Objective measurement of gait parameters in healthy and cognitively impaired elderly using the dual-task paradigm

Dual-task gait has prognostic value for cognitive decline in Parkinson's disease

INTRODUCTION

Cognitive decline frequently occurs in individuals with Parkinson's disease (PD), but the clinical methods to predict the onset of cognitive changes are limited. Given preliminary evidence of the link between gait and cognition, the purpose of this study was to determine if dual task (DT) gait was related to declines in cognition over two years in PD.

METHODS

A retrospective two-year longitudinal study of 48 individuals with PD using data from the Parkinson's Progression Markers Initiative of the Michael J. Fox Foundation. The following data were extracted at baseline: spatiotemporal gait (during single and DT), demographics (age, sex), PD disease duration (time since diagnosis), motor function (Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS)), and cognition (Montreal Cognitive Assessment (MoCA)), with MoCA scores also extracted after two years.

RESULTS

A binomial logistic regression was conducted, with all covariates (above) in block 1 and DT effect (DTE) of gait characteristics in block 2 entered in a stepwise fashion. The final model was statistically significant (χ2(6) = 23.20, p < 0.001) and correctly classified 78.7 % of participants by cognitive status after two years. Only DTE of arm swing asymmetry (ASA) (p = 0.030) was included in block 2 such that a 1 % decline in DTE resulted in 1.6 % increased odds of cognitive decline.

CONCLUSIONS

Individuals with greater change in arm swing asymmetry from single to DT gait may be more likely to experience a decline in cognition within two years. These results suggested that reduced automaticity or poor utilization of attentional resources may be indicative of subtle changes in cognition and indicate that DT paradigms may hold promise as a marker of future cognitive decline.

Digital biomarkers for non-motor symptoms in Parkinson's disease: the state of the art

Digital biomarkers that remotely monitor symptoms have the potential to revolutionize outcome assessments in future disease-modifying trials in Parkinson's disease (PD), by allowing objective and recurrent measurement of symptoms and signs collected in the participant's own living environment. This biomarker field is developing rapidly for assessing the motor features of PD, but the non-motor domain lags behind. Here, we systematically review and assess digital biomarkers under development for measuring non-motor symptoms of PD. We also consider relevant developments outside the PD field. We focus on technological readiness level and evaluate whether the identified digital non-motor biomarkers have potential for measuring disease progression, covering the spectrum from prodromal to advanced disease stages. Furthermore, we provide perspectives for future deployment of these biomarkers in trials. We found that various wearables show high promise for measuring autonomic function, constipation and sleep characteristics, including REM sleep behavior disorder. Biomarkers for neuropsychiatric symptoms are less well-developed, but show increasing accuracy in non-PD populations. Most biomarkers have not been validated for specific use in PD, and their sensitivity to capture disease progression remains untested for prodromal PD where the need for digital progression biomarkers is greatest. External validation in real-world environments and large longitudinal cohorts remains necessary for integrating non-motor biomarkers into research, and ultimately also into daily clinical practice.

Medial Temporal Lobe Atrophy in Older Adults With Subjective Cognitive Impairments Affects Gait Parameters in the Spatial Navigation Task

Both navigation abilities and gait can be affected by the atrophy in the medial temporal cortex. This study aimed to determine whether navigation abilities could differentiate seniors with and without medial temporal lobe atrophy who complained about their cognitive status. The participants, classified to either the medial temporal atrophy group (n = 23) or the control group (n = 22) underwent neuropsychological assessment and performed a spatial navigation task while their gait parameters were recorded. The study showed no significant differences between the two groups in memory, fluency, and semantic knowledge or typical measures of navigating abilities. However, gait parameters, particularly the propulsion index during certain phases of the navigation task, distinguished between seniors with and without medial temporal lobe lesions. These findings suggest that the gait parameters in the navigation task may be a valuable tool for identifying seniors with cognitive complaints and subtle medial temporal atrophy.

