Endurance and gait speed relationships with mild cognitive impairment and dementia

Machine Learning Approaches for Dementia Detection Through Speech and Gait Analysis: A Systematic Literature Review

Background

Dementia is a general term for several progressive neurodegenerative disorders including Alzheimer's disease. Timely and accurate detection is crucial for early intervention. Advancements in artificial intelligence present significant potential for using machine learning to aid in early detection.

Objective

Summarize the state-of-the-art machine learning-based approaches for dementia prediction, focusing on non-invasive methods, as the burden on the patients is lower. Specifically, the analysis of gait and speech performance can offer insights into cognitive health through clinically cost-effective screening methods.

Methods

A systematic literature review was conducted following the PRISMA protocol (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). The search was performed on three electronic databases (Scopus, Web of Science, and PubMed) to identify the relevant studies published between 2017 to 2022. A total of 40 papers were selected for review.

Results

The most common machine learning methods employed were support vector machine followed by deep learning. Studies suggested the use of multimodal approaches as they can provide comprehensive and better prediction performance. Deep learning application in gait studies is still in the early stages as few studies have applied it. Moreover, including features of whole body movement contribute to better classification accuracy. Regarding speech studies, the combination of different parameters (acoustic, linguistic, cognitive testing) produced better results.

Conclusions

The review highlights the potential of machine learning, particularly non-invasive approaches, in the early prediction of dementia. The comparable prediction accuracies of manual and automatic speech analysis indicate an imminent fully automated approach for dementia detection.

Gait variability predicts cognitive impairment in older adults with subclinical cerebral small vessel disease

Introduction

Advanced methods of gait research, including approaches to quantify variability, and orderliness/regularity/predictability, are increasingly used to identify patients at risk for the development of cognitive impairment. Cerebral small vessel disease (CSVD) is highly prevalent in older adults and is known to contribute to the development of vascular cognitive impairment and dementia (VCID). Studies in preclinical models demonstrate that subclinical alterations precede CSVD-related cognitive impairment in gait coordination. In humans, CSVD also associates with gait abnormalities. The present study was designed to test the hypothesis that increased gait variability and gait asymmetry predict a decline in cognitive performance in older adults with CSVD.

Methods

To test this hypothesis, we compared cognitive performance and gait function in patients with CSVD (age: 69.8 ± 5.3 years; n = 11) and age- and sex-matched control participants (age: 70.7 ± 5.8 years; n = 11). Based on imaging findings, patients with CSVD were identified [presence of white matter hyperintensities plus silent brain infarcts and/or microhemorrhages on magnetic resonance imaging (MRI) assessment]. Cognitive performance was assessed using the Cambridge Neuropsychological Test Automated Battery (CANTAB). Gait parameters were measured during the single and dual tasks, during which participants, in addition to the motor task, completed a series of mental arithmetic calculations. Spatial and temporal parameters of gait variability, symmetry, and permutation entropy were determined using a pressure-sensitive gait mat during single and dual cognitive task conditions.

Results

Patients with CSVD exhibited lower performance in a visual learning test (p = 0.030) and in a sustained attention test (p = 0.007). CSVD also affected step time variability (p = 0.009) and step length variability (p = 0.017). Step lengths of CSVD participants were more asymmetric (p = 0.043) than that of controls, while the two groups were statistically similar regarding step time symmetry and entropy of step time and length. Gait variability was inversely associated with sustained attention, especially among CSVD patients, and this relationship was significantly different between the two groups. The association of sustained attention with gait symmetry was also significantly different between the two groups.

Discussion

Our findings provide additional evidence in support of the concept that increased gait variability and asymmetry may predict cognitive impairment in older adults with CSVD.

Plasma metabolomic profiles of dementia: a prospective study of 110,655 participants in the UK Biobank

BACKGROUND

Plasma metabolomic profile is disturbed in dementia patients, but previous studies have discordant conclusions.

METHODS

Circulating metabolomic data of 110,655 people in the UK Biobank study were measured with nuclear magnetic resonance technique, and incident dementia records were obtained from national health registers. The associations between plasma metabolites and dementia were estimated using Cox proportional hazard models. The 10-fold cross-validation elastic net regression models selected metabolites that predicted incident dementia, and a 10-year prediction model for dementia was constructed by multivariable logistic regression. The predictive values of the conventional risk model, the metabolites model, and the combined model were discriminated by comparison of area under the receiver operating characteristic curves (AUCs). Net reclassification improvement (NRI) was used to estimate the change of reclassification ability when adding metabolites into the conventional prediction model.

