Validation of Walking Speed Estimation from Trunk Mounted Accelerometers for a Range of Walking Speeds.

Walking speed is a strong indicator of the health status of older people and patients. Using algorithms, the walking speed can be estimated from wearable accelerometers, which enables minimally obtrusive (longitudinal) monitoring. We evaluated the performance of two algorithms, the inverted pendulum (IP) algorithm, and a novel adaptation correcting for lateral step movement, which aimed to improve accuracy during slow walking. To evaluate robustness, we gathered data from different groups (healthy adults, elderly, and elderly patients) of volunteers (n = 159) walking under various conditions (over ground, treadmill, using walking aids) at a broad range of speeds (0.11-1.93 m/s). Both of the algorithms showed good agreement with the reference values and similar root-mean-square errors (RMSEs) for walking speeds ≥0.5 m/s, which ranged from 0.09-0.16 m/s for the different positions, in line with the results from others. However, for slower walking, RMSEs were significantly better for the new method (0.06-0.09 m/s versus 0.15-0.19 m/s). Pearson correlation improved for speeds <0.5 m/s (from 0.67-0.72 to 0.73-0.82) as well as higher speeds (0.87-0.97 to 0.90-0.98) with the new method. Overall, we found that IP(-based) walking speed estimation proved to be applicable for a variety of wearing positions, conditions and speeds, indicating its potential value for health assessment applications.

Detection of Talking in Respiratory Signals: A Feasibility Study Using Machine Learning and Wearable Textile-Based Sensors.

Social isolation and loneliness are major health concerns in young and older people. Traditional approaches to monitor the level of social interaction rely on self-reports. The goal of this study was to investigate if wearable textile-based sensors can be used to accurately detect if the user is talking as a future indicator of social interaction. In a laboratory study, fifteen healthy young participants were asked to talk while performing daily activities such as sitting, standing and walking. It is known that the breathing pattern differs significantly between normal and speech breathing (i.e., talking). We integrated resistive stretch sensors into wearable elastic bands, with a future integration into clothing in mind, to record the expansion and contraction of the chest and abdomen while breathing. We developed an algorithm incorporating machine learning and evaluated its performance in distinguishing between periods of talking and non-talking. In an intra-subject analysis, our algorithm detected talking with an average accuracy of 85%. The highest accuracy of 88% was achieved during sitting and the lowest accuracy of 80.6% during walking. Complete segments of talking were correctly identified with 96% accuracy. From the evaluated machine learning algorithms, the random forest classifier performed best on our dataset. We demonstrate that wearable textile-based sensors in combination with machine learning can be used to detect when the user is talking. In the future, this approach may be used as an indicator of social interaction to prevent social isolation and loneliness.

Accuracy of a wavelet-based fall detection approach using an accelerometer and a barometric pressure sensor.

GOAL: Falls are a major source of morbidity in older adults, and 50% of older adults who fall cannot rise independently after falling. Wearable sensor-based fall detection devices may assist in preventing long lies after falls. The goal of this study was to determine the accuracy of a novel wavelet-based approach to automatically detect falls based on accelerometer and barometric pressure sensor data. METHODS: Participants (n=15) mimicked a range of falls, near falls, and activities of daily living (ADLs) while wearing accelerometer and barometric pressure sensors on the lower back, chest, wrists and thighs. The wavelet transform using pattern adapted wavelets was applied to detect falls from the sensor data. RESULTS: In total, 525 trials (194 falls, 105 near-falls and 226 ADLs) were included in our analysis. When we applied the wavelet-based method on only accelerometer data, classification accuracies ranged from 82% to 96%, with the chest sensor providing the highest accuracy. Accuracy improved by 3.4% on average (p=0.041; SD=3.0%) when we also included the barometric pressure sensor data. The highest classification accuracies (of 98%) were achieved when we combined wavelet-based features and traditional statistical features in a multiphase fall detection model using machine learning. CONCLUSION: We show that the wavelet-based approach accurately distinguishes falls from near-falls and ADLs, and that it can be applied on wearable sensor data generated from various body locations. Additionally, we show that the accuracy of a wavelet-based fall detection system can be further improved by combining accelerometer and barometric pressure sensor data, and by incorporating wavelet and statistical features in a machine learning classification algorithm.