Ecological validity of a deep learning algorithm to detect gait events from real-life walking bouts in mobility-limiting diseases

Introduction

The clinical assessment of mobility, and walking specifically, is still mainly based on functional tests that lack ecological validity. Thanks to inertial measurement units (IMUs), gait analysis is shifting to unsupervised monitoring in naturalistic and unconstrained settings. However, the extraction of clinically relevant gait parameters from IMU data often depends on heuristics-based algorithms that rely on empirically determined thresholds. These were mainly validated on small cohorts in supervised settings.

Methods

Here, a deep learning (DL) algorithm was developed and validated for gait event detection in a heterogeneous population of different mobility-limiting disease cohorts and a cohort of healthy adults. Participants wore pressure insoles and IMUs on both feet for 2.5 h in their habitual environment. The raw accelerometer and gyroscope data from both feet were used as input to a deep convolutional neural network, while reference timings for gait events were based on the combined IMU and pressure insoles data.

Results and discussion

The results showed a high-detection performance for initial contacts (ICs) (recall: 98%, precision: 96%) and final contacts (FCs) (recall: 99%, precision: 94%) and a maximum median time error of -0.02 s for ICs and 0.03 s for FCs. Subsequently derived temporal gait parameters were in good agreement with a pressure insoles-based reference with a maximum mean difference of 0.07, -0.07, and <0.01 s for stance, swing, and stride time, respectively. Thus, the DL algorithm is considered successful in detecting gait events in ecologically valid environments across different mobility-limiting diseases.

Cognitive-motor dual-task interference in Alzheimer's disease, Parkinson's disease, and prodromal neurodegeneration: A scoping review

BACKGROUND

Cognitive-motor interference (CMI) is a common deficit in Alzheimer's (AD) disease and Parkinson's disease (PD) and may have utility in identification of prodromal neurodegeneration. There is lack of consensus regarding measurement of CMI resulting from dual task paradigms.

RESEARCH QUESTION

How are individuals with AD, PD, and prodromal neurodegeneration impacted by CMI as measured by dual-task (DT) performance?

METHODS

A systematic literature search was performed in six datasets using the PRISMA guidelines. Studies were included if they had samples of participants with AD, PD, or prodromal neurodegeneration and reported at least one measure of cognitive-motor DT performance.

RESULTS

4741 articles were screened and 95 included as part of this scoping review. Articles were divided into three non-mutually exclusive groups based on diagnoses, with 26 articles in AD, 56 articles in PD, and 29 articles in prodromal neurodegeneration, and results presented accordingly.

SIGNIFICANCE

Individuals with AD and PD are both impacted by CMI, though the impact is likely different for each disease. We found a robust body of evidence regarding the utility of measures of DT performance in the detection of subtle deficits in prodromal AD and some signals of utility in prodromal PD. There are several key methodological challenges related to DT paradigms for the measurement of CMI in neurodegeneration. Overall, DT paradigms show good potential as a clinical method to probe specific brain regions, networks, and function; however, task selection and effect measurement should be carefully considered.

Cortical thickness is related to cognitive-motor automaticity and attention allocation in individuals with Alzheimer's disease: a regions of interest study

