Premium
IC‐P‐090: Combining Cortical Thickness Analysis and Clinical Measures to Predict Alzheimer's Disease
Author(s) -
Julkunen Valtteri J.,
Koikkalainen Juha,
Niskanen Eini,
Pihlajamäki Maija,
Herukka Sanna-Kaisa,
Hallikainen Merja,
Kivipelto Miia,
Vanninen Ritva,
Lötjönen Jyrki,
Soininen Hilkka
Publication year - 2010
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2010.05.104
Subject(s) - parahippocampal gyrus , sulcus , naive bayes classifier , artificial intelligence , support vector machine , posterior cingulate , precuneus , medicine , psychology , temporal lobe , pattern recognition (psychology) , cognition , computer science , neuroscience , epilepsy
clinical and research reasons. Identifying these subjects as soon as possible is crucial, first for the patient benefit and second for speeding up and track the effect of different therapeutical interventions. We hypothesise that the prediction will be facilitated by structural MRI investigating whole brain atrophy, as well as regional alterations in cortical features (grey matter volume, thickness, and surface area) by using serial MRI scans, with a three months follow up period. Methods: This study focused on the MCI group (N 1⁄4 92), out of which 23 were diagnosed with AD 12 months from the baseline clinical assessment. Whole brain atrophy was assessed by iterative principal component analysis (IPCA). The Fischl and Dale pipeline provided measures of regional brain atrophy and cortical thickness. The structural MRI data collected at baseline and at 3 months were analysed cross-sectionally and longitudinally using multivariate statistical models. Results: Preliminary analyses point out that whole-brain measurements are not robust predictors of conversion to AD. Conclusions: This is the first study exploring structural MRI data with only 3 months follow-up period. Although a 3 months follow up period may seem too short given the complexity of neurodegenerative processes leading to overt brain atrophy, we are currently only exploring the potential information provided by the wealth of data produced by complex image analysis procedures.