Association between Sedentary Behaviour and Physical, Cognitive, and Psychosocial Status among Older Adults in Assisted Living.

OBJECTIVE: Identification of the factors that influence sedentary behaviour in older adults is important for the design of appropriate intervention strategies. In this study, we determined the prevalence of sedentary behaviour and its association with physical, cognitive, and psychosocial status among older adults residing in Assisted Living (AL). METHODS: Participants (n = 114, mean age = 86.7) from AL sites in British Columbia wore waist-mounted activity monitors for 7 consecutive days, after being assessed with the Timed Up and Go (TUG), Montreal Cognitive Assessment (MoCA), Short Geriatric Depression Scale (GDS), and Modified Fall Efficacy Scale (MFES). RESULTS: On average, participants spent 87% of their waking hours in sedentary behaviour, which accumulated in 52 bouts per day with each bout lasting an average of 13 minutes. Increased sedentary behaviour associated significantly with scores on the TUG (r = 0.373, p < 0.001) and MFES (r = -0.261, p = 0.005), but not with the MoCA or GDS. Sedentary behaviour also associated with male gender, use of mobility aid, and multiple regression with increased age. CONCLUSION: We found that sedentary behaviour among older adults in AL associated with TUG scores and falls-related self-efficacy, which are modifiable targets for interventions to decrease sedentary behaviour in this population.

Comparison between clinical gait and daily-life gait assessments of fall risk in older people.

AIM: Falls are a leading cause of disability in older people. Here we investigate if daily-life gait assessments are better than clinical gait assessments at discriminating between older people with and without a history of falls. METHODS: A total of 96 independent-living participants (age 75.5 ± 7.8) underwent sensorimotor, psychological and cognitive assessments, and the Timed Up and Go and 10-m walk tests. Participants wore a small pendant sensor device for a week in their home environment, from which the new remote assessments of daily-life gait were determined. RESULTS: During daily-life, fallers had significantly lower gait quality (lower gait endurance, higher within-walk variability and lower between-walk adaptability), but not reduced gait quantity (total steps) or gait intensity (mean cadence). In the clinic, fallers had slower Timed Up and Go, but not 10-m walk test times. After adjusting for demographics, only the daily-life assessments of gait endurance and within-walk variability remained significant. Reduced daily-life gait assessments were significantly correlated with older age, higher body mass index, multiple medications, disability, more concern about falling, poor executive function and higher physiological fall risk. CONCLUSIONS: The new daily-life gait assessments were better than the clinical gait assessments at identifying fall risk in our sample of independent living older people. However, further research is required to validate these findings in other populations or those living in residential aged care. Daily-life gait was not only associated with demographics and physiological capacity, but also general health, executive function and the ability to undertake a variety of activities of daily living without excessive concern about falling. Geriatr Gerontol Int 2017; 17: 2274-2282.

Wavelet-Based Sit-To-Stand Detection and Assessment of Fall Risk in Older People Using a Wearable Pendant Device.

GOAL: Wearable devices provide new ways to identify people who are at risk of falls and track long-term changes of mobility in daily life of older people. The aim of this study was to develop a wavelet-based algorithm to detect and assess quality of sit-to-stand movements with a wearable pendant device. METHODS: The algorithm used wavelet transformations of the accelerometer and barometric air pressure sensor data. Detection accuracy was tested in 25 older people performing 30 min of typical daily activities. The ability to differentiate between people who are at risk of falls from people who are not at risk was investigated by assessing group differences of sensor-based sit-to-stand measurements in 34 fallers and 60 nonfallers (based on 12-month fall history) performing sit-to-stand movements as part of a laboratory study. RESULTS: Sit-to-stand movements were detected with 93.1% sensitivity and a false positive rate of 2.9% during activities of daily living. In the laboratory study, fallers had significantly lower maximum acceleration, velocity, and power during the sit-to-stand movement compared to nonfallers. CONCLUSION: The new wavelet-based algorithm accurately detected sit-to-stand movements in older people and differed significantly between older fallers and nonfallers. SIGNIFICANCE: Accurate detection and quantification of sit-to-stand movements may provide objective assessment and monitoring of fall risk during daily life in older people.

Kinect-based choice reaching and stepping reaction time tests for clinical and in-home assessment of fall risk in older people: a prospective study.

