Wearable Technology to Detect Motor Fluctuations in Parkinson's Disease Patients: Current State and Challenges

Innovative Detection and Segmentation of Mobility Activities in Patients Living with Parkinson's Disease Using a Single Ankle-Positioned Smartwatch

BACKGROUND

The automatic detection of activities of daily living (ADL) is necessary to improve long-term home-based monitoring of Parkinson's disease (PD) symptoms. While most body-worn sensor algorithms for ADL detection were developed using laboratory research systems covering full-body kinematics, it is now crucial to achieve ADL detection using a single body-worn sensor that remains commercially available and affordable for ecological use.

AIM

to detect and segment Walking, Turning, Sitting-down, and Standing-up activities of patients with PD using a Smartwatch positioned at the ankle.

METHOD

Twenty-two patients living with PD performed a Timed Up and Go (TUG) task three times before engaging in cleaning ADL in a simulated free-living environment during a 3 min trial. Accelerations and angular velocities of the right or left ankle were recorded in three dimensions using a Smartwatch. The TUG task was used to develop detection algorithms for Walking, Turning, Sitting-down, and Standing-up, while the 3 min trial in the free-living environment was used to test and validate these algorithms. Sensitivity, specificity, and F-scores were calculated based on a manual segmentation of ADL.

RESULTS

Sensitivity, specificity, and F-scores were 96.5%, 94.7%, and 96.0% for Walking; 90.0%, 93.6%, and 91.7% for Turning; 57.5%, 70.5%, and 52.3% for Sitting-down; and 57.5%, 72.9%, and 54.1% for Standing-up. The median of time difference between the manual and automatic segmentation was 1.31 s for Walking, 0.71 s for Turning, 2.75 s for Sitting-down, and 2.35 s for Standing-up.

CONCLUSION

The results of this study demonstrate that segmenting ADL to characterize the mobility of people with PD based on a single Smartwatch can be comparable to manual segmentation while requiring significantly less time. While Walking and Turning were well detected, Sitting-down and Standing-up will require further investigation to develop better algorithms. Nonetheless, these achievements increase the odds of success in implementing wearable technologies for PD monitoring in ecological environments.

Devices for remote continuous monitoring of people with Parkinson's disease: a systematic review and cost-effectiveness analysis

Background

Parkinson's disease is a brain condition causing a progressive loss of co ordination and movement problems. Around 145,500 people have Parkinson's disease in the United Kingdom. Levodopa is the most prescribed treatment for managing motor symptoms in the early stages. Patients should be monitored by a specialist every 6-12 months for disease progression and treatment of adverse effects. Wearable devices may provide a novel approach to management by directly monitoring patients for bradykinesia, dyskinesia, tremor and other symptoms. They are intended to be used alongside clinical judgement.

Objectives

To determine the clinical and cost-effectiveness of five devices for monitoring Parkinson's disease: Personal KinetiGraph, Kinesia 360, KinesiaU, PDMonitor and STAT-ON.

Methods

We performed systematic reviews of all evidence on the five devices, outcomes included: diagnostic accuracy, impact on decision-making, clinical outcomes, patient and clinician opinions and economic outcomes. We searched MEDLINE and 12 other databases/trial registries to February 2022. Risk of bias was assessed. Narrative synthesis was used to summarise all identified evidence, as the evidence was insufficient for meta-analysis. One included trial provided individual-level data, which was re-analysed. A de novo decision-analytic model was developed to estimate the cost-effectiveness of Personal KinetiGraph and Kinesia 360 compared to standard of care in the UK NHS over a 5-year time horizon. The base-case analysis considered two alternative monitoring strategies: one-time use and routine use of the device.

Results

Fifty-seven studies of Personal KinetiGraph, 15 of STAT-ON, 3 of Kinesia 360, 1 of KinesiaU and 1 of PDMonitor were included. There was some evidence to suggest that Personal KinetiGraph can accurately measure bradykinesia and dyskinesia, leading to treatment modification in some patients, and a possible improvement in clinical outcomes when measured using the Unified Parkinson's Disease Rating Scale. The evidence for STAT-ON suggested it may be of value for diagnosing symptoms, but there is currently no evidence on its clinical impact. The evidence for Kinesia 360, KinesiaU and PDMonitor is insufficient to draw any conclusions on their value in clinical practice. The base-case results for Personal KinetiGraph compared to standard of care for one-time and routine use resulted in incremental cost-effectiveness ratios of £67,856 and £57,877 per quality-adjusted life-year gained, respectively, with a beneficial impact of the Personal KinetiGraph on Unified Parkinson's Disease Rating Scale domains III and IV. The incremental cost-effectiveness ratio results for Kinesia 360 compared to standard of care for one-time and routine use were £38,828 and £67,203 per quality-adjusted life-year gained, respectively.

