Premium
[P2–254]: SERUM FERRITIN IS INCREASED IN A SUBSET OF PATIENTS WITH FRONTOTEMPORAL DEMENTIA
Author(s) -
Foiani Martha S.,
Heller Carolin,
Woollacott Ione OC.,
Marshall Charles R.,
Heslegrave Amanda J.,
Warren Jason D.,
Zetterberg Henrik,
Rohrer Jonathan D.
Publication year - 2017
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.2017.06.906
Subject(s) - frontotemporal dementia , progressive supranuclear palsy , ferritin , medicine , primary progressive aphasia , frontotemporal lobar degeneration , gastroenterology , c9orf72 , pathology , dementia , psychology , disease
(AD1⁄433; MCI1⁄452; Normal Controls1⁄4 43). First, we investigate the capacity of the defined oculomotor parameters (in a combination of 556 variables) to distinguish between AD and normal control groups. With this purpose, we selected the most capable parameters using an heuristic algorithm. Then, classification models were trained on a multidimensional representation of eye movements from a subsample of the 3 groups. MCI data was tested with the classifiers after training the model, to verify if MCI subjects would have an oculomotor pattern similar to either AD or normal control group. Results: Automatic classification algorithms were able to distinguish between AD subjects and normal controls, with good levels of performance, having the best results reached 85% of accuracy, 70% of sensitivity and 18% of error. Also, the classifiers successfully classified 18 MCI subjects with an AD oculomotor profile. Conclusions: Different oculomotor parameters, when combined together, significantly improve the ability to accurately distinguish between healthy and impaired subjects. Eye movement analysis reveals the potential to detect early oculomotor deficits in MCI patients similar to AD subjects, and suggests a promising approach for detecting early AD.