Premium
P1‐445: AUTOMATIC ASSESSMENT OF CONVENTIONAL COGNITIVE TESTS FOR DIFFERENTIATING MILD COGNITIVE IMPAIRMENT: A PROOF OF CONCEPT STUDY OF THE DIGIT SPAN TASK
Author(s) -
Asgari Meysam,
Gale Robert,
Kaye Jeffrey A.,
Dodge Hiroko H.
Publication year - 2019
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.2019.06.1050
Subject(s) - memory span , task (project management) , set (abstract data type) , cognition , numerical digit , computer science , cognitive test , test (biology) , artificial intelligence , speech recognition , pattern recognition (psychology) , arithmetic , natural language processing , machine learning , working memory , psychology , mathematics , engineering , systems engineering , neuroscience , paleontology , biology , programming language
functioning, with results shown to be unaffected by language, educational level, or cultural background. Our study aims to assess the capacity of CG to predict future cognitive decline in mild cognitive impairment (MCI) and cognitively normal (CN) subjects. Baseline data is presented here. Methods: Subjects were characterized by a full cognitive assessment at a tertiary memory clinic. 13 MCI and 8 CN subjects are presented here (target 30 MCI and 30 CN). CG testing and MoCA testing sessions will be performed at baseline, 3, 6, 9, 12, 24, and 36 months. Clinical and neuropsychological evaluations will be performed at baseline, 12, 24, and 36 months. Results: MCI participants’ average age is 77.8 years old and 70.0 years old for CN. 23% of MCI and 100% of CN participants are female. The average baseline MoCA score for MCI was 22.2 and 28.13 for CN. The average baseline CG composite score for psychomotor function/attention for MCI was 85.0 and 97.0 for CN. The average baseline CG composite score for learning/ working memory was 89.0 for MCI and 100.0 for CN participants. Conclusions: In this baseline data, both average CG composite scores forMCI participants werewithin the borderline performance range of 80-89, compared to the CN participants whose scores were within the normal performance range of 92-150. Preliminary findings to date show that CG can help distinguish cognitive differences between MCI and CN participants at baseline. The next three years of data collection will hopefully identify early changes in CG scoring that are predictive of further decline in the subset of patients with MCI that will convert to dementia.