Premium
P2‐230: Sixty‐Six Percent Screen Failures in a Prospective Multicenter Neuroimaging Trial on the Diagnosis of Clinically Uncertain Cognitive Impairment in Geriatric Inpatients
Author(s) -
Apostolova Ivayla,
Lange Catharina,
Roberts Anna,
Joachim Hans,
Maeurer Anja,
Liese Stephanie,
Estrella Melanie,
Prasad Vikas,
Stechl Elisabeth,
Laemmler Gernot,
Steinhagen-Thiessen Elisabeth,
Buchert Ralph
Publication year - 2016
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.2016.06.1398
Subject(s) - medicine , geriatrics , prospective cohort study , clinical trial , cognitive impairment , neuroimaging , etiology , cognition , pediatrics , emergency medicine , physical therapy , psychiatry
copy of the Apolipoprotein E e4 allele (APOE4) and conversion to AD in males (odds ratio: OR 1⁄4 2.7) and females (OR 1⁄4 4.9). Conclusions: Compared to other reports of machine learning, the classification accuracy of 89% in women is the highest achieved. Our accuracy of 73% in men is comparable to other applications of machine learning to this problem. Therefore, it appears that women with MCI send stronger signals of imminent conversion to AD than men. The results suggest that separate algorithms for the early detection of AD should be designed for men and women, and that the algorithms for women will be more accurate. Structural MRImay be used as an alternative to positron emission tomography (PET) and cerebrospinal fluid (CSF) when making an initial diagnosis.