Dataset Information

Latent information in fluency lists predicts functional decline in persons at risk for Alzheimer disease.

ABSTRACT: We constructed random forest classifiers employing either the traditional method of scoring semantic fluency word lists or new methods. These classifiers were then compared in terms of their ability to diagnose Alzheimer disease (AD) or to prognosticate among individuals along the continuum from cognitively normal (CN) through mild cognitive impairment (MCI) to AD.Semantic fluency lists from 44 cognitively normal elderly individuals, 80 MCI patients, and 41 AD patients were transcribed into electronic text files and scored by four methods: traditional raw scores, clustering and switching scores, "generalized" versions of clustering and switching, and a method based on independent components analysis (ICA). Random forest classifiers based on raw scores were compared to "augmented" classifiers that incorporated newer scoring methods. Outcome variables included AD diagnosis at baseline, MCI conversion, increase in Clinical Dementia Rating-Sum of Boxes (CDR-SOB) score, or decrease in Financial Capacity Instrument (FCI) score. Receiver operating characteristic (ROC) curves were constructed for each classifier and the area under the curve (AUC) was calculated. We compared AUC between raw and augmented classifiers using Delong's test and assessed validity and reliability of the augmented classifier.Augmented classifiers outperformed classifiers based on raw scores for the outcome measures AD diagnosis (AUC .97 vs. .95), MCI conversion (AUC .91 vs. .77), CDR-SOB increase (AUC .90 vs. .79), and FCI decrease (AUC .89 vs. .72). Measures of validity and stability over time support the use of the method.Latent information in semantic fluency word lists is useful for predicting cognitive and functional decline among elderly individuals at increased risk for developing AD. Modern machine learning methods may incorporate latent information to enhance the diagnostic value of semantic fluency raw scores. These methods could yield information valuable for patient care and clinical trial design with a relatively small investment of time and money.

SUBMITTER: Clark DG

PROVIDER: S-EPMC4039569 | biostudies-literature | 2014 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Latent information in fluency lists predicts functional decline in persons at risk for Alzheimer disease.

Clark D G DG Kapur P P Geldmacher D S DS Brockington J C JC Harrell L L DeRamus T P TP Blanton P D PD Lokken K K Nicholas A P AP Marson D C DC

Cortex; a journal devoted to the study of the nervous system and behavior 20140116

<h4>Objective</h4>We constructed random forest classifiers employing either the traditional method of scoring semantic fluency word lists or new methods. These classifiers were then compared in terms of their ability to diagnose Alzheimer disease (AD) or to prognosticate among individuals along the continuum from cognitively normal (CN) through mild cognitive impairment (MCI) to AD.<h4>Method</h4>Semantic fluency lists from 44 cognitively normal elderly individuals, 80 MCI patients, and 41 AD pa ...[more]

PMID: 24556551

Similar Datasets

Project description:Background Cardiovascular risk factors co-occur with one another, and little is known about the extent of their clustering and risk of Alzheimer disease (AD). We identify groups of cardiovascular risk factors in cognitively normal individuals and investigate between-group differences in incident AD and death. Methods and Results Cognitively normal individuals were recruited from the National Alzheimer's Coordinator Center. A latent class analysis was conducted with hypertension, hypercholesterolemia, heart condition, stroke, smoking history, diabetes, and high body mass index. Between-group differences in the incidence of AD, mortality, and mortality-adjusted AD were investigated. This study included 12 412 cognitively normal individuals (average follow-up, 65 months). Three groups were identified: (1) low probabilities of cardiovascular risk factors (reference; N=5398 [43%]), (2) hypertension and hypercholesterolemia (vascular-dominant; N=5721 [46%]), and (3) hypertension, hypercholesterolemia, diabetes, and high body mass index (vascular-metabolic; N=1293 [10%]). Both vascular groups were significantly older, had more men, were slightly less educated, and were slightly more cognitively impaired than the reference group (all P<0.05). However, only the vascular-metabolic group had a significantly younger age of death compared with the reference group (84.3 versus 88.7 years, P<0.001). Only the vascular-dominant group had a greater incidence of AD (odds ratio [OR], 1.30; P<0.001) compared with the reference group. Mortality was greater in the vascular-dominant (OR, 3.26; P<0.001) and vascular-metabolic groups (OR, 1.84; P=0.02). Mortality-adjusted AD was greater in the vascular-dominant (OR, 1.54; P=0.02) and vascular-metabolic groups (OR, 1.46; P=0.04). Conclusions Three distinct cardiovascular risk factor groups were identified in cognitively normal elderly individuals. Only the vascular-dominant group was associated with a greater incidence of AD. Selective mortality may contribute to the attenuated association between the vascular-metabolic group and incident AD.

Dataset Information

Latent information in fluency lists predicts functional decline in persons at risk for Alzheimer disease.

Publications

Latent information in fluency lists predicts functional decline in persons at risk for Alzheimer disease.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets