Validation of 24-hour ambulatory gait assessment in Parkinson's disease with simultaneous video observation

Does Impaired Plantar Cutaneous Vibration Perception Contribute to Axial Motor Symptoms in Parkinson's Disease? Effects of Medication and Subthalamic Nucleus Deep Brain Stimulation

OBJECTIVE

To investigate whether impaired plantar cutaneous vibration perception contributes to axial motor symptoms in Parkinson's disease (PD) and whether anti-parkinsonian medication and subthalamic nucleus deep brain stimulation (STN-DBS) show different effects.

METHODS

Three groups were evaluated: PD patients in the medication "on" state (PD-MED), PD patients in the medication "on" state and additionally "on" STN-DBS (PD-MED-DBS), as well as healthy subjects (HS) as reference. Motor performance was analyzed using a pressure distribution platform. Plantar cutaneous vibration perception thresholds (VPT) were investigated using a customized vibration exciter at 30 Hz.

RESULTS

Motor performance of PD-MED and PD-MED-DBS was characterized by greater postural sway, smaller limits of stability ranges, and slower gait due to shorter strides, fewer steps per minute, and broader stride widths compared to HS. Comparing patient groups, PD-MED-DBS showed better overall motor performance than PD-MED, particularly for the functional limits of stability and gait. VPTs were significantly higher for PD-MED compared to those of HS, which suggests impaired plantar cutaneous vibration perception in PD. However, PD-MED-DBS showed less impaired cutaneous vibration perception than PD-MED.

CONCLUSIONS

PD patients suffer from poor motor performance compared to healthy subjects. Anti-parkinsonian medication in tandem with STN-DBS seems to be superior for normalizing axial motor symptoms compared to medication alone. Plantar cutaneous vibration perception is impaired in PD patients, whereas anti-parkinsonian medication together with STN-DBS is superior for normalizing tactile cutaneous perception compared to medication alone. Consequently, based on our results and the findings of the literature, impaired plantar cutaneous vibration perception might contribute to axial motor symptoms in PD.

Probabilistic Modelling of Gait for Robust Passive Monitoring in Daily Life

Passive monitoring in daily life may provide valuable insights into a person's health throughout the day. Wearable sensor devices play a key role in enabling such monitoring in a non-obtrusive fashion. However, sensor data collected in daily life reflect multiple health and behavior-related factors together. This creates the need for a structured principled analysis to produce reliable and interpretable predictions that can be used to support clinical diagnosis and treatment. In this work we develop a principled modelling approach for free-living gait (walking) analysis. Gait is a promising target for non-obtrusive monitoring because it is common and indicative of many different movement disorders such as Parkinson's disease (PD), yet its analysis has largely been limited to experimentally controlled lab settings. To locate and characterize stationary gait segments in free-living using accelerometers, we present an unsupervised probabilistic framework designed to segment signals into differing gait and non-gait patterns. We evaluate the approach using a new video-referenced dataset including 25 PD patients with motor fluctuations and 25 age-matched controls, performing unscripted daily living activities in and around their own houses. Using this dataset, we demonstrate the framework's ability to detect gait and predict medication induced fluctuations in PD patients based on free-living gait. We show that our approach is robust to varying sensor locations, including the wrist, ankle, trouser pocket and lower back.

Continuous home monitoring of Parkinson's disease using inertial sensors: A systematic review

Parkinson’s disease (PD) is a progressive neurological disorder of the central nervous system that deteriorates motor functions, while it is also accompanied by a large diversity of non-motor symptoms such as cognitive impairment and mood changes, hallucinations, and sleep disturbance. Parkinsonism is evaluated during clinical examinations and appropriate medical treatments are directed towards alleviating symptoms. Tri-axial accelerometers, gyroscopes, and magnetometers could be adopted to support clinicians in the decision-making process by objectively quantifying the patient’s condition. In this context, at-home data collections aim to capture motor function during daily living and unobstructedly assess the patients’ status and the disease’s symptoms for prolonged time periods. This review aims to collate existing literature on PD monitoring using inertial sensors while it focuses on papers with at least one free-living data capture unsupervised either directly or via videotapes. Twenty-four papers were selected at the end of the process: fourteen investigated gait impairments, eight of which focused on walking, three on turning, two on falls, and one on physical activity; ten articles on the other hand examined symptoms, including bradykinesia, tremor, dyskinesia, and motor state fluctuations in the on/off phenomenon. In summary, inertial sensors are capable of gathering data over a long period of time and have the potential to facilitate the monitoring of people with Parkinson’s, providing relevant information about their motor status. Concerning gait impairments, kinematic parameters (such as duration of gait cycle, step length, and velocity) were typically used to discern PD from healthy subjects, whereas for symptoms’ assessment, researchers were capable of achieving accuracies of over 90% in a free-living environment. Further investigations should be focused on the development of ad-hoc hardware and software capable of providing real-time feedback to clinicians and patients. In addition, features such as the wearability of the system and user comfort, set-up process, and instructions for use, need to be strongly considered in the development of wearable sensors for PD monitoring.

