Acceleration-based gait analysis for frailty assessment in older adults

Gait parameters and daily physical activity for distinguishing pre-frail, frail, and non-frail older adults: A scoping review

OBJECTIVE

This scoping review aimed to gather current knowledge on accurately identifying and distinguishing between non-frail, pre-frail, and frail older adults using gait and daily physical activity (DPA) parameters and/or models that combine gait with DPA parameters in both controlled and daily life environments.

METHODS

Following PRISMA-ScR guidelines, a systematic search was conducted across seven databases using key terms: "frail", "gait or walk", "IMU", and "age". Studies were included if they focused on gait analysis using Inertial Measurement Units (IMUs) for walking distances greater than 10 meters. Extracted data included study design, gait and DPA outcomes, walking conditions, and classification model performance. Gait parameters were grouped into four domains: spatio-temporal, frequency, amplitude, and dynamic gait. DPA parameters were synthesized into three categories: postural and transition, variability, and physical activity pattern.

RESULTS

A total of 15 cross-sectional studies involving 2,366 participants met the inclusion criteria. Gait analysis showed (pre)frail individuals had slower, shorter steps with longer stride times compared to non-frail individuals. Pre-frail individuals showed distinct gait patterns in periodicity, magnitude range, and variability. In daily activities, (pre)frail individuals displayed shorter, fragmented walking periods and longer transitions between positions. Walking variation identified pre-frail status, showing progressive decreases from non-frail to frail states. Combined gait and daily physical activity models achieved over 97% accuracy, sensitivity and specificity in distinguishing between groups.

DISCUSSION

This review provides an updated synthesis of the relationship between various gait and/or DPA parameters and physical frailty, highlighting gaps in pre-frailty detection and the variability in measurement protocols. It underscores the potential of long-term, sensor-based monitoring of daily physical activity for advancing pre-frailty screening and guiding future clinical trials. Structured Abstract BACKGROUND: Changes in gait and physical activity are critical indicators of frailty. With advancements in wearable sensor technology, long-term gait analysis using acceleration data has become more feasible. However, the contribution of parameters beyond gait speed, such as gait dynamics and daily physical activity (DPA), in identifying frail and pre-frail individuals remains unclear.

OBJECTIVE

This scoping review aimed to gather knowledge on accurately identifying and differentiating physical pre-frail and frail individuals from non-frail individuals using gait parameters alone or models that combine gait and DPA parameters, both in controlled settings and daily life environments.

METHODS

The review followed PRISMA-ScR guidelines. A search strategy incorporating key terms-"frail", "gait or walk", "IMU", and "age"-was applied across seven databases from inception to March 1, 2024. Studies were included if they focused on gait analysis in controlled or daily environments using Inertial Measurement Units (IMUs) and involved walking distances longer than 10 meters. Data on walking conditions, gait outcomes, classification methods, and results were extracted. Gait parameters were categorized into four domains: spatio-temporal, frequency, amplitude, and dynamic gait. DPA parameters were synthesized into three categories: postural and transition, variability, physical activity pattern.

RESULTS

A total of 15 cross-sectional observational studies met the eligibility criteria, covering 2,366 participants, with females representing 27%-80% of the sample and ages ranging from 60 to 92 years. Regarding gait parameters, (pre)frail individuals exhibited longer stride times, slower walking speeds, shorter steps, and reduced cadence compared to non-frail individuals. In three studies, pre-frail could be distinguished from the non-frail and frail group through gait periodicity, range of magnitude, and gait variability. DPA patterns differed between groups, with (pre)frail individuals showing shorter and more fragmented walking periods, brief walking bouts and longer postural transitions. Walking bout variation (CoV) effectively identified pre-frail status, decreasing 53.73% from non-frail to pre-frail, and another 30.87% from pre-frail to frail. Models combining both gait and DPA parameters achieved the highest accuracy (97.25%), sensitivity (98.25%), and specificity (98.25%) in distinguishing between groups.

DISCUSSION

This scoping review provides an updated overview of the current knowledge and gaps in understanding the relationship between gait parameters across different domains and DPA parameters along with physical frailty. Significant variability in gait measurement methods and protocols complicates direct comparisons between studies. The review emphasizes the need for further research, particularly in pre-frailty screening, and underscores the potential of inertial sensor-based long-term monitoring of daily physical activity for future clinical trials.