Alzheimer's disease (AD) is characterized by a distinct pattern of cortical thinning and resultant changes in cognition and function. These result in prominent deficits in cognitive-motor automaticity. The relationship between AD-related cortical thinning and decreased automaticity is not well-understood. We aimed to investigate the relationship between cortical thickness regions-of-interest (ROI) and automaticity and attention allocation in AD using hypothesis-driven and exploratory approaches. We performed an ROI analysis of 46 patients with AD. Data regarding MR images, demographic characteristics, cognitive-motor dual task performance, and cognition were extracted from medical records. Cortical thickness was calculated from MR T1 images using FreeSurfer. Data from the dual task assessment was used to calculate the combined dual task effect (cDTE), a measure of cognitive-motor automaticity, and the modified attention allocation index (mAAI). Four hierarchical multiple linear regression models were conducted regressing cDTE and mAAI separately on (1) hypothesis-generated ROIs and (2) exploratory ROIs. For cDTE, cortical thicknesses explained 20.5% (p = 0.014) and 25.9% (p = 0.002) variability in automaticity in the hypothesized ROI and exploratory models, respectively. The dorsal lateral prefrontal cortex (DLPFC) (β =  - 0.479, p = 0.018) and superior parietal cortex (SPC) (β = 0.467, p = 0.003), and were predictors of automaticity. For mAAI, cortical thicknesses explained 20.7% (p = 0.025) and 28.3% (p = 0.003) variability in attention allocation in the hypothesized ROI and exploratory models, respectively. Thinning of SPC and fusiform gyrus were associated with motor prioritization (β =  - 0.405, p = 0.013 and β =  - 0.632, p = 0.004, respectively), whereas thinning of the DLPFC was associated with cognitive prioritization (β = 0.523, p = 0.022). Cortical thinning in AD was related to cognitive-motor automaticity and task prioritization, particularly in the DLPFC and SPC. This suggests that these regions may play a primary role in automaticity and attentional strategy during dual-tasking.

Volumetric regional MRI and neuropsychological predictors of motor task variability in cognitively unimpaired, Mild Cognitive Impairment, and probable Alzheimer's disease older adults

INTRODUCTION

The mechanisms linking motor function to Alzheimer's disease (AD) progression have not been well studied, despite evidence of AD pathology within motor brain regions. Thus, there is a need for new motor measure that is sensitive and specific to AD.

METHODS

In a sample of 121 older adults (54 cognitive unimpaired [CU], 35 amnestic Mild Cognitive Impairment [aMCI], and 32 probable mild AD), intrasubject standard deviation (ISD) across six trials of a novel upper-extremity motor task was predicted with volumetric regional gray matter and neuropsychological scores using classification and regression tree (CART) analyses.

RESULTS

Both gray matter and neuropsychological CART models indicated that motor task ISD (our measure of motor learning) was related to cortical regions and cognitive test scores associated with memory, executive function, and visuospatial skills. CART models also accurately distinguished motor task ISD of MCI and probable mild AD from CU.

DISCUSSION

Variability in motor task performance across practice trials may be valuable for understanding preclinical and early-stage AD.

SMARTfit Dual-Task Exercise Improves Cognition and Physical Function in Older Adults With Mild Cognitive Impairment: Results of a Community-Based Pilot Study

Mild cognitive impairment is an intermediate state between the cognitive decline often experienced in normal aging and dementia that affects 15% of Americans over 65 years of age. Our communities have an opportunity to support the development and adoption of evidence-based programs to help older adults preserve cognition and physical function. In partnership with a local urban YMCA in an underserved, predominantly minority neighborhood, we tested the appeal and therapeutic benefits of SMARTfit training among older adults with mild cognitive impairment. The participants reported a positive training experience. After 12 weeks of dual-task training, Trail-Making Test and Stroop Color-Word Interference Test scores improved, as did scores on the Short Physical Performance Battery. Results of our SMARTfit dual-task training intervention are encouraging. Larger randomized controlled trials must further investigate the development, implementation, and therapeutic impacts of SMARTfit dual-task training on cognitive and physical function in aging.

Capturing Measures That Matter: The Potential Value of Digital Measures of Physical Behavior for Alzheimer's Disease Drug Development

Alzheimer's disease (AD) is a devastating neurodegenerative disease and the primary cause of dementia worldwide. Despite the magnitude of AD's impact on patients, caregivers, and society, nearly all AD clinical trials fail. A potential contributor to this high rate of failure is that established clinical outcome assessments fail to capture subtle clinical changes, entail high burden for patients and their caregivers, and ineffectively address the aspects of health deemed important by patients and their caregivers. AD progression is associated with widespread changes in physical behavior that have impacts on the ability to function independently, which is a meaningful aspect of health for patients with AD and important for diagnosis. However, established assessments of functional independence remain underutilized in AD clinical trials and are limited by subjective biases and ceiling effects. Digital measures of real-world physical behavior assessed passively, continuously, and remotely using digital health technologies have the potential to address some of these limitations and to capture aspects of functional independence in patients with AD. In particular, measures of real-world gait, physical activity, and life-space mobility captured with wearable sensors may offer value. Additional research is needed to understand the validity, feasibility, and acceptability of these measures in AD clinical research.

