Harnessing digital health to objectively assess cognitive impairment in people undergoing hemodialysis process: The Impact of cognitive impairment on mobility performance measured by wearables

A Scoping Review of Wearable Technologies for Use in Individuals With Intellectual Disabilities and Diabetic Peripheral Neuropathy

BACKGROUND

Individuals with intellectual disabilities (IDs) are at risk of diabetes mellitus (DM) and diabetic peripheral neuropathy (DPN), which can lead to foot ulcers and lower-extremity amputations. However, cognitive differences and communication barriers may impede some methods for screening and prevention of DPN. Wearable and mobile technologies-such as smartphone apps and pressure-sensitive insoles-could help to offset these barriers, yet little is known about the effectiveness of these technologies among individuals with ID.

METHODS

We conducted a scoping review of the databases Embase, PubMed, and Web of Science using search terms for DM, DPN, ID, and technology to diagnose or monitor DPN. Finding a lack of research in this area, we broadened our search terms to include any literature on technology to diagnose or monitor DPN and then applied these findings within the context of ID.

RESULTS

We identified 88 articles; 43 of 88 (48.9%) articles were concerned with gait mechanics or foot pressures. No articles explicitly included individuals with ID as the target population, although three articles involved individuals with other cognitive impairments (two among patients with a history of stroke, one among patients with hemodialysis-related cognitive changes).

CONCLUSIONS

Individuals with ID are not represented in studies using technology to diagnose or monitor DPN. This is a concern given the risk of DM complications among patients with ID and the potential for added benefit of such technologies to reduce barriers to screening and prevention. More studies should investigate how wearable devices can be used among patients with ID.

Establishing the Validity and Reliability of an Online Motor Learning Game: Applications for Alzheimer's Disease Research Within MindCrowd

Objective: Motor practice effects (i.e., improvements in motor task performance with practice) are emerging as a unique variable that can predict Alzheimer's disease (AD) progression and biomarker positivity. However, the tasks used to study motor practice effects have involved face-to-face assessment, making them difficult to integrate into large internet-based cohorts that represent the next generation of AD research. The purpose of this study was to validate an online computer game against its in-lab version, which has been shown previously to characterize motor practice effects. Materials and Methods: This study leveraged young adult participants within the MindCrowd electronic cohort, a large nationwide cohort for AD research collected entirely through the internet. Validation compared performance on the online version among MindCrowd users against an age-matched cohort's performance on an in-lab version using a different controller (Xbox 360 controller joystick for in-lab sample versus keyboard arrow keys for online sample). Results: Data indicated that the rate of skill acquisition among MindCrowd users were not significantly different from those of the in-lab cohort. Furthermore, the contact-to-consent rate observed in this study (although low) was similar to that of other online AD cohorts. Conclusion: Overall, this study demonstrates that implementing online games designed to study and measure motor practice effects into online research cohorts is feasible and valid. Future research will explore how online game performance is associated with age and dementia risk factors that may help further an understanding of AD.

Effect of exercise on cognitive impairment in patients undergoing haemodialyses: A systematic review and meta-analysis of randomised controlled trials

BACKGROUND

The influence of exercise on cognition in patients undergoing haemodialysis has been examined. However, evidence elucidating the effects in this patient group is scarce.

OBJECTIVE

To examine the effect of exercise on cognitive impairment in patients undergoing haemodialysis and provide insight into the effects of various characteristics of exercise on cognitive impairment in this population.

DESIGN

A systematic review and meta-analysis, following the guidance of PRISMA was undertaken.

PARTICIPANTS

Adult patients undergoing haemodialysis.

RESULTS

This review found that exercise significantly improved cognitive impairment in patients undergoing haemodialysis (SMD = 0.37, 95% CI: 0.13, 0.60, p = 0.002). Subgroup analyses demonstrated that both intradialytic exercise (SMD = 0.82, 95% CI: 0.37, 1.26, p < 0.001) and interdialytic exercise (SMD = 0.24, 95% CI: 0.01, 0.47, p = 0.038), exercise for 16 weeks or over (SMD = 0.33, 95% CI: 0.07, 0.58, p = 0.012), and lasting for more than 30 minutes (SMD = 0.52, 95% CI: 0.17, 0.86, p = 0.004) significantly alleviated cognitive impairment. The effect of exercise on cognitive impairment in patients less than 65 years of age (SMD = 0.39, 95% CI: 0.10, 0.68, p = 0.009) was significantly better than those over 65.

CONCLUSION

Exercise significantly improves cognitive impairment in patients undergoing haemodialysis. Both Intradialytic and interdialytic exercise of at least 30 minutes duration, 3 times weekly, and at least for 16 weeks may play a significant role in alleviating cognitive impairment in patients under 65 years of age.

The application of artificial intelligence and custom algorithms with inertial wearable devices for gait analysis and detection of gait-altering pathologies in adults: A scoping review of literature

Background

The purpose of this scoping review was to explore the current applications of objective gait analysis using inertial measurement units, custom algorithms and artificial intelligence algorithms in detecting neurological and musculoskeletal gait altering pathologies from healthy gait patterns.

Methods

Literature searches were conducted of four electronic databases (Medline, PubMed, Embase and Web of Science) to identify studies that assessed the accuracy of these custom gait analysis models with inputs derived from wearable devices. Data was collected according to the preferred reporting items for systematic reviews and meta-analysis statement guidelines.

Results

A total of 23 eligible studies were identified for inclusion in the present review, including 10 custom algorithms articles and 13 artificial intelligence algorithms articles. Nine studies evaluated patients with Parkinson’s disease of varying severity and subtypes. Support vector machine was the commonest adopted artificial intelligence algorithm model, followed by random forest and neural networks. Overall classification accuracy was promising for articles that use artificial intelligence algorithms, with nine articles achieving more than 90% accuracy.

Conclusions

Current applications of artificial intelligence algorithms are reasonably effective discrimination between pathological and non-pathological gait. Of these, machine learning algorithms demonstrate the additional capacity to handle complicated data input, when compared to other custom algorithms. Notably, there has been increasing application of machine learning algorithms for conducting gait analysis. More studies are needed with unsupervised methods and in non-clinical settings to better reflect the community and home-based usage.

Cognitive disorders in patients with chronic kidney disease: specificities of clinical assessment

Neurocognitive disorders are frequent among chronic kidney disease (CKD) patients. Identifying and characterizing cognitive impairment (CI) can help to assess the ability of adherence to CKD risk reduction strategy, identify potentially reversible causes of cognitive decline, modify pharmacotherapy, educate the patient and caregiver and provide appropriate patient and caregiver support. Numerous factors are associated with the development and progression of CI in CKD patients and various conditions can influence the results of cognitive assessment in these patients. Here we review clinical warning signs that should lead to cognitive screening; conditions frequent in CKD at risk to interfere with cognitive testing or performance, including specificities of cognitive assessment in dialysis patients or after kidney transplantation; and available tests for screening and observed cognitive patterns in CKD patients.

Remote, Unsupervised Functional Motor Task Evaluation in Older Adults across the United States Using the MindCrowd Electronic Cohort

COVID-19 has impacted the ability to evaluate motor function in older adults, as motor assessments typically require face-to-face interaction. One hundred seventy-seven older adults nationwide completed an unsupervised functional upper-extremity assessment at home. Data were compared to data from an independent sample of community-dwelling older adults (N = 250) assessed in lab. The effect of age on performance was similar between the in-lab and at-home groups. Practice effects were also similar. Assessing upper-extremity motor function remotely is feasible and reliable in community-dwelling older adults. This test offers a practical solution for telehealth practice and other research involving remote or geographically isolated individuals.

Harnessing digital health to objectively assess cancer-related fatigue: The impact of fatigue on mobility performance

Objective

Cancer-related fatigue (CRF) is highly prevalent among cancer survivors, which may have long-term effects on physical activity and quality of life. CRF is assessed by self-report or clinical observation, which may limit timely diagnosis and management. In this study, we examined the effect of CRF on mobility performance measured by a wearable pendant sensor.

Methods

This is a secondary analysis of a clinical trial evaluating the benefit of exercise in cancer survivors with chemotherapy-induced peripheral neuropathy (CIPN). CRF status was classified based on a Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F) score ≤ 33. Among 28 patients (age = 65.7±9.8 years old, BMI = 26.9±4.1kg/m², sex = 32.9%female) with database variables of interest, twenty-one subjects (75.9%) were classified as non-CRF. Mobility performance, including behavior (sedentary, light, and moderate to vigorous activity (MtV)), postures (sitting, standing, lying, and walking), and locomotion (e.g., steps, postural transitions) were measured using a validated pendant-sensor over 24-hours. Baseline psychosocial, Functional Assessment of Cancer Therapy–General (FACT-G), Falls Efficacy Scale–International (FES-I), and motor-capacity assessments including gait (habitual speed, fast speed, and dual-task speed) and static balance were also performed.

Results

Both groups had similar baseline clinical and psychosocial characteristics, except for body-mass index (BMI), FACT-G, FACIT-F, and FES-I (p<0.050). The groups did not differ on motor-capacity. However, the majority of mobility performance parameters were different between groups with large to very large effect size, Cohen’s d ranging from 0.91 to 1.59. Among assessed mobility performance, the largest effect sizes were observed for sedentary-behavior (d = 1.59, p = 0.006), light-activity (d = 1.48, p = 0.009), and duration of sitting+lying (d = 1.46, p = 0.016). The largest correlations between mobility performance and FACIT-F were observed for sitting+lying (rho = -0.67, p<0.001) and the number of steps per day (rho = 0.60, p = 0.001).

Conclusion

The results of this study suggest that sensor-based mobility performance monitoring could be considered as a potential digital biomarker for CRF assessment. Future studies warrant evaluating utilization of mobility performance to track changes in CRF over time, response to CRF-related interventions, and earlier detection of CRF.