Digital biomarkers for real-life, home-based monitoring of frailty: a systematic review and meta-analysis

BACKGROUND

Frailty, characterised by decreased physiological function and increased vulnerability to stressors, was associated with an increase in numerous adverse outcomes. Although the number of digital biomarkers for detecting frailty in older adults is increasing, there remains a lack of evidence regarding their effectiveness for early detection and follow-up in real-world, home-based settings.

METHODS

Five databases were searched from inception until 1 August 2024. Standardised forms were utilised for data extraction. The Quality Assessment of Diagnostic Accuracy Studies was used to assess the risk of bias and applicability of included studies. A meta-analysis was conducted to assess the overall sensitivity and specificity for frailty detection.

RESULTS

The systematic review included 16 studies, identifying digital biomarkers relevant for frailty detection, including gait, activity, sleep, heart rate, hand movements and room transition. Meta-analysis further revealed pooled sensitivity of 0.78 [95% confidence interval (CI): 0.70-0.86] and specificity of 0.79 (95% CI: 0.72-0.86) to classify robust and pre-frailty/frailty participants. The overall risk of bias indicated that all the included studies were characterised as having a high or unclear risk of bias.

CONCLUSION

This study offers a thorough characterisation of digital biomarkers for detecting frailty, underscoring their potential for early prediction in home settings. These findings are instrumental in bridging the gap between evidence and practice, enabling more proactive and personalised healthcare monitoring. Further longitudinal studies involving larger sample sizes are necessary to validate the effectiveness of these digital biomarkers as diagnostic tools or prognostic indicators.

On the Necessity of Multidisciplinarity in the Development of at-Home Health Monitoring Platforms for Older Adults: Systematic Review

Background

The growth of aging populations globally has increased the demand for new models of care. At-home, computerized health care monitoring is a growing paradigm, which explores the possibility of reducing workloads, lowering the demand for resource-intensive secondary care, and providing more precise and personalized care. Despite the potential societal benefit of autonomous monitoring systems when implemented properly, uptake in health care institutions is slow, and a great volume of research across disciplines encounters similar common barriers to real-world implementation.

Objective

The goal of this systematic review was to construct an evaluation framework that can assess research in terms of how well it addresses already identified barriers to application and then use that framework to analyze the literature across disciplines and identify trends between multidisciplinarity and the likelihood of research being developed robustly.

Methods

This paper introduces a scoring framework for evaluating how well individual pieces of research address key development considerations using 10 identified common barriers to uptake found during meta-review from different disciplines across the domain of health care monitoring. A scoping review is then conducted using this framework to identify the impact that multidisciplinarity involvement has on the effective development of new monitoring technologies. Specifically, we use this framework to measure the relationship between the use of multidisciplinarity in research and the likelihood that a piece of research will be developed in a way that gives it genuine practical application.

Results

We show that viewpoints of multidisciplinarity; namely across computer science and medicine alongside public and patient involvement (PPI) have a significant positive impact in addressing commonly encountered barriers to application research and development according to the evaluation criteria. Using our evaluation metric, multidisciplinary teams score on average 54.3% compared with 35% for teams made up of medical experts and social scientists, and 2.68 for technical-based teams, encompassing computer science and engineering. Also identified is the significant effect that involving either caregivers or end users in the research in a co-design or PPI-based capacity has on the evaluation score (29.3% without any input and between 48.3% and 36.7% for end user or caregiver input respectively, on average).

Conclusions

This review recommends that, to limit the volume of novel research arbitrarily re-encountering the same issues in the limitations of their work and hence improve the efficiency and effectiveness of research, multidisciplinarity should be promoted as a priority to accelerate the rate of advancement in this field and encourage the development of more technology in this domain that can be of tangible societal benefit.

A review of utility of wearable sensor technologies for older person frailty assessment

Frailty is one of the most concerning aspects of global population aging, and early identification is crucial to prevent or reverse its progression. Simple, universal, and efficient frailty assessment technologies are essential for the timely detection of frailty in older patients. Various multi-dimensional assessment instruments have been developed to quantify frailty phenotypes; we review the literature on wearable sensor technologies leveraged for older person frailty assessment. This review examines representative studies on older person frailty assessment published up to 2024, summarizing pertinent wearable sensor technologies utilized for frailty assessment. Our findings suggest that objective, simple, rapid, and affordable sensor-based frailty screening holds utility across diverse applications including diagnostic aid, prognostication, and endpoint ascertainment in research.

The Fundamentals and Applications of Wearable Sensor Devices in Sports Medicine: A Scoping Review

PURPOSE

To (1) characterize the various forms of wearable sensor devices (WSDs) and (2) review the peer-reviewed literature of applied wearable technology within sports medicine.