Evaluation of Wearable Sensor Devices in Parkinson's Disease: A Review of Current Status and Future Prospects

Parkinson's disease (PD) decreases the quality of life of the affected individuals. The incidence of PD is expected to increase given the growing aging population. Motor symptoms associated with PD render the patients unable to self-care and function properly. Given that several drugs have been developed to control motor symptoms, highly sensitive scales for clinical evaluation of drug efficacy are needed. Among such scales, the objective and continuous evaluation of wearable devices is increasingly utilized by clinicians and patients. Several electronic technologies have revolutionized the clinical monitoring of PD development, especially its motor symptoms. Here, we review and discuss the recent advances in the development of wearable devices for bradykinesia, tremor, gait, and myotonia. Our aim is to capture the experiences of patients and clinicians, as well as expand our understanding on the application of wearable technology. In so-doing, we lay the foundation for further research into the use of wearable technology in the management of PD.

Lateralized Connectivity between Globus Pallidus and Motor Cortex is Associated with Freezing of Gait in Parkinson's Disease

Freezing of gait (FoG) is a brief, episodic absence or marked reduction of forward progression of the feet, despite the intention to walk, that is common in people with Parkinson's disease (PD). We hypothesized that not only motor, but higher level cognitive and attention areas may be impaired in freezers. In this study, we aimed to characterize differences in cortical and subcortical functional connectivity specific to FoG. We examined resting state neuroimaging and objective measures of FoG severity and gait from 103 individuals (28 PD + FoG, 36 PD - FoG and 39 healthy controls). Inertial sensors were used to quantify freezing severity and gait. Groups with and without FoG were matched on age, disease severity, cognitive status, and levodopa medication. MRI data was processed using surface-based registration. High-quality imaging data were used to characterize differences in connectivity specific to FoG using a pre-defined set of Regions of Interest (ROIs) and validated using whole-brain connectivity analysis. Associations between functional connectivity and objective measures of FoG were determined via predictive modeling using hold-out cross validation. We found that connectivity between the left globus pallidus (GP) and left somatosensory cortex and between two brain areas in the default and insular/vestibular networks exhibited significant differences specific to FoG and were also strong and significant predictors of FoG severity. Our findings suggest that the interplay among motor, default and vestibular areas of the left cortex are critical in the pathology of FoG.

Clinical and methodological challenges for assessing freezing of gait: Future perspectives

Freezing of gait, defined as sudden and usually brief episodes of inability to produce effective stepping, often results in falls and is both disabling and common in parkinsonism. In this narrative review, sprung from the 2nd International Workshop on freezing of gait in Leuven, we summarize the latest insights into clinical and methodological challenges for assessing freezing of gait. We also highlight the role of emerging wearable technology to improve the management of this debilitating symptom. © 2019 International Parkinson and Movement Disorder Society.

Quantification of discrete behavioral components of the MDS-UPDRS

INTRODUCTION

The Movement Disorder Society's Unified Parkinson's Disease Rating Scale (MDS-UPDRS) is the current gold standard means of assessing disease state in Parkinson's disease (PD). Objective measures in the form of wearable sensors have the potential to improve our ability to monitor symptomology in PD, but numerous methodological challenges remain, including integration into the MDS-UPDRS. We applied a structured video coding scheme to temporally quantify clinical, scripted, motor tasks in the MDS-UPDRS for the alignment and integration of objective measures collected in parallel.

