Frontiers in Aging Neuroscience (Dec 2015)

Discrimination between Alzheimer’s Disease and Late Onset Bipolar Disorder using multivariate analysis

  • Ariadna eBesga,
  • Ariadna eBesga,
  • Ariadna eBesga,
  • Itxaso eGonzalez Ortega,
  • Itxaso eGonzalez Ortega,
  • Itxaso eGonzalez Ortega,
  • Ana Maria Gonzalez-Pinto Arrillaga,
  • Ana Maria Gonzalez-Pinto Arrillaga,
  • Enrique eEcheburua,
  • Enrique eEcheburua,
  • Jose Luis eMuñoz Madrigal,
  • Jose Luis eMuñoz Madrigal,
  • Juan Carlos eLeza,
  • Juan Carlos eLeza,
  • Alexandre eSavio,
  • Alexandre eSavio,
  • Manuel eGrana,
  • Manuel eGrana,
  • Darya eChyzhyk,
  • Darya eChyzhyk,
  • Borja eAyerdi

DOI
https://doi.org/10.3389/fnagi.2015.00231
Journal volume & issue
Vol. 7

Abstract

Read online

textbf{Background} Late Onset Bipolar Disorder (LOBD) is often difficultto 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 isincreasing. Both pathologies share pathophysiological features relatedto neuroinflammation. Improved means to differentiate between LOBDand AD in elder subjects will help to select the best personalizedtreatment. textbf{Objective} The aim of this study textcolor{red}{was}to assess the relative significance of clinical observations, neuropsychologicaltests, 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 basedcomputer aided diagnosis (CAD) systems based on these variables. textbf{Materials}A sample of healthy controls (HC) (n=26), AD patients (n=37), andLOBD patients (n=32) textcolor{red}{was} recruited at the Alava UniversityHospital. Clinical observations, neuropsychological tests, and plasmabiomarkers textcolor{red}{were} obtained at recruitment time. textbf{Methods}We appltextcolor{red}{ied} multivariate machine learning classificationmethods to discriminate subjects from HC, AD and LOBD populationsin the study. We analyzetextcolor{red}{d} of feature sets textcolor{red}{combining}clinical observations, neuropshycological measures, and biologicalmarkers, including inflammation biomarkers. textcolor{red}{A featureset containing variables showing significative differences for eachclassification contrast was tested also.} Furthermore, a battery ofclassifier approaches textcolor{red}{were} applied, encompassinglinear and non-linear Support Vector Machines (SVM), Random Forests(RF), Classification and regression trees (CART), and their performancetextcolor{red}{was} evaluated in a leave-one-out textcolor{red}{(LOO)}cross-validation scheme. Post-hoc analysis of Gini index in CART classifiersprovided 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: Discriminationof HC vs. AD patients reaches accuracy 97.21%, AUC 98.17%. Discriminationof LOBD vs. AD patients reaches accuracy 90.26%, AUC 89.57%. Discriminationof HC vs LOBD patients achieves accuracy 95.76%, AUC 88.46%.} textbf{Conclusions}It is feasible to build CAD systems for discrimination among LOBDand AD textcolor{red}{on the basis of a reduced set of clinical variables}to assist the clinician in this difficult differential

Keywords