METHODS

A systematic search of PubMed and EMBASE databases, from inception through 2023, was conducted to identify eligible studies using WSDs within sports medicine. Data extraction was performed of study demographics and sensor specifications. Included studies were categorized by application: athletic training, rehabilitation, and research.

RESULTS

In total, 43 studies met criteria for inclusion in this review. Forms of WSDs include pedometers, accelerometers, encoders (consisting of magnetometers and gyroscopes), force sensors, global positioning system trackers, and inertial measurement units. Outcome metrics include step counts; gait, limb motion, and angular positioning; foot and skin pressure; change of direction and inclination, including analysis of both body parts and athletes on a field; displacement and velocity of body segments and joints; heart rate; plethysmography; sport-specific kinematics; range of motion, symmetry, and alignment; head impact; sleep; throwing biomechanics; and kinetic and spatiotemporal running metrics. WSDs are used in athletic training to assess sport-specific biomechanics and workload with a goal of injury prevention and training optimization, as well as for rehabilitation monitoring and research such as for risk predicting and aiding diagnosis.

CONCLUSIONS

WSDs enable real-time monitoring of human performance across a variety of implementations and settings, allowing collection of metrics otherwise not achievable. WSDs are powerful tools with multiple applications within athletic training, patient rehabilitation, and orthopaedic and sports medicine research.

CLINICAL RELEVANCE

Wearable technology may represent the missing link to quantitatively addressing return to play and previous performance. WSDs are commercially available and portable adjuncts that allow clinicians, trainers, and individual athletes to monitor biomechanical parameters, workload, and recovery status to better contextualize personalized training, injury risk, and rehabilitation.

Improving Surgical Outcomes for Older Adults with Adoption of Technological Advances in Comprehensive Geriatric Assessment

Frailty is a well-recognized predictor of poor surgical outcomes for older adults, yet effective measurements and interventions remain limited. Technological advances offer an opportunity to address this gap and improve surgical care for older adults. This paper reviews the background of frailty and comprehensive geriatric assessments in surgery, and how technological innovations can advance frailty measurement and intervention in surgical settings. We review two broad areas of technological advancement as applied to frailty in surgery: 1) Innovation in the use of electronic health records (EHR) using Artificial Intelligence (AI) and Machine Learning (ML), and 2) Novel uses for wearable sensors and mobile health (mHealth) applications. We explore the integration of AI and ML with EHR systems, which can surpass traditional comorbidity indices by providing comprehensive health assessments and enhancing prediction models. Innovations like the electronic Frailty Index (eFI) show promise in expanding the reach of frailty assessments and facilitating real-time screening. Additionally, wearable devices and mobile health (mHealth) applications offer new ways to monitor and improve physical activity, nutrition, and psychological well-being, supporting perioperative rehabilitation. While these technologies present challenges, such as the need for infrastructure, training, and data interoperability, they offer promising strategies to facilitate the assessment and management of frailty among surgical patients. Continued research and tailored implementation strategies will be essential to fully realize the potential of these advancements in improving surgical outcomes for frail older adults.

Machine learning approaches for frailty detection, prediction and classification in elderly people: A systematic review

BACKGROUND

Frailty in older people is a syndrome related to aging that is becoming increasingly common and problematic as the average age of the world population increases. Detecting frailty in its early stages or, even better, predicting its appearance can greatly benefit health in later years of life and save the healthcare system from high costs. Machine Learning models fit the need to develop a tool for supporting medical decision-making in detecting or predicting frailty.

METHODS

In this review, we followed the PRISMA methodology to conduct a systematic search of the most relevant Machine Learning models that have been developed so far in the context of frailty. We selected 41 publications and compared them according to their purpose, the type of dataset used, the target variables, and the results they obtained, highlighting their shortcomings and strengths.

RESULTS

The variety of frailty definitions allows many problems to fall into this field, and it is often challenging to compare results due to the differences in target variables. The data types can be divided into gait data, usually collected with sensors, and medical records, often in the context of aging studies. The most common algorithms are well-known models available from every Machine Learning library. Only one study developed a new framework for frailty classification, and only two considered Explainability.

CONCLUSIONS

This review highlights some gaps in the field of Machine Learning applied to the assessment and prediction of frailty, such as the need for a universal quantitative definition. It emphasizes the need for close collaboration between medical professionals and data scientists to unlock the potential of data collected in hospital and clinical settings. As a suggestion for future work, the area of Explainability, which is crucial for models in medicine and health care, was considered in very few studies.