METHODS

25 PD subjects completed two video-recorded MDS-UPDRS administrations. Visual cues of task performance reliably identifiable in video recordings were used to construct a structured video coding scheme. Postural transitions were also defined and coded. Videos were independently coded by two trained non-expert coders and a third expert coder to derive indices of inter-rater agreement.

RESULTS

50 videos of MDS-UPDRS performance were fully coded. Non-expert coders achieved a high level of agreement (Cohen's κ > 0.8) on all postural transitions and scripted motor tasks except for Postural Stability (κ = 0.617); this level of agreement was largely maintained even when more stringent thresholds for agreement were applied. Durations coded by non-expert coders and expert coders were significantly different (p < 0.05) for only Postural Stability and Rigidity, Left Upper Limb.

CONCLUSIONS

Non-expert coders consistently and accurately quantified discrete behavioral components of the MDS-UPDRS using a structured video coding scheme; this represents a novel, promising approach for integrating objective and clinical measures into unified, longitudinal datasets.

Impact of motor fluctuations on real-life gait in Parkinson's patients

BACKGROUND

People with PD (PWP) have an increased risk of becoming inactive. Wearable sensors can provide insights into daily physical activity and walking patterns.

RESEARCH QUESTIONS

(1) Is the severity of motor fluctuations associated with sensor-derived average daily walking quantity? (2) Is the severity of motor fluctuations associated with the amount of change in sensor-derived walking quantity after levodopa intake?

METHODS

304 Dutch PWP from the Parkinson@Home study were included. At baseline, all participants received a clinical examination. During the follow-up period (median: 97 days; 25-Interquartile range-IQR: 91 days, 75-IQR: 188 days), participants used the Fox Wearable Companion app and streamed smartwatch accelerometer data to a cloud platform. The first research question was assessed by linear regression on the sensor-derived mean time spent walking/day with the severity of fluctuations (MDS-UPDRS item 4.4) as independent variable, controlled for age and MDS-UPDRS part-III score. The second research question was assessed by linear regression on the sensor-derived mean post-levodopa walking quantity, with the sensor-derived mean pre-levodopa walking quantity and severity of fluctuations as independent variables, controlled for mean time spent walking per day, age and MDS-UPDRS part-III score.

RESULTS

PWP spent most time walking between 8am and 1pm, summing up to 72 ± 39 (mean ± standard deviation) minutes of walking/day. The severity of motor fluctuations did not influence the mean time spent walking (B = 2.4 ± 1.9, p = 0.20), but higher age (B = -1.3 ± 0.3, p = < 0.001) and greater severity of motor symptoms (B = -0.6 ± 0.2, p < 0.001) was associated with less time spent walking (F(3216) = 14.6, p < .001, R² = .17). The severity of fluctuations was not associated with the amount of change in time spent walking in relation to levodopa intake in any part of the day.

SIGNIFICANCE

Analysis of sensor-derived gait quantity suggests that the severity of motor fluctuations is not associated with changes in real-life walking patterns in mildly to moderate affected PWP.

External validation of a collar-mounted triaxial accelerometer for second-by-second monitoring of eight behavioural states in dogs

Early detection of disease by an animal owner may motivate them to seek early veterinary advice. Presentation before a more advanced clinical manifestation is evident could lead to more effective treatment and thus benefit the animal’s health and welfare. Accelerometers are able to detect changes in specific activities or behaviours, thus indicating early signs of possible adverse health events. The objective of this validation study was to determine whether the detection of eight behavioural states: walk, trot, canter/gallop, sleep, static/inactive, eat, drink, and headshake, by an accelerometer device was sufficiently accurate to be useful in a clinical setting. This fully independent external validation estimated the accuracy of a specific triaxial, collar-mounted accelerometer on a second-by second basis in 51 healthy dogs of different breeds, aged between 6 months and 13 years, weighing >10 kg. The overall diagnostic effectiveness was estimated as: % record correctly classified of > 95% in walk, trot, canter/gallop, eat, drink and headshake and >90% in sleep and static/inactive. The positive predictive values ranged from 93–100%, while the negative predictive values ranged from 96–100%, with exception of static/inactive (86%).This was probably because dogs were placed in unfamiliar kennels where they did not exhibit their typical resting behaviour. The device is worn on a collar, making its use feasible for anyone wanting to monitor their dog’s behaviour. The high accuracy in detecting various kinds of behaviour appears promising in assessing canine health and welfare states.

