onset Bipolar Disorder Using Multivariate Analysis

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 7 - 2015

YEAR=2015

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2015.00231

DOI=10.3389/fnagi.2015.00231

ISSN=1663-4365

ABSTRACT=textbf{Background} Late Onset Bipolar Disorder (LOBD) is often difficult
to distinguish from degenerative dementias, such as Alzheimer Disease
(AD), due to comorbidities and common cognitive symptoms. Moreover,
LOBD prevalence in the elder population is not negligible and it is
increasing. Both pathologies share pathophysiological features related
to neuroinflammation. Improved means to differentiate between LOBD
and AD in elder subjects will help to select the best personalized
treatment. \textbf{Objective} The aim of this study \textcolor{red}{was}
to assess the relative significance of clinical observations, neuropsychological
tests, and \textcolor{red}{specific} \textcolor{red}{blood plasma
}biomarkers (inflammatory and neurotrophic), separately and combined,
in the \textcolor{red}{differential diagnosis} of LOBD versus AD.
The \textcolor{red}{significance} assessment \textcolor{red}{was}
carried out evaluating the accuracy achieved by classification based
computer aided diagnosis (CAD) systems based on these variables. \textbf{Materials}
A sample of healthy controls (HC) (n=26), AD patients (n=37), and
LOBD patients (n=32) \textcolor{red}{was} recruited at the Alava University
Hospital. Clinical observations, neuropsychological tests, and plasma
biomarkers \textcolor{red}{were} obtained at recruitment time. \textbf{Methods}
We appl\textcolor{red}{ied} multivariate machine learning classification
methods to discriminate subjects from HC, AD and LOBD populations
in the study. We analyze\textcolor{red}{d} of feature sets \textcolor{red}{combining}
clinical observations, neuropshycological measures, and biological
markers, including inflammation biomarkers. \textcolor{red}{A feature
set containing variables showing significative differences for each
classification contrast was tested also.} Furthermore, a battery of
classifier approaches \textcolor{red}{were} applied, encompassing
linear and non-linear Support Vector Machines (SVM), Random Forests
(RF), Classification and regression trees (CART), and their performance
\textcolor{red}{was} evaluated in a leave-one-out \textcolor{red}{(LOO)}
cross-validation scheme. Post-hoc analysis of Gini index in CART classifiers
provided a measure of each variable importance. \textbf{Results}
Welch's t-test found one biomarker (Malondialdehyde)
with significative differences (p<0.001) in LOBD vs. AD contrast.
Classification results with the best features are as follows: Discrimination
of HC vs. AD patients reaches accuracy 97.21\%, AUC 98.17\%. Discrimination
of LOBD vs. AD patients reaches accuracy 90.26\%, AUC 89.57\%. Discrimination
of HC vs LOBD patients achieves accuracy 95.76\%, AUC 88.46\%.} \textbf{Conclusions}
It is feasible to build CAD systems for discrimination among  LOBD
and AD \textcolor{red}{on the basis of a reduced set of clinical variables}
to assist the clinician in this difficult differential