Limitations

The evidence was limited in extent and often low quality. For all devices, except Personal KinetiGraph, there was little to no evidence on the clinical impact of the technology.

Conclusions

Personal KinetiGraph could reasonably be used in practice to monitor patient symptoms and modify treatment where required. There is too little evidence on STAT-ON, Kinesia 360, KinesiaU or PDMonitor to be confident that they are clinically useful. The cost-effectiveness of remote monitoring appears to be largely unfavourable with incremental cost-effectiveness ratios in excess of £30,000 per quality-adjusted life-year across a range of alternative assumptions. The main driver of cost-effectiveness was the durability of improvements in patient symptoms.

Study registration

This study is registered as PROSPERO CRD42022308597.

Funding

This award was funded by the National Institute for Health and Care Research (NIHR) Evidence Synthesis programme (NIHR award ref: NIHR135437) and is published in full in Health Technology Assessment; Vol. 28, No. 30. See the NIHR Funding and Awards website for further award information.

The use of commercial wrist-worn technology to track physiological outcomes in behavioral interventions

PURPOSE OF REVIEW

The aim of this review is to provide an overview of the use of commercial wrist-worn mobile health devices to track and monitor physiological outcomes in behavioral interventions as well as discuss considerations for selecting the optimal device.

RECENT FINDINGS

Wearable technology can enhance intervention design and implementation. The use of wrist-worn wearables provides the opportunity for tracking physiological outcomes, thus providing a unique approach for assessment and delivery of remote interventions. Recent findings support the utility, acceptability, and benefits of commercial wrist-worn wearables in interventions, and they can be used to continuously monitor outcomes, remotely administer assessments, track adherence, and personalize interventions. Wrist-worn devices show acceptable accuracy when measuring heart rate, blood pressure, step counts, and physical activity; however, accuracy is dependent on activity type, intensity, and device brand. These factors should be considered when designing behavioral interventions that utilize wearable technology.

SUMMARY

With the continuous advancement in technology and frequent product upgrades, the capabilities of commercial wrist-worn devices will continue to expand, thus increasing their potential use in intervention research. Continued research is needed to examine and validate the most recent devices on the market to better inform intervention design and implementation.

Objective measurement versus clinician-based assessment for Parkinson's disease

INTRODUCTION

Although clinician-based assessment through standardized clinical rating scales is currently the gold standard for quantifying motor impairment in Parkinson's disease (PD), it is not without limitations, including intra- and inter-rater variability and a degree of approximation. There is increasing evidence supporting the use of objective motion analyses to complement clinician-based assessment. Objective measurement tools hold significant potential for improving the accuracy of clinical and research-based evaluations of patients.

AREAS COVERED

The authors provide several examples from the literature demonstrating how different motion measurement tools, including optoelectronics, contactless and wearable systems allow for both the objective quantification and monitoring of key motor symptoms (such as bradykinesia, rigidity, tremor, and gait disturbances), and the identification of motor fluctuations in PD patients. Furthermore, they discuss how, from a clinician's perspective, objective measurements can help in various stages of PD management.

EXPERT OPINION

In our opinion, sufficient evidence supports the assertion that objective monitoring systems enable accurate evaluation of motor symptoms and complications in PD. A range of devices can be utilized not only to support diagnosis but also to monitor motor symptom during the disease progression and can become relevant in the therapeutic decision-making process.

A smartphone-based tapping task as a marker of medication response in Parkinson's disease: a proof of concept study

Tapping tasks have the potential to distinguish between ON-OFF fluctuations in Parkinson's disease (PD) possibly aiding assessment of medication status in e-diaries and research. This proof of concept study aims to assess the feasibility and accuracy of a smartphone-based tapping task (developed as part of the cloudUPDRS-project) to discriminate between ON-OFF used in the home setting without supervision. 32 PD patients performed the task before their first medication intake, followed by two test sessions after 1 and 3 h. Testing was repeated for 7 days. Index finger tapping between two targets was performed as fast as possible with each hand. Self-reported ON-OFF status was also indicated. Reminders were sent for testing and medication intake. We studied task compliance, objective performance (frequency and inter-tap distance), classification accuracy and repeatability of tapping. Average compliance was 97.0% (± 3.3%), but 16 patients (50%) needed remote assistance. Self-reported ON-OFF scores and objective tapping were worse pre versus post medication intake (p < 0.0005). Repeated tests showed good to excellent test-retest reliability in ON (0.707 ≤ ICC ≤ 0.975). Although 7 days learning effects were apparent, ON-OFF differences remained. Discriminative accuracy for ON-OFF was particularly good for right-hand tapping (0.72 ≤ AUC ≤ 0.80). Medication dose was associated with ON-OFF tapping changes. Unsupervised tapping tests performed on a smartphone have the potential to classify ON-OFF fluctuations in the home setting, despite some learning and time effects. Replication of these results are needed in a wider sample of patients.

Chaudhuri's Dashboard of Vitals in Parkinson's syndrome: an unmet need underpinned by real life clinical tests

We have recently published the notion of the "vitals" of Parkinson's, a conglomeration of signs and symptoms, largely nonmotor, that must not be missed and yet often not considered in neurological consultations, with considerable societal and personal detrimental consequences. This "dashboard," termed the Chaudhuri's vitals of Parkinson's, are summarized as 5 key vital symptoms or signs and comprise of (a) motor, (b) nonmotor, (c) visual, gut, and oral health, (d) bone health and falls, and finally (e) comorbidities, comedication, and dopamine agonist side effects, such as impulse control disorders. Additionally, not addressing the vitals also may reflect inadequate management strategies, leading to worsening quality of life and diminished wellness, a new concept for people with Parkinson's. In this paper, we discuss possible, simple to use, and clinically relevant tests that can be used to monitor the status of these vitals, so that these can be incorporated into clinical practice. We also use the term Parkinson's syndrome to describe Parkinson's disease, as the term "disease" is now abandoned in many countries, such as the U.K., reflecting the heterogeneity of Parkinson's, which is now considered by many as a syndrome.

Deep Clinical Phenotyping of Parkinson's Disease: Towards a New Era of Research and Clinical Care

Despite recent advances in technology, clinical phenotyping of Parkinson's disease (PD) has remained relatively limited as current assessments are mainly based on empirical observation and subjective categorical judgment at the clinic. A lack of comprehensive, objective, and quantifiable clinical phenotyping data has hindered our capacity to diagnose, assess patients' conditions, discover pathogenesis, identify preclinical stages and clinical subtypes, and evaluate new therapies. Therefore, deep clinical phenotyping of PD patients is a necessary step towards understanding PD pathology and improving clinical care. In this review, we present a growing community consensus and perspective on how to clinically phenotype this disease, that is, to phenotype the entire course of disease progression by integrating capacity, performance, and perception approaches with state-of-the-art technology. We also explore the most studied aspects of PD deep clinical phenotypes, namely, bradykinesia, tremor, dyskinesia and motor fluctuation, gait impairment, speech impairment, and non-motor phenotypes.

Sensor Verification and Analytical Validation of Algorithms to Measure Gait and Balance and Pronation/Supination in Healthy Volunteers

Numerous studies have sought to demonstrate the utility of digital measures of motor function in Parkinson’s disease. Frameworks, such as V3, document digital measure development: technical verification, analytical and clinical validation. We present the results of a study to (1) technically verify accelerometers in an Apple iPhone 8 Plus and ActiGraph GT9X versus an oscillating table and (2) analytically validate software tasks for walking and pronation/supination on the iPhone plus passively detect walking measures with the ActiGraph in healthy volunteers versus human raters. In technical verification, 99.4% of iPhone and 91% of ActiGraph tests show good or excellent agreement versus the oscillating table as the gold standard. For the iPhone software task and algorithms, intraclass correlation coefficients (ICCs) > 0.75 are achieved versus the human raters for measures when walking distance is >10 s and pronation/supination when the arm is rotated more than two times. Passively detected walking start and end time was accurate to approx. 1 s and walking measures were accurate to one unit, e.g., one step. The results suggest that the Apple iPhone and ActiGraph GT9X accelerometers are fit for purpose and that task and passively collected measures are sufficiently analytically valid to assess usability and clinical validity in Parkinson’s patients.