Evaluating physical function and activity in the elderly patient using wearable motion sensors

Wearable sensors, in particular inertial measurement units (IMUs) allow the objective, valid, discriminative and responsive assessment of physical function during functional tests such as gait, stair climbing or sit-to-stand.

Applied to various body segments, precise capture of time-to-task achievement, spatiotemporal gait and kinematic parameters of demanding tests or specific to an affected limb are the most used measures.

In activity monitoring (AM), accelerometry has mainly been used to derive energy expenditure or general health related parameters such as total step counts.

In orthopaedics and the elderly, counting specific events such as stairs or high intensity activities were clinimetrically most powerful; as were qualitative parameters at the ‘micro-level’ of activity such as step frequency or sit-stand duration.

Low cost and ease of use allow routine clinical application but with many options for sensors, algorithms, test and parameter definitions, choice and comparability remain difficult, calling for consensus or standardisation.

Cite this article: Grimm B, Bolink S. Evaluating physical function and activity in the elderly patient using wearable motion sensors. EFORT Open Rev 2016;1:112–120. DOI: 10.1302/2058-5241.1.160022.

The clinical significance of freezing while turning in Parkinson's disease

Freezing of gait (FoG) in people with Parkinson's disease (PD) is an environmentally sensitive, intermittent problem that occurs most often during turning. FoG is difficult for clinicians to evaluate and treat because it can be difficult to elicit during a clinical visit. Here, we aimed to develop a clinically valid objective measure of freezing severity during a 2-min 360-degree turning-in-place. Twenty-eight subjects with PD (16 freezers, FoG+, and 12 nonfreezers, FoG-) in the "off" state and 14 healthy control subjects were tested. Subjects wore three inertial sensors (one on each shin and one on the waist) while (1) turning in place for 2 min (alternating 360 degrees to the right with 360 degrees to the left) and (2) performing an Instrumented 7-m Timed Up and Go test (ITUG). Performance was videotaped, and clinical severity of FoG was independently rated by two movement disorders specialists (co-authors). Turning in place consistently resulted in FoG (13 out of 16 subjects with PD) while FoG was clinically observed in only two subjects with PD during the ITUG test. The Freezing Ratio during the turning test was significantly correlated with the clinical ratings (ρ=0.7, p=0.003) and with score on the new FoG questionnaire (ρ=0.5, p=0.03). After correcting for symptom severity (UPDRS-III), out of the four objective measures of the turning test (total number of turns, average turn peak speed and average turn smoothness), only the Freezing Ratio was significantly different across groups (p=0.04). Freezing can be well quantified with body-worn inertial sensors during a 2-min turning-in-place protocol.

Ambulatory Gait Behavior in Patients With Dementia: A Comparison With Parkinson's Disease

Accelerometry-based gait analysis is a promising approach in obtaining insightful information on the gait characteristics of patients with neurological disorders such as dementia and Parkinson's disease (PD). In order to improve its practical use outside the laboratory or hospital, it is required to design new metrics capable of quantifying ambulatory gait and their extraction procedures from long-term acceleration data. This paper presents a gait analysis method developed for such a purpose. Our system is based on a single trunk-mounted accelerometer and analytical algorithm for the assessment of gait behavior that may be context dependent. The algorithm consists of the detection of gait peaks from acceleration data and the analysis of multimodal patterns in the relationship between gait cycle and vertical gait acceleration. A set of six new measures can be obtained by applying the algorithm to a 24-h motion signal. To examine the performance and utility of our method, we recorded acceleration data from 13 healthy, 26 PD, and 26 mild cognitive impairment or dementia subjects. Each patient group was further classified into two, comprising 13 members each, according to the severity of the disease, and the gait behavior of the five groups was compared. We found that the normal, PD, and MCI/dementia groups show characteristic walking patterns which can be distinguished from one another by the developed gait measure set. We also examined conventional parameters such as gait acceleration, gait cycle, and gait variability, but failed to reproduce the distinct differences among the five groups. These findings suggest that the proposed gait analysis may be useful in capturing disease-specific gait features in a community setting.

Remote Physical Activity Monitoring in Neurological Disease: A Systematic Review

OBJECTIVE

To perform a systematic review of studies using remote physical activity monitoring in neurological diseases, highlighting advances and determining gaps.