A Deep Learning Approach for Gait Event Detection from a Single Shank-Worn IMU: Validation in Healthy and Neurological Cohorts

Many algorithms use 3D accelerometer and/or gyroscope data from inertial measurement unit (IMU) sensors to detect gait events (i.e., initial and final foot contact). However, these algorithms often require knowledge about sensor orientation and use empirically derived thresholds. As alignment cannot always be controlled for in ambulatory assessments, methods are needed that require little knowledge on sensor location and orientation, e.g., a convolutional neural network-based deep learning model. Therefore, 157 participants from healthy and neurologically diseased cohorts walked 5 m distances at slow, preferred, and fast walking speed, while data were collected from IMUs on the left and right ankle and shank. Gait events were detected and stride parameters were extracted using a deep learning model and an optoelectronic motion capture (OMC) system for reference. The deep learning model consisted of convolutional layers using dilated convolutions, followed by two independent fully connected layers to predict whether a time step corresponded to the event of initial contact (IC) or final contact (FC), respectively. Results showed a high detection rate for both initial and final contacts across sensor locations (recall ≥92%, precision ≥97%). Time agreement was excellent as witnessed from the median time error (0.005 s) and corresponding inter-quartile range (0.020 s). The extracted stride-specific parameters were in good agreement with parameters derived from the OMC system (maximum mean difference 0.003 s and corresponding maximum limits of agreement (−0.049 s, 0.051 s) for a 95% confidence level). Thus, the deep learning approach was considered a valid approach for detecting gait events and extracting stride-specific parameters with little knowledge on exact IMU location and orientation in conditions with and without walking pathologies due to neurological diseases.

A Novel Way of Measuring Dual-Task Interference: The Reliability and Construct Validity of the Dual-Task Effect Battery in Neurodegenerative Disease

BACKGROUND

Decreased automaticity is common among individuals with neurodegenerative disease and is often assessed using dual-task (DT) paradigms. However, the best methods for assessing performance changes related to DT demands remain inconclusive.

OBJECTIVE

To investigate the reliability and validity of a novel battery of DT measures (DT Effect-Battery (DTE-B)) encompassing three domains: task-specific interference, task prioritization, and automaticity.

METHODS

Data for this retrospective cross-sectional study included 125 participants with Parkinson's disease (PD), 127 participants with Alzheimer's disease (AD), and 84 healthy older adults. Reliability analyses were conducted using a subset of each population. DTE-B measures were calculated from single and DT performance on the Timed Up and Go test and a serial subtraction task. Construct validity was evaluated via associations within the DTE-B and with theoretically supported measures as well as known-groups validity analyses.

RESULTS

Good to excellent reliability was found for DTE-B measures of task interference (motor and cognitive DT effects) (ICCs≥.658) and automaticity (combined DT effect (cDTE)) (ICCs≥.938). Evidence for convergent validity was found with associations within the hypothesized constructs. Known-groups validity analyses revealed differences in the DTE-B among the healthy group and PD and AD groups (ps≤.001), excepting task prioritization (ps≥.061).

CONCLUSIONS

This study provides evidence to support the DTE-B as a reliable measure of multiple constructs pertinent to DT performance. The cDTE demonstrated evidence to support its validity as a measure of automaticity. Further investigation of the utility of the DTE-B in both PD and AD, as well as other populations, is warranted.

