Short-Term Memory Binding Distinguishing Amnestic Mild Cognitive Impairment from Healthy Aging: A Machine Learning Study

Abstract

Background: Amnestic mild cognitive impairment (aMCI) is the most common preclinical stage of Alzheimer’s disease (AD). A strategy to reduce the impact of AD is the early aMCI diagnosis and clinical intervention. Neuroimaging, neurobiological, and genetic markers have proved to be sensitive and specific for the early diagnosis of AD. However, the high cost of these procedures is prohibitive in low-income and middle-income countries (LIMCs). The neuropsychological assessments currently aim to identify cognitive markers that could contribute to the early diagnosis of dementia. Objective: Compare machine learning (ML) architectures classifying and predicting aMCI and asset the contribution of cognitive measures including binding function in distinction and prediction of aMCI. Methods: We conducted a two-year follow-up assessment of a sample of 154 subjects with a comprehensive multidomain neuropsychological battery. Statistical analysis was proposed using complete ML architectures to compare subjects’ performance to classify and predict aMCI. Additionally, permutation importance and Shapley additive explanations (SHAP) routines were implemented for feature importance selection. Results: AdaBoost, gradient boosting, and XGBoost had the highest performance with over 80%success classifying aMCI, and decision tree and random forest had the highest performance with over 70%success predictive routines. Feature importance points, the auditory verbal learning test, short-term memory binding tasks, and verbal and category fluency tasks were used as variables with the first grade of importance to distinguish healthy cognition and aMCI. Conclusion: Although neuropsychological measures do not replace biomarkers’ utility, it is a relatively sensitive and specific diagnostic tool for aMCI. Further studies with ML must identify cognitive performance that differentiates conversion from average MCI to the pathological MCI observed in AD.

Sex Differences in Memory: Do Female Reproductive Factors Explain the Differences?

“…However, a similar association with our finding was found, showing that a longer reproductive period was related with a higher risk of all-cause dementia and Alzheimer's disease in a 44year longitudinal population study (43). It may indicate that memory impairment was a strong predictor of future dementia (1)(2)(3)(4). The risk of objective memory impairment in women who experienced menopause at the age of was the lowest one among groups of menopausal age and comparable with the risk in men.…”

“…With the rapid increase of the older population, cognitive decline is becoming an important topic in relevant areas of clinic and research. Among various domains of cognition, episodic memory impairment was a prominent risk factor of future mild cognitive impairment (1) and dementia (2), and it could increase mortality at 10 years in the older adults without dementia (3). It was also suggested as a required clinical phenotype in the diagnostic criteria for Alzheimer's disease (4).…”

Short-term memory conjunctive binding in Alzheimer’s disease: A systematic review and meta-analysis.

“…We aimed to analyze the performance of patients within the ADcontinuum and related healthy controls on conjunctive STM binding tests. Different paradigms were included: change detection (Cecchini et al, 2020(Cecchini et al, , 2021Della Sala et al, 2018;Fernández et al, 2018;Fernández & Parra, 2021;Koppara et al, 2015b;Kozlova et al, 2021;Martínez-Florez et al, 2021;Norton et al, 2020;Parra et al, 2010aParra et al, , 2010bParra et al, , 2011bParra et al, , 2015bParra et al, , 2017aParra et al, , 2019Pietto et al, 2016;Valdés Hernández et al, 2020), free recall (Cecchini et al, 2017a(Cecchini et al, , 2020Della Sala et al, 2012;Parra et al, 2009b), and cued recall modalities (Guazzo et al, 2020), different set sizes This document is copyrighted by the American Psychological Association or one of its allied publishers.…”

“…They also used different set sizes in the task, ranging from 2 to 4 items per trial, and some studies titrated the cognitive load by using different set sizes for the control and experimental groups. All studies reviewed used small sample sizes, except for Martínez-Florez et al (2021), which included 109 controls and 45 MCIs. The other studies varied from 6 to 37 participants within each group, with a mean sample size of 22.24 and a median of 23 participants per group.…”

“…However, longitudinal studies should be carried out to verify whether the STM binding test is a good predictor of dementia in patients with suspected AD. Only one such study could be gleaned from the literature (Martínez-Florez et al, 2021). However, in this study the authors did not test the STM binding as a predictor of MCI to AD dementia conversion but used machine learning techniques to extract classifiers distinguishing between controls and MCI at baseline and tested their accuracy at the follow-up assessment (i.e., by comparing MCI vs. controls).…”