METHODS

Studies were systematically identified in PubMed/MEDLINE, CINAHL and SCOPUS from January 2004 to December 2014 that monitored physical activity for ≥24 hours in adults with neurological diseases. Studies that measured only involuntary motor activity (tremor, seizures), energy expenditure or sleep were excluded. Feasibility, findings, and protocols were examined.

RESULTS

137 studies met inclusion criteria in multiple sclerosis (MS) (61 studies); stroke (41); Parkinson's Disease (PD) (20); dementia (11); traumatic brain injury (2) and ataxia (1). Physical activity levels measured by remote monitoring are consistently low in people with MS, stroke and dementia, and patterns of physical activity are altered in PD. In MS, decreased ambulatory activity assessed via remote monitoring is associated with greater disability and lower quality of life. In stroke, remote measures of upper limb function and ambulation are associated with functional recovery following rehabilitation and goal-directed interventions. In PD, remote monitoring may help to predict falls. In dementia, remote physical activity measures correlate with disease severity and can detect wandering.

CONCLUSIONS

These studies show that remote physical activity monitoring is feasible in neurological diseases, including in people with moderate to severe neurological disability. Remote monitoring can be a psychometrically sound and responsive way to assess physical activity in neurological disease. Further research is needed to ensure these tools provide meaningful information in the context of specific neurological disorders and patterns of neurological disability.

Contribution of Step Length to Increase Walking and Turning Speed as a Marker of Parkinson's Disease Progression

When increasing ambulation speed in Parkinson’s disease, step cadence increases more than stride length, indicating movement scaling difficulties that affect step generation in particular. We investigated whether step length variation when increasing ambulation speed was related to disease progression. Patients with Parkinson’s disease (N = 39) and controls (N = 152) performed two timed ambulation tasks: at a 'free' (self-selected) pace and then at 'maximal' speed. The total number of steps (including during turns) and time to complete the task were clinically measured. The relative contribution of step length and cadence to increased ambulation speed was determined using two methods: the ratios of change in step length or in cadence to the change in ambulation speed, and the step length index. While the relative contribution of step length and cadence to increased ambulation speed was independent of age in both control and patient groups, in Parkinson’s disease there was a negative correlation between time from diagnosis and the ratio of change in step length to change in ambulation speed (R = 0.54; p = 0.0004) and the step length index (R = 0.56, p = 0.0002). In parallel, there was a positive correlation between time since diagnosis and the ratio of change in cadence to change in ambulation speed (R = 0.57; p = 0.0002). The relative contribution of step length and cadence to increased ambulation speed is age invariant but a marker of Parkinson's disease advancement, and can be easily determined in the clinical setting.

Motion tracking and gait feature estimation for recognising Parkinson's disease using MS Kinect

Background

Analysis of gait features provides important information during the treatment of neurological disorders, including Parkinson’s disease. It is also used to observe the effects of medication and rehabilitation. The methodology presented in this paper enables the detection of selected gait attributes by Microsoft (MS) Kinect image and depth sensors to track movements in three-dimensional space.

Methods

The experimental part of the paper is devoted to the study of three sets of individuals: 18 patients with Parkinson’s disease, 18 healthy aged-matched individuals, and 15 students. The methodological part of the paper includes the use of digital signal-processing methods for rejecting gross data-acquisition errors, segmenting video frames, and extracting gait features. The proposed algorithm describes methods for estimating the leg length, normalised average stride length (SL), and gait velocity (GV) of the individuals in the given sets using MS Kinect data.

Results

The main objective of this work involves the recognition of selected gait disorders in both the clinical and everyday settings. The results obtained include an evaluation of leg lengths, with a mean difference of 0.004 m in the complete set of 51 individuals studied, and of the gait features of patients with Parkinson’s disease (SL: 0.38 m, GV: 0.61 m/s) and an age-matched reference set (SL: 0.54 m, GV: 0.81 m/s). Combining both features allowed for the use of neural networks to classify and evaluate the selectivity, specificity, and accuracy. The achieved accuracy was 97.2 %, which suggests the potential use of MS Kinect image and depth sensors for these applications.

Conclusions

Discussion points include the possibility of using the MS Kinect sensors as inexpensive replacements for complex multi-camera systems and treadmill walking in gait-feature detection for the recognition of selected gait disorders.