Assessing a Sensory-Motor-Cognition Triad in Amnestic Mild Cognitive Impairment With Dichotic Listening While Walking: A Dual-Task Paradigm

A contemporary topic in aging research relates to the significance of cognitive changes proper to mild cognitive impairment (MCI) to higher risk of falls and gait deteriorations. The present study addresses this question in the amnestic type of MCI (aMCI) by examining a triad of interrelated comorbidities occurring in the MCI condition: attentional impairments, hearing loss and gait disturbances. To this end, we applied a dichotic listening (DL) test during over-ground walking. DL assesses spontaneous and lateralized auditory attention in three conditions (i.e., free report or Non-forced (NF), Forced-Right (FR) ear and Forced-Left (FL) ear). Earlier reports suggest that this dual-task paradigm evoke asymmetric gait effects on healthy controls, which are moderated by degree of hearing loss. Therefore, the aim of the present study was to evaluate the effects of DL on bilateral (data from both limbs) and lateralized (each limb separately) gait outcomes in a group of forty-three aMCI participants (mean = 71.19) and fifty-two healthy older controls (mean = 70.90) by using hearing loss as a covariate in all analyses. Results showed the aMCI group presented overall compromised gait parameters, especially higher gait variability in all DL conditions during lateralized attentional control. These findings were observed bilaterally, and no lateralized effects on gait were observed. Only after controlling for hearing acuity, gait asymmetries on step length variability emerged almost exclusively in healthy controls. It was concluded that hearing loss in the aMCI group together with higher attentional impairments preclude aMCI individuals to properly execute DL and therefore, they do not display gait asymmetries. The present data demonstrate that varied demands on attentional control dependent on hearing acuity affects gait negatively in healthy older adults and aMCI individuals in very different ways. The appearance of asymmetric effects seems to be a perturbation related to normal aging, while the lack of asymmetries but exaggerated gait variability characterizes aMCI. The present findings show the intricate interplay of sensory, cognitive, and motor deteriorations in different group of older adults, which stresses the need of addressing co-occurring comorbidities behind gait perturbations in individuals prone to develop a dementia state.

Kyphosis and bizarre patterns impair spontaneous gait performance in end-of-life mice with Alzheimer's disease pathology while gait is preserved in normal aging

The shorter life spans of mice provide an exceptional experimental gerontology scenario. We previously described increased bizarre (disruptive) behaviors in the 6-month-old 3xTg-AD mice model for Alzheimer's disease (AD), compared to C57BL/6J wildtype (NTg), when confronting new environments. In the present work, we evaluated spontaneous gait and exploratory activity at old age, using 16-month-old mice. Male sex was chosen since sex-dependent psychomotor effects of aging are stronger in NTg males than females and, at this age, male 3 × Tg-AD mice are close to an end-of-life status due to increased mortality rates. Mice's behavior was evaluated in a transparent test box during the neophobia response. Stretching, jumping, backward movements and bizarre circling were identified during the gait and exploratory activity. The results corroborate that in the face of novelty and recognition of places, old 3xTg-AD mice exhibit increased bizarre behaviors than mice with normal aging. Furthermore, bizarre circling and backward movements delayed the elicitation of locomotion and exploration, in an already frail scenario, as shown by highly prevalent kyphosis in both groups. Thus, the translational study of co-occurrence of psychomotor impairments and anxiety-like behaviors can be helpful for understanding and managing the progressive functional deterioration shown in aging, especially in AD.

Dual task walking costs in older adults with mild cognitive impairment: a systematic review and meta-analysis

OBJECTIVE

The objective of this systematic review and meta-analysis (PROSPERO registration No CRD42020192121) is to review existing literature focusing on effects of different dual task paradigms on walking speed in older adults with and without Mild Cognitive Impairment.

METHODS

(1) Data Sources: PubMEd, Cumulative Index of Nursing and Allied Health, Cochrane library, and Web of Science. (2) Study Selection: The key terms searched included those associated with dual task, walking speed, executive function, older adults, and MCI. (3) Data Extraction: The search yielded 140 results with 20 studies meeting the inclusion criteria, which were rated by two independent reviewers using the Quality Assessment Tool. Descriptions of each study including the single and dual task protocol, outcome measure, and final outcomes were extracted. Meta-analysis was performed to evaluate the dual task effects on walking costs in older adults with and without MCI.