Survey of Machine Learning Techniques in the Analysis of EEG Signals for Parkinson’s Disease: A Systematic Review

Background: Parkinson’s disease (PD) affects 7–10 million people worldwide. Its diagnosis is clinical and can be supported by image-based tests, which are expensive and not always accessible. Electroencephalograms (EEG) are non-invasive, widely accessible, low-cost tests. However, the signals obtained are difficult to analyze visually, so advanced techniques, such as Machine Learning (ML), need to be used. In this article, we review those studies that consider ML techniques to study the EEG of patients with PD. Methods: The review process was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, which are used to provide quality standards for the objective evaluation of various studies. All publications before February 2022 were included, and their main characteristics and results were evaluated and documented through three key points associated with the development of ML techniques: dataset quality, data preprocessing, and model evaluation. Results: 59 studies were included. The predominating models were Support Vector Machine (SVM) and Artificial Neural Networks (ANNs). In total, 31 articles diagnosed PD with a mean accuracy of 97.35 ± 3.46%. There was no standard cleaning protocol for EEG and a great heterogeneity in EEG characteristics was shown, although spectral features predominated by 88.37%. Conclusions: Neither the cleaning protocol nor the number of EEG channels influenced the classification results. A baseline value was provided for the PD diagnostic problem, although recent studies focus on the identification of cognitive impairment.

A New Paradigm in Parkinson's Disease Evaluation With Wearable Medical Devices: A Review of STAT-ONTM

In the past decade, the use of wearable medical devices has been a great breakthrough in clinical practice, trials, and research. In the Parkinson's disease field, clinical evaluation is time limited, and healthcare professionals need to rely on retrospective data collected through patients' self-filled diaries and administered questionnaires. As this often leads to inaccurate evaluations, a more objective system for symptom monitoring in a patient's daily life is claimed. In this regard, the use of wearable medical devices is crucial. This study aims at presenting a review on STAT-ONTM, a wearable medical device Class IIa, which provides objective information on the distribution and severity of PD motor symptoms in home environments. The sensor analyzes inertial signals, with a set of validated machine learning algorithms running in real time. The device was developed for 12 years, and this review aims at gathering all the results achieved within this time frame. First, a compendium of the complete journey of STAT-ONTM since 2009 is presented, encompassing different studies and developments in funded European and Spanish national projects. Subsequently, the methodology of database construction and machine learning algorithms design and development is described. Finally, clinical validation and external studies of STAT-ONTM are presented.

Machine Learning Approach to Support the Detection of Parkinson's Disease in IMU-Based Gait Analysis

The aim of this study was to determine which supervised machine learning (ML) algorithm can most accurately classify people with Parkinson’s disease (pwPD) from speed-matched healthy subjects (HS) based on a selected minimum set of IMU-derived gait features. Twenty-two gait features were extrapolated from the trunk acceleration patterns of 81 pwPD and 80 HS, including spatiotemporal, pelvic kinematics, and acceleration-derived gait stability indexes. After a three-level feature selection procedure, seven gait features were considered for implementing five ML algorithms: support vector machine (SVM), artificial neural network, decision trees (DT), random forest (RF), and K-nearest neighbors. Accuracy, precision, recall, and F1 score were calculated. SVM, DT, and RF showed the best classification performances, with prediction accuracy higher than 80% on the test set. The conceptual model of approaching ML that we proposed could reduce the risk of overrepresenting multicollinear gait features in the model, reducing the risk of overfitting in the test performances while fostering the explainability of the results.