Multidirectional walk test in individuals with Parkinson's disease: a validity study

Gait parameters of forward, backward, and sideways walk were studied when the participants walked overground in four directions at their self-selected speed and were compared with walking in the four directions on an instrumented GAITRite walkway. Intraclass correlation coefficients between the overground walk test measures and the instrumented walkway measures of gait speed, cadence, and stride length for the forward walk were 0.85, 0.88, and 0.87, respectively. For the backward walk, the coefficients were 0.91 for gait speed, 0.75 for cadence, and 0.93 for stride length. For the sideways walk, the coefficients were 0.92 for gait speed, 0.93 for cadence, and 0.94 for stride length. Gait parameters of forward, backward, and sideways walk obtained by the overground walk test had excellent agreement with those obtained by the instrumented walkway. The quick timed test provided quantitative data for gait evaluation and was valid for clinical use.

Objective assessment of fall risk in Parkinson's disease using a body-fixed sensor worn for 3 days

BACKGROUND

Patients with Parkinson's disease (PD) suffer from a high fall risk. Previous approaches for evaluating fall risk are based on self-report or testing at a given time point and may, therefore, be insufficient to optimally capture fall risk. We tested, for the first time, whether metrics derived from 3 day continuous recordings are associated with fall risk in PD.

METHODS AND MATERIALS

107 patients (Hoehn & Yahr Stage: 2.6±0.7) wore a small, body-fixed sensor (3D accelerometer) on lower back for 3 days. Walking quantity (e.g., steps per 3-days) and quality (e.g., frequency-derived measures of gait variability) were determined. Subjects were classified as fallers or non-fallers based on fall history. Subjects were also followed for one year to evaluate predictors of the transition from non-faller to faller.

RESULTS

The 3 day acceleration derived measures were significantly different in fallers and non-fallers and were significantly correlated with previously validated measures of fall risk. Walking quantity was similar in the two groups. In contrast, the fallers walked with higher step-to-step variability, e.g., anterior-posterior width of the dominant frequency was larger (p = 0.012) in the fallers (0.78 ± 0.17 Hz) compared to the non-fallers (0.71 ± 0.07 Hz). Among subjects who reported no falls in the year prior to testing, sensor-derived measures predicted the time to first fall (p = 0.0034), whereas many traditional measures did not. Cox regression analysis showed that anterior-posterior width was significantly (p = 0.0039) associated with time to fall during the follow-up period, even after adjusting for traditional measures.

CONCLUSIONS/SIGNIFICANCE

These findings indicate that a body-fixed sensor worn continuously can evaluate fall risk in PD. This sensor-based approach was able to identify transition from non-faller to faller, whereas many traditional metrics were not successful. This approach may facilitate earlier detection of fall risk and may in the future, help reduce high costs associated with falls.

Continuous monitoring of turning in patients with movement disability

Difficulty with turning is a major contributor to mobility disability and falls in people with movement disorders, such as Parkinson's disease (PD). Turning often results in freezing and/or falling in patients with PD. However, asking a patient to execute a turn in the clinic often does not reveal their impairments. Continuous monitoring of turning with wearable sensors during spontaneous daily activities may help clinicians and patients determine who is at risk of falls and could benefit from preventative interventions. In this study, we show that continuous monitoring of natural turning with wearable sensors during daily activities inside and outside the home is feasible for people with PD and elderly people. We developed an algorithm to detect and characterize turns during gait, using wearable inertial sensors. First, we validate the turning algorithm in the laboratory against a Motion Analysis system and against a video analysis of 21 PD patients and 19 control (CT) subjects wearing an inertial sensor on the pelvis. Compared to Motion Analysis and video, the algorithm maintained a sensitivity of 0.90 and 0.76 and a specificity of 0.75 and 0.65, respectively. Second, we apply the turning algorithm to data collected in the home from 12 PD and 18 CT subjects. The algorithm successfully detects turn characteristics, and the results show that, compared to controls, PD subjects tend to take shorter turns with smaller turn angles and more steps. Furthermore, PD subjects show more variability in all turn metrics throughout the day and the week.