RESULTS

Meta-analysis revealed that there were significant differences in the dual task walking costs among older adults with or without MCI (p < .05). Pooled effect sizes of the serial subtraction (9.54; 95%CI, 3.93-15.15) and verbal fluency tasks (10.06; 95%CI, 6.26-15.65) showed that there are higher motor dual-task costs in older adults with MCI than age-matched controls. For quality assessment, all studies ranged from 12 to 16 in score, out of 18 (high quality).

CONCLUSIONS

In the studies included in this review, mental tracking tasks, consisting of serial subtraction and verbal fluency, were found to be the most sensitive in detecting MCI-related changes in older adults, and could serve an important role as a target measure for evaluating the efficacy of interventions aimed at improving cognitive and motor function in older adults.

Corrigendum: Evaluation of Wearable Technology in Dementia: A Systematic Review and Meta-Analysis

[This corrects the article DOI: 10.3389/fmed.2020.501104.].

Functional objective parameters which may discriminate patients with mild cognitive impairment from cognitively healthy individuals: a systematic review and meta-analysis using an instrumented kinematic assessment

BACKGROUND

a systematic review in 2015 showed kinematic gait and balance parameters which can discriminate patients with mild cognitive impairment (MCI) from cognitively healthy individuals.

OBJECTIVE

this systematic review and meta-analysis aims to summarise and synthesise the evidence published after the previous review about the functional objective parameters obtained by an instrumented kinematic assessment which could discriminate patients with MCI from cognitively healthy individuals, as well as to assess the level of evidence per outcome.

METHODS

major electronic databases were searched from inception to August 2019 for cross-sectional studies published after 2015 examining kinematic gait and balance parameters, which may discriminate patients with MCI from cognitively healthy individuals. Meta-analysis was carried out for each parameter reported in two or more studies.

RESULTS

Ten cross-sectional studies with a total of 1,405 patients with MCI and 2,277 cognitively healthy individuals were included. Eight of the included studies reported a low risk of bias. Patients with MCI showed a slower gait speed than cognitively healthy individuals. Thus, single-task gait speed (d = -0.44, 95%CI [-0.60 to -0.28]; P < 0.001), gait speed at fast pace (d = -0.48, 95%CI [-0.72 to -0.24]; P < 0.001) and arithmetic dual-task gait speed (d = -1.20, 95%CI [-2.12 to -0.28]; P = 0.01) were the functional objective parameters which best discriminated both groups.

CONCLUSION

the present review shows kinematic gait parameters which may discriminate patients with MCI from cognitively healthy individuals. Most of the included studies reported a low risk of bias, but the grading of recommendations assessment, development and evaluation criteria showed a low level of evidence per outcome.

Evaluation of Wearable Technology in Dementia: A Systematic Review and Meta-Analysis

Background: The objective of this analysis was to systematically review studies employing wearable technology in patients with dementia by quantifying differences in digitally captured physiological endpoints.

Methods: This systematic review and meta-analysis was based on web searches of Cochrane Database, PsycInfo, Pubmed, Embase, and IEEE between October 25–31st, 2017. Observational studies providing physiological data measured by wearable technology on participants with dementia with a mean age ≥50. Data were extracted according to PRISMA guidelines and methodological quality assessed independently using Downs and Black criteria. Standardized mean differences between cases and controls were estimated using random-effects models.

Results: Forty-eight studies from 18,456 screened abstracts (Dementia: n = 2,516, Control: n = 1,224) met inclusion criteria for the systematic review. Nineteen of these studies were included in one or multiple meta-analyses (Dementia: n = 617, Control: n = 406). Participants with dementia demonstrated lower levels of daily activity (standardized mean difference (SMD), −1.60; 95% CI, −2.66 to −0.55), decreased sleep efficiency (SMD, −0.52; 95% CI, −0.89 to −0.16), and greater intradaily circadian variability (SMD, 0.46; 95% CI, 0.27 to 0.65) than controls, among other measures. Statistical between-study heterogeneity was observed, possibly due to variation in testing duration, device type or patient setting.