Motor Fluctuations Development Is Associated with Non-Mostor Symptoms Burden Progression in Parkinson’s Disease Patients: A 2-Year Follow-Up Study

Objective: The aim of the present study was to analyze the progression of non-motor symptoms (NMS) burden in Parkinson’s disease (PD) patients regarding the development of motor fluctuations (MF). Methods: PD patients without MF at baseline, who were recruited from January 2016 to November 2017 (V0) and evaluated again at a 2-year follow-up (V2) from 35 centers of Spain from the COPPADIS cohort, were included in this analysis. MF development at V2 was defined as a score ≥ 1 in the item-39 of the UPDRS-Part IV, whereas NMS burden was defined according to the Non-motor Symptoms Scale (NMSS) total score. Results: Three hundred and thirty PD patients (62.67 ± 8.7 years old; 58.8% males) were included. From V0 to V2, 27.6% of the patients developed MF. The mean NMSS total score at baseline was higher in those patients who developed MF after the 2-year follow-up (46.34 ± 36.48 vs. 34.3 ± 29.07; p = 0.001). A greater increase in the NMSS total score from V0 to V2 was observed in patients who developed MF (+16.07 ± 37.37) compared to those who did not develop MF (+6.2 ± 25.8) (p = 0.021). Development of MF after a 2-year follow-up was associated with an increase in the NMSS total score (β = 0.128; p = 0.046) after adjustment to age, gender, years from symptoms onset, levodopa equivalent daily dose (LEDD) and the NMSS total score at baseline, and the change in LEDD from V0 to V2. Conclusions: In PD patients, the development of MF is associated with a greater increase in the NMS burden after a 2-year follow-up.

Internet of Things Technologies and Machine Learning Methods for Parkinson's Disease Diagnosis, Monitoring and Management: A Systematic Review

Parkinson's disease is a chronic neurodegenerative disease that affects a large portion of the population, especially the elderly. It manifests with motor, cognitive and other types of symptoms, decreasing significantly the patients' quality of life. The recent advances in the Internet of Things and Artificial Intelligence fields, including the subdomains of machine learning and deep learning, can support Parkinson's disease patients, their caregivers and clinicians at every stage of the disease, maximizing the treatment effectiveness and minimizing the respective healthcare costs at the same time. In this review, the considered studies propose machine learning models, trained on data acquired via smart devices, wearable or non-wearable sensors and other Internet of Things technologies, to provide predictions or estimations regarding Parkinson's disease aspects. Seven hundred and seventy studies have been retrieved from three dominant academic literature databases. Finally, one hundred and twelve of them have been selected in a systematic way and have been considered in the state-of-the-art systematic review presented in this paper. These studies propose various methods, applied on various sensory data to address different Parkinson's disease-related problems. The most widely deployed sensors, the most commonly addressed problems and the best performing algorithms are highlighted. Finally, some challenges are summarized along with some future considerations and opportunities that arise.

Validation and clinical value of the MANAGE-PD tool: A clinician-reported tool to identify Parkinson's disease patients inadequately controlled on oral medications

INTRODUCTION

Making Informed Decisions to Aid Timely Management of Parkinson's Disease (MANAGE-PD) is a clinician-reported tool designed to facilitate timely identification and management of patients with advancing Parkinson's disease (PD) with suboptimal symptom control while on standard therapy. The objective of this study was to evaluate the validity and clinical value of the tool.

METHODS

Driven by structured inputs from a steering committee and panel of PD experts, the tool was developed to classify patients into 3 categories. Validity and clinical value were elucidated using a two-pronged approach: (i) hypothetical patient vignettes (n = 10) developed based on the MANAGE-PD tool and rated by 17 PD specialists and 400 general neurologists (GN) and (ii) patients with PD (n = 2546) managed in real-world clinical settings. Vignette validity was based on concordance between PD experts' clinical judgement and MANAGE-PD vignette categorization. Patient-level data was used for known-group comparisons (validity) and discordant pair analysis (clinical value).

RESULTS

The tool demonstrated strong validity and clinical value among PD specialists (intraclass coefficient [ICC] 0.843; Fleiss weighted kappa [ƙ_weighted] 0.79) and GN (ICC 0.690; ƙ_weighted 0.65) using patient vignettes. MANAGE-PD also demonstrated real-world validity and clinical value based on ability to identify patients with incrementally higher clinical, economic, and humanistic PD burden across categories of the tool (p < 0.01).

CONCLUSIONS

MANAGE-PD demonstrated robust validity and clinical value in identifying patients with suboptimal PD symptom control. Clinical use of MANAGE-PD may complement treatment decision-making and facilitate timely and comprehensive management of patients with advancing PD.