RESULTS

Amongst 110,655 participants, the mean (standard deviation) age was 56.5 (8.1) years, and 51 186 (46.3%) were male. A total of 1439 (13.0%) developed dementia during a median follow-up of 12.2 years (interquartile range: 11.5-12.9 years). A total of 38 metabolites, including lipids and lipoproteins, ketone bodies, glycolysis-related metabolites, and amino acids, were found to be significantly associated with incident dementia. Adding selected metabolites (n=24) to the conventional dementia risk prediction model significantly improved the prediction for incident dementia (AUC: 0.824 versus 0.817, p =0.042) and reclassification ability (NRI = 4.97%, P = 0.009) for identifying high risk groups.

CONCLUSIONS

Our analysis identified various metabolomic biomarkers which were significantly associated with incident dementia. Metabolomic profiles also provided opportunities for dementia risk reclassification. These findings may help explain the biological mechanisms underlying dementia and improve dementia prediction.

Gait Speed and Sleep Duration Is Associated with Increased Risk of MCI in Older Community-Dwelling Adults

This study aimed to examine the linear and nonlinear associations between sleep duration and gait speed and the risk of developing mild cognitive impairment (MCI) in community-dwelling older adults. Participants were 233 older adults who met the study inclusion criteria. The MCI diagnosis was based on medical evaluations through a clinical interview conducted by a dementia specialist. Self-reported sleep duration was evaluated using the Pittsburgh Sleep Quality Index. The usual gait speed was calculated from the time taken to walk along a 4 m walkway. Multivariate logistic regression analysis was used to calculate the odds ratio (OR) and the 95% confidence interval (95% CI) of developing MCI in relation to sleep duration and gait speed. Generalized additive models were used to examine the dose−response relationships between sleep duration, gait speed, and the risk of developing MCI. Slower gait speed (OR: 1.84, 95%; CI: 1.00−3.13) and poor sleep duration (OR: 1.76, 95%; CI: 1.00−3.35) were associated with the risk of developing MCI, compared with their optimal status. In addition, the combination of poor sleep and slower gait was associated with a higher risk of developing MCI than optimal sleep duration and gait speed (OR: 3.13, 95%; CI: 1.93−5.14). Furthermore, gait speed and sleep duration were non-linearly associated with the risk of developing MCI. These results highlight the complex interplay and synergism between sleep duration and gait abilities on the risk of developing MCI in older adults. In addition, our results suggest that slower gait speed (<1.0 m/s) and short (<330 min) and long (>480 min) sleep duration may be linked to MCI risks through underlying pathways.

Gait Speed as a Biomarker of Cognitive Vulnerability: A Population-Based Study with Cognitively Normal Older Adults

We aimed to examine associations between cognitive vulnerability and gait speed (GS) in a large older sample. A cross-sectional study analyzed data from the “Health, Lifestyle and Fitness in Adults and Seniors in Amazonas” (SEVAAI) project. In total, 697 participants were included (mean age 70.35 ± 6.86 years). Usual and fast GS were evaluated, and cognitive performance was examined by the COGTEL test battery. There was a positive and large correlation between cognition (COGTEL score) and usual GS (r = 0.510; p < 0.001) and fast GS (r = 0.503; p < 0.001). The usual GS, as a continuous variable, indicated a chance of improved cognitive performance by up to 55%, and fast GS by up to 82%. After controlling for potential confounders (i.e., sex, age, MMSE and years of education), usual and fast GS indicated a chance of improving cognition, respectively, in 57% and 85%. Analysis of GS in quartiles (Q) showed high and significant associations between usual and fast GS and cognitive vulnerability. GS classified as Q1 (slower), Q2 and Q3 represented a greater chance of presenting cognitive deficits, respectively, than in participants with both GS classified as Q4 (highest). Cognitive vulnerability was associated with low GS. Usual and fast GS can be used as complementary measures for the evaluation of cognitively normal Brazilian older adults.