Conclusions and Relevance: Digitally captured data using wearable devices revealed that adults with dementia were less active, demonstrated increased fragmentation of their sleep-wake cycle and a loss of typical diurnal variation in circadian rhythm as compared to controls.

The use of Motor and Cognitive Dual-Task quantitative assessment on subjects with mild cognitive impairment: A systematic review

Dementia and Alzheimer's Disease (AD) represent a health emergency. The identification of valid and noninvasive markers to identify people with Mild Cognitive Impairment (MCI) is profoundly advocated. This review outlines the use of quantitative Motor and Cognitive Dual-Task (MCDT) on MCI, by technologies aid. We describe the framework and the most valuable researches, displaying the adopted protocols, and the available technologies. PubMed Central, Web of Science, and Scopus were inspected between January 2010 and May 2020. 1939 articles were found in the initial quest. Exclusion criteria allowed the selection of the most relevant papers; 38 papers were included. The articles, regarding four technological solutions "wearable sensors", "personal devices", "optokinetic systems", and "electronic walkways", are organized into three categories: "Quantitative MCDT", "MCDT Inspired by Neuropsychological Test", and "MCDT for MCI Stimulation". MCDT might furnish clinical landmarks, supplying aid for disease stratication, risk prediction, and intervention optimization. Such protocols could foster the use of data mining and machine learning techniques. Notwithstanding, there is still a need to standardize and harmonize such protocols.

Dual-Task Gait Assessment and Machine Learning for Early-detection of Cognitive Decline

Alzheimer's disease (AD) affects approximately 30 million people worldwide, and this number is predicted to triple by 2050 unless further discoveries facilitate the early detection and prevention of the disease. Computerized walkways for simultaneous assessment of motor-cognitive performance, known as a dual-task assessment, has been used to associate changes in gait characteristics to mild cognitive impairment (MCI) with early-stage disease. However, to our best knowledge, there is no validated method to detect MCI using the collective analysis of these gait characteristics. In this paper, we develop a machine learning approach to analyze the gait data from the dual-task assessment in order to detect subjects with cognitive impairment from healthy individuals. We collected dual-task gait data from a computerized walkway of a total of 92 subjects with 31 healthy control (HC) and 61 MCI. Using support vector machine (SVM) and gradient tree boosting, we developed a classifier to differentiate MCI from HC subjects and compared the results with a paper-based questionnaire assessment that has been commonly used in clinical practice. SVM provided the highest accuracy of 77.17% with 81.97% sensitivity and 67.74% specificity. Our results indicate the potential of machine learning + dual-task assessment to enable early diagnosis of cognitive decline before it advances to dementia and AD, so that early intervention or prevention strategies can be initiated.

Detection of Mild Cognitive Impairment and Alzheimer's Disease using Dual-task Gait Assessments and Machine Learning

Objective

Early detection of mild cognitive impairment (MCI) and Alzheimer's disease (AD) can increase access to treatment and assist in advance care planning. However, the development of a diagnostic system that d7oes not heavily depend on cognitive testing is a major challenge. We describe a diagnostic algorithm based solely on gait and machine learning to detect MCI and AD from healthy.

Methods

We collected "single-tasking" gait (walking) and "dual-tasking" gait (walking with cognitive tasks) from 32 healthy, 26 MCI, and 20 AD participants using a computerized walkway. Each participant was assessed with the Montreal Cognitive Assessment (MoCA). A set of gait features (e.g., mean, variance and asymmetry) were extracted. Significant features for three classifications of MCI/healthy, AD/healthy, and AD/MCI were identified. A support vector machine model in a one-vs.-one manner was trained for each classification, and the majority vote of the three models was assigned as healthy, MCI, or AD.