BACKGROUND: Quick protective reactions such as reaching or stepping are important to avoid a fall or minimize injuries. We developed Kinect-based choice reaching and stepping reaction time tests (Kinect-based CRTs) and evaluated their ability to differentiate between older fallers and non-fallers and the feasibility of administering them at home. METHODS: A total of 94 community-dwelling older people were assessed on the Kinect-based CRTs in the laboratory and were followed-up for falls for 6 months. Additionally, a subgroup (n = 20) conducted the Kinect-based CRTs at home. Signal processing algorithms were developed to extract features for reaction, movement and the total time from the Kinect skeleton data. RESULTS: Nineteen participants (20.2 %) reported a fall in the 6 months following the assessment. The reaction time (fallers: 797 ± 136 ms, non-fallers: 714 ± 89 ms), movement time (fallers: 392 ± 50 ms, non-fallers: 358 ± 51 ms) and total time (fallers: 1189 ± 170 ms, non-fallers: 1072 ± 109 ms) of the reaching reaction time test differentiated well between the fallers and non-fallers. The stepping reaction time test did not significantly discriminate between the two groups in the prospective study. The correlations between the laboratory and in-home assessments were 0.689 for the reaching reaction time and 0.860 for stepping reaction time. CONCLUSION: The study findings indicate that the Kinect-based CRT tests are feasible to administer in clinical and in-home settings, and thus represents an important step towards the development of sensor-based fall risk self-assessments. With further validation, the assessments may prove useful as a fall risk screen and home-based assessment measures for monitoring changes over time and effects of fall prevention interventions.

Kinect-Based Five-Times-Sit-to-Stand Test for Clinical and In-Home Assessment of Fall Risk in Older People.

BACKGROUND: Accidental falls remain an important problem in older people. The five-times-sit-to-stand (5STS) test is commonly used as a functional test to assess fall risk. Recent advances in sensor technologies hold great promise for more objective and accurate assessments. OBJECTIVE: The aims of this study were: (1) to examine the feasibility of a low-cost and portable Kinect-based 5STS test to discriminate between fallers and nonfallers and (2) to investigate whether this test can be used for supervised clinical, supervised and unsupervised in-home fall risk assessments. METHODS: A total of 94 community-dwelling older adults were assessed by the Kinect-based 5STS test in the laboratory and 20 participants were tested in their own homes. An algorithm was developed to automatically calculate timing- and speed-related measurements from the Kinect-based sensor data to discriminate between fallers and nonfallers. The associations of these measurements with standard clinical fall risk tests and the results of supervised and unsupervised in-home assessments were examined. RESULTS: Fallers were significantly slower than nonfallers on Kinect-based measures. The mean velocity of the sit-to-stand transitions discriminated well between the fallers and nonfallers based on 12-month retrospective fall data. The Kinect-based measures collected in the laboratory correlated strongly with those collected in the supervised (r = 0.704-0.832) and unsupervised (r = 0.775-0.931) in-home assessments. CONCLUSION: In summary, we found that the Kinect-based 5STS test discriminated well between the fallers and nonfallers and was feasible to administer in clinical and supervised in-home settings. This test may be useful in clinical settings for identifying high-risk fallers for further intervention or for regular in-home assessments in the future.

ICT-based system to predict and prevent falls (iStoppFalls): results from an international multicenter randomized controlled trial.

