D G Clark1, P Kapur2, D S Geldmacher3, J C Brockington3, L Harrell3, T P DeRamus4, P D Blanton5, K Lokken5, A P Nicholas6, D C Marson3. 1. Birmingham VA Medical Center, USA; Department of Neurology, University of Alabama at Birmingham, USA. Electronic address: dgclark@uab.edu. 2. Department of Biomedical Sciences, Georgia Health Science University, USA. 3. Department of Neurology, University of Alabama at Birmingham, USA. 4. Department of Psychology and Behavioral Neuroscience, University of Alabama at Birmingham, USA. 5. Birmingham VA Medical Center, USA. 6. Birmingham VA Medical Center, USA; Department of Neurology, University of Alabama at Birmingham, USA.
Abstract
OBJECTIVE: 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. METHOD: 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. RESULTS: 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. CONCLUSION: 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. Published by Elsevier Ltd.
OBJECTIVE: 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. METHOD: Semantic fluency lists from 44 cognitively normal elderly individuals, 80 MCI patients, and 41 ADpatients 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. RESULTS: 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. CONCLUSION: 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. Published by Elsevier Ltd.
Authors: K B Walhovd; A M Fjell; J Brewer; L K McEvoy; C Fennema-Notestine; D J Hagler; R G Jennings; D Karow; A M Dale Journal: AJNR Am J Neuroradiol Date: 2010-01-14 Impact factor: 3.825
Authors: Kathleen D Askland; Sarah Garnaat; Nicholas J Sibrava; Christina L Boisseau; David Strong; Maria Mancebo; Benjamin Greenberg; Steve Rasmussen; Jane Eisen Journal: Int J Methods Psychiatr Res Date: 2015-05-21 Impact factor: 4.035
Authors: Kacie D Deters; Kwangsik Nho; Shannon L Risacher; Sungeun Kim; Vijay K Ramanan; Paul K Crane; Liana G Apostolova; Andrew J Saykin Journal: Brain Lang Date: 2017-05-31 Impact factor: 2.381
Authors: David Glenn Clark; Paula M McLaughlin; Ellen Woo; Kristy Hwang; Sona Hurtz; Leslie Ramirez; Jennifer Eastman; Reshil-Marie Dukes; Puneet Kapur; Thomas P DeRamus; Liana G Apostolova Journal: Alzheimers Dement (Amst) Date: 2016-02-15