Quantitative wearable sensors for objective assessment of Parkinson's disease

There is a rapidly growing interest in the quantitative assessment of Parkinson's disease (PD)-associated signs and disability using wearable technology. Both persons with PD and their clinicians see advantages in such developments. Specifically, quantitative assessments using wearable technology may allow for continuous, unobtrusive, objective, and ecologically valid data collection. Also, this approach may improve patient-doctor interaction, influence therapeutic decisions, and ultimately ameliorate patients' global health status. In addition, such measures have the potential to be used as outcome parameters in clinical trials, allowing for frequent assessments; eg, in the home setting. This review discusses promising wearable technology, addresses which parameters should be prioritized in such assessment strategies, and reports about studies that have already investigated daily life issues in PD using this new technology.

Accelerometry-based gait analysis and its application to Parkinson's disease assessment- part 2: a new measure for quantifying walking behavior

Gait analysis is a valuable tool for obtaining quantitative information on motor deficits in Parkinson's disease (PD). Since the characteristic gait patterns of PD patients may not be fully identified by brief examination in a clinic, long-term, and unobtrusive monitoring of their activities is essential, especially in a nonclinical setting. This paper describes a single accelerometer-based gait analysis system for the assessment of ambulatory gait properties. Acceleration data were recorded continuously for up to 24 h from normal and PD subjects, from which gait peaks were picked out and the relationship between gait cycle and vertical gait acceleration was evaluated. By fitting a model equation to the relationships, a quantitative index was obtained for characterizing the subjects' walking behavior. The averaged index for PD patients with gait disorder was statistically smaller than the value for normal subjects. The proposed method could be used to evaluate daily gait characteristics and thus contribute to a more refined diagnosis and treatment of the disease.

Clinical assessment of freezing of gait in Parkinson's disease from computer-generated animation

The current 'gold standard' for clinical evaluation of freezing of gait (FOG) in Parkinson's disease (PD) is determination of the number of FOG episodes from video by independent raters. We have previously described a robust technique for objective FOG assessment from lower-limb acceleration. However, there is no existing method for validation of autonomous FOG measures in the absence of video documentation. In this study we compared the results of clinical evaluation of FOG from computer-generated animations (derived from body-mounted inertial sensors) during a timed up and go test with the 'gold standard' of clinical video assessment, utilizing a cohort of 10 experienced raters from four PD centers. Agreement between the 10 clinical observers for scoring of FOG from computer animations was more robust for the relative duration of freeze events (percent time frozen; intraclass correlation coefficient of 0.65) than number of FOG episodes, and was comparable with clinical evaluation of the patient from video (intraclass correlation coefficient 0.73). This result suggests that percent time frozen should be considered (along with number of FOG events) to better convey FOG severity. The ability of clinical observers to quantify FOG from computer-generated animation derived from lower-limb motion data provides a potential approach to validation of accelerometry-based FOG identification outside of the clinic.

Unsupervised machine-learning method for improving the performance of ambulatory fall-detection systems

BACKGROUND

Falls can cause trauma, disability and death among older people. Ambulatory accelerometer devices are currently capable of detecting falls in a controlled environment. However, research suggests that most current approaches can tend to have insufficient sensitivity and specificity in non-laboratory environments, in part because impacts can be experienced as part of ordinary daily living activities.

METHOD

We used a waist-worn wireless tri-axial accelerometer combined with digital signal processing, clustering and neural network classifiers. The method includes the application of Discrete Wavelet Transform, Regrouping Particle Swarm Optimization, Gaussian Distribution of Clustered Knowledge and an ensemble of classifiers including a multilayer perceptron and Augmented Radial Basis Function (ARBF) neural networks.

RESULTS

Preliminary testing with 8 healthy individuals in a home environment yields 98.6% sensitivity to falls and 99.6% specificity for routine Activities of Daily Living (ADL) data. Single ARB and MLP classifiers were compared with a combined classifier. The combined classifier offers the greatest sensitivity, with a slight reduction in specificity for routine ADL and an increased specificity for exercise activities. In preliminary tests, the approach achieves 100% sensitivity on in-group falls, 97.65% on out-group falls, 99.33% specificity on routine ADL, and 96.59% specificity on exercise ADL.

CONCLUSION

The pre-processing and feature-extraction steps appear to simplify the signal while successfully extracting the essential features that are required to characterize a fall. The results suggest this combination of classifiers can perform better than MLP alone. Preliminary testing suggests these methods may be useful for researchers who are attempting to improve the performance of ambulatory fall-detection systems.