Results

The average classification accuracy of 5-fold cross-validation using only the gait features was 78% (77% F1-score), which was plausible when compared with the MoCA score with 83% accuracy (84% F1-score). The performance of healthy vs. MCI or AD was 86% (88% F1-score), which was comparable to 88% accuracy (90% F1-score) with MoCA.

Conclusion

Our results indicate the potential of machine learning and gait assessments as objective cognitive screening and diagnostic tools.

Significance

Gait-based cognitive screening can be easily adapted into clinical settings and may lead to early identification of cognitive impairment, so that early intervention strategies can be initiated.

Verification, analytical validation, and clinical validation (V3): the foundation of determining fit-for-purpose for Biometric Monitoring Technologies (BioMeTs)

Digital medicine is an interdisciplinary field, drawing together stakeholders with expertize in engineering, manufacturing, clinical science, data science, biostatistics, regulatory science, ethics, patient advocacy, and healthcare policy, to name a few. Although this diversity is undoubtedly valuable, it can lead to confusion regarding terminology and best practices. There are many instances, as we detail in this paper, where a single term is used by different groups to mean different things, as well as cases where multiple terms are used to describe essentially the same concept. Our intent is to clarify core terminology and best practices for the evaluation of Biometric Monitoring Technologies (BioMeTs), without unnecessarily introducing new terms. We focus on the evaluation of BioMeTs as fit-for-purpose for use in clinical trials. However, our intent is for this framework to be instructional to all users of digital measurement tools, regardless of setting or intended use. We propose and describe a three-component framework intended to provide a foundational evaluation framework for BioMeTs. This framework includes (1) verification, (2) analytical validation, and (3) clinical validation. We aim for this common vocabulary to enable more effective communication and collaboration, generate a common and meaningful evidence base for BioMeTs, and improve the accessibility of the digital medicine field.

Estimation of spatio-temporal parameters of gait from magneto-inertial measurement units: multicenter validation among Parkinson, mildly cognitively impaired and healthy older adults

Background

The use of miniaturized magneto-inertial measurement units (MIMUs) allows for an objective evaluation of gait and a quantitative assessment of clinical outcomes. Spatial and temporal parameters are generally recognized as key metrics for characterizing gait. Although several methods for their estimate have been proposed, a thorough error analysis across different pathologies, multiple clinical centers and on large sample size is still missing. The aim of this study was to apply a previously presented method for the estimate of spatio-temporal parameters, named Trusted Events and Acceleration Direct and Reverse Integration along the direction of Progression (TEADRIP), on a large cohort (236 patients) including Parkinson, mildly cognitively impaired and healthy older adults collected in four clinical centers. Data were collected during straight-line gait, at normal and fast walking speed, by attaching two MIMUs just above the ankles. The parameters stride, step, stance and swing durations, as well as stride length and gait velocity, were estimated for each gait cycle. The TEADRIP performance was validated against data from an instrumented mat.

Results

Limits of agreements computed between the TEADRIP estimates and the reference values from the instrumented mat were − 27 to 27 ms for Stride Time, − 68 to 44 ms for Stance Time, − 31 to 31 ms for Step Time and − 67 to 52 mm for Stride Length. For each clinical center, the mean absolute errors averaged across subjects for the estimation of temporal parameters ranged between 1 and 4%, being on average less than 3% (< 30 ms). Stride length mean absolute errors were on average 2% (≈ 25 mm). Error comparisons across centers did not show any significant difference. Significant error differences were found exclusively for stride and step durations between healthy elderly and Parkinsonian subjects, and for the stride length between walking speeds.

Conclusions

The TEADRIP method was effectively validated on a large number of healthy and pathological subjects recorded in four different clinical centers. Results showed that the spatio-temporal parameters estimation errors were consistent with those previously found on smaller population samples in a single center. The combination of robustness and range of applicability suggests the use of the TEADRIP as a suitable MIMU-based method for gait spatio-temporal parameter estimate in the routine clinical use. The present paper was awarded the “SIAMOC Best Methodological Paper 2017”.