BACKGROUND: Falls and fall-related injuries are a serious public health issue. Exercise programs can effectively reduce fall risk in older people. The iStoppFalls project developed an Information and Communication Technology-based system to deliver an unsupervised exercise program in older people's homes. The primary aims of the iStoppFalls randomized controlled trial were to assess the feasibility (exercise adherence, acceptability and safety) of the intervention program and its effectiveness on common fall risk factors. METHODS: A total of 153 community-dwelling people aged 65+ years took part in this international, multicentre, randomized controlled trial. Intervention group participants conducted the exercise program for 16 weeks, with a recommended duration of 120 min/week for balance exergames and 60 min/week for strength exercises. All intervention and control participants received educational material including advice on a healthy lifestyle and fall prevention. Assessments included physical and cognitive tests, and questionnaires for health, fear of falling, number of falls, quality of life and psychosocial outcomes. RESULTS: The median total exercise duration was 11.7 h (IQR = 22.0) over the 16-week intervention period. There were no adverse events. Physiological fall risk (Physiological Profile Assessment, PPA) reduced significantly more in the intervention group compared to the control group (F1,127 = 4.54, p = 0.035). There was a significant three-way interaction for fall risk assessed by the PPA between the high-adherence (>90 min/week; n = 18, 25.4 %), low-adherence (<90 min/week; n = 53, 74.6 %) and control group (F2,125 = 3.12, n = 75, p = 0.044). Post hoc analysis revealed a significantly larger effect in favour of the high-adherence group compared to the control group for fall risk (p = 0.031), postural sway (p = 0.046), stepping reaction time (p = 0.041), executive functioning (p = 0.044), and quality of life (p for trend = 0.052). CONCLUSIONS: The iStoppFalls exercise program reduced physiological fall risk in the study sample. Additional subgroup analyses revealed that intervention participants with better adherence also improved in postural sway, stepping reaction, and executive function. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry Trial ID: ACTRN12614000096651 International Standard Randomised Controlled Trial Number: ISRCTN15932647.

The effect of sensor-based exercise at home on functional performance associated with fall risk in older people - a comparison of two exergame interventions.

BACKGROUND: There is good evidence that balance challenging exercises can reduce falls in older people. However, older people often find it difficult to incorporate such programs in their daily life. Videogame technology has been proposed to promote enjoyable, balance-challenging exercise. As part of a larger analysis, we compared feasibility and efficacy of two exergame interventions: step-mat-training (SMT) and Microsoft-Kinect® (KIN) exergames. METHODS: 148 community-dwelling people, aged 65+ years participated in two exergame studies in Sydney, Australia (KIN: n = 57, SMT: n = 91). Both interventions were delivered as unsupervised exercise programs in participants' homes for 16 weeks. Assessment measures included overall physiological fall risk, muscle strength, finger-press reaction time, proprioception, vision, balance and executive functioning. RESULTS: For participants allocated to the intervention arms, the median time played each week was 17 min (IQR 32) for KIN and 48 min (IQR 94) for SMT. Compared to the control group, SMT participants improved their fall risk score (p = 0.036), proprioception (p = 0.015), reaction time (p = 0.003), sit-to-stand performance (p = 0.011) and executive functioning (p = 0.001), while KIN participants improved their muscle strength (p = 0.032) and vision (p = 0.010), and showed a trend towards improved fall risk scores (p = 0.057). CONCLUSIONS: The findings suggest that it is feasible for older people to conduct an unsupervised exercise program at home using exergames. Both interventions reduced fall risk and SMT additionally improved specific cognitive functions. However, further refinement of the systems is required to improve adherence and maximise the benefits of exergames to deliver fall prevention programs in older people's homes. TRIAL REGISTRATIONS: ACTRN12613000671763 (Step Mat Training RCT) ACTRN12614000096651 (MS Kinect RCT).

The design of a purpose-built exergame for fall prediction and prevention for older people.

BACKGROUND: Falls in older people represent a major age-related health challenge facing our society. Novel methods for delivery of falls prevention programs are required to increase effectiveness and adherence to these programs while containing costs. The primary aim of the Information and Communications Technology-based System to Predict and Prevent Falls (iStoppFalls) project was to develop innovative home-based technologies for continuous monitoring and exercise-based prevention of falls in community-dwelling older people. The aim of this paper is to describe the components of the iStoppFalls system. METHODS: The system comprised of 1) a TV, 2) a PC, 3) the Microsoft Kinect, 4) a wearable sensor and 5) an assessment and training software as the main components. RESULTS: The iStoppFalls system implements existing technologies to deliver a tailored home-based exercise and education program aimed at reducing fall risk in older people. A risk assessment tool was designed to identify fall risk factors. The content and progression rules of the iStoppFalls exergames were developed from evidence-based fall prevention interventions targeting muscle strength and balance in older people. CONCLUSIONS: The iStoppFalls fall prevention program, used in conjunction with the multifactorial fall risk assessment tool, aims to provide a comprehensive and individualised, yet novel fall risk assessment and prevention program that is feasible for widespread use to prevent falls and fall-related injuries. This work provides a new approach to engage older people in home-based exercise programs to complement or provide a potentially motivational alternative to traditional exercise to reduce the risk of falling.

ICT-based system to predict and prevent falls (iStoppFalls): study protocol for an international multicenter randomized controlled trial.

BACKGROUND: Falls are very common, especially in adults aged 65 years and older. Within the current international European Commission's Seventh Framework Program (FP7) project 'iStoppFalls' an Information and Communication Technology (ICT) based system has been developed to regularly assess a person's risk of falling in their own home and to deliver an individual and tailored home-based exercise and education program for fall prevention. The primary aims of iStoppFalls are to assess the feasibility and acceptability of the intervention program, and its effectiveness to improve balance, muscle strength and quality of life in older people. METHODS/DESIGN: This international, multicenter study is designed as a single-blinded, two-group randomized controlled trial. A total of 160 community-dwelling older people aged 65 years and older will be recruited in Germany (n = 60), Spain (n = 40), and Australia (n = 60) between November 2013 and May 2014. Participants in the intervention group will conduct a 16-week exercise program using the iStoppFalls system through their television set at home. Participants are encouraged to exercise for a total duration of 180 minutes per week. The training program consists of a variety of balance and strength exercises in the form of video games using exergame technology. Educational material about a healthy lifestyle will be provided to each participant. Final reassessments will be conducted after 16 weeks. The assessments include physical and cognitive tests as well as questionnaires assessing health, fear of falling, quality of life and psychosocial determinants. Falls will be followed up for six months by monthly falls calendars. DISCUSSION: We hypothesize that the regular use of this newly developed ICT-based system for fall prevention at home is feasible for older people. By using the iStoppFalls sensor-based exercise program, older people are expected to improve in balance and strength outcomes. In addition, the exercise training may have a positive impact on quality of life by reducing the risk of falls. Taken together with expected cognitive improvements, the individual approach of the iStoppFalls program may provide an effective model for fall prevention in older people who prefer to exercise at home. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry Trial ID: ACTRN12614000096651.International Standard Randomised Controlled Trial Number: ISRCTN15932647.

New methods for fall risk prediction.

PURPOSE OF REVIEW: Accidental falls are the leading cause of injury-related death and hospitalization in old age, with over one-third of the older adults experiencing at least one fall or more each year. Because of limited healthcare resources, regular objective fall risk assessments are not possible in the community on a large scale. New methods for fall prediction are necessary to identify and monitor those older people at high risk of falling who would benefit from participating in falls prevention programmes. RECENT FINDINGS: Technological advances have enabled less expensive ways to quantify physical fall risk in clinical practice and in the homes of older people. Recently, several studies have demonstrated that sensor-based fall risk assessments of postural sway, functional mobility, stepping and walking can discriminate between fallers and nonfallers. SUMMARY: Recent research has used low-cost, portable and objective measuring instruments to assess fall risk in older people. Future use of these technologies holds promise for assessing fall risk accurately in an unobtrusive manner in clinical and daily life settings.

Choice stepping reaction time test using exergame technology for fall risk assessment in older people.

Accidental falls remain an important problem in older people. Stepping is a common task to avoid a fall and requires good interplay between sensory functions, central processing and motor execution. Increased choice stepping reaction time has been associated with recurrent falls in older people. The aim of this study was to examine if a sensor-based Exergame Choice Stepping Reaction Time test can successfully discriminate older fallers from non-fallers. The stepping test was conducted in a cohort of 104 community-dwelling older people (mean age: 80.7 ± 7.0 years). Participants were asked to step laterally as quickly as possible after a light stimulus appeared on a TV screen. Spatial and temporal measurements of the lower and upper body were derived from a low-cost and portable 3D-depth sensor (i.e. Microsoft Kinect) and 3D-accelerometer. Fallers had a slower stepping reaction time (970 ± 228 ms vs. 858 ± 123 ms, P = 0.001) and a slower reaction of their upper body (719 ± 289 ms vs. 631 ± 166 ms, P = 0.052) compared to non-fallers. It took fallers significantly longer than non-fallers to recover their balance after initiating the step (2147 ± 800 ms vs. 1841 ± 591 ms, P = 0.029). This study demonstrated that a sensor-based, low-cost and easy to administer stepping test, with the potential to be used in clinical practice or regular unsupervised home assessments, was able to identify significant differences between performances